From 0a612556b124d71e81831fbda31c39a7a60b9ce4 Mon Sep 17 00:00:00 2001 From: "Alan, CHUNG" <49159105+dizzyi@users.noreply.github.com> Date: Mon, 19 Apr 2021 14:42:16 +0800 Subject: [PATCH 1/8] update README.md test --- README.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/README.md b/README.md index 87a5b1d..ac00012 100644 --- a/README.md +++ b/README.md @@ -1,2 +1,6 @@ # binPicking Central repository for our large-scale and efficient bin-picking project. wiki will be a central repository for all aspects of this work. + +## Visual Feedback +For visual feedback, this project will utilize a Graph-based neural network deep reinforcement learning for object detection and tracking. + From 3b49061b13e83d5ce2fbe1184f5ee162765bc1b3 Mon Sep 17 00:00:00 2001 From: dizzyi Date: Thu, 22 Jul 2021 10:42:17 +0800 Subject: [PATCH 2/8] added notebook and readme for downloading data, preprocessing data and training the model --- .../script/detectron/Preprocess_Data.ipynb | 148 +++++++++ binPicking/script/detectron/README.md | 0 .../detectron/Train_on_modify_COCO.ipynb | 281 ++++++++++++++++++ 3 files changed, 429 insertions(+) create mode 100644 binPicking/script/detectron/Preprocess_Data.ipynb create mode 100644 binPicking/script/detectron/README.md create mode 100644 binPicking/script/detectron/Train_on_modify_COCO.ipynb diff --git a/binPicking/script/detectron/Preprocess_Data.ipynb b/binPicking/script/detectron/Preprocess_Data.ipynb new file mode 100644 index 0000000..8789dbc --- /dev/null +++ b/binPicking/script/detectron/Preprocess_Data.ipynb @@ -0,0 +1,148 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "id": "eb245494", + "metadata": {}, + "outputs": [], + "source": [ + "import json" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "2f94b5da", + "metadata": {}, + "outputs": [], + "source": [ + "# The modified category, stripe all useless classes\n", + "# TODO: change it to real modified category\n", + "modified_category = [\n", + " {'supercategory': 'indoor', 'id': 84, 'name': 'book'},\n", + " {'supercategory': 'indoor', 'id': 85, 'name': 'clock'},\n", + " {'supercategory': 'indoor', 'id': 86, 'name': 'vase'},\n", + " {'supercategory': 'indoor', 'id': 87, 'name': 'scissors'},\n", + " {'supercategory': 'indoor', 'id': 88, 'name': 'teddy bear'},\n", + " {'supercategory': 'indoor', 'id': 89, 'name': 'hair drier'},\n", + " {'supercategory': 'indoor', 'id': 90, 'name': 'toothbrush'}\n", + "]\n", + "with open(f\"./modified_category.json\", 'w') as f:\n", + " json.dump(modified_category,f)" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "bab1716b", + "metadata": {}, + "outputs": [], + "source": [ + "## data's directory\n", + "DATA_ROOT = './coco'\n", + "PREPROCESS_DATA_ROOT = './coco'" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "2ee597db", + "metadata": {}, + "outputs": [], + "source": [ + "with open(f\"{DATA_ROOT}/instances_val2017.json\") as f:\n", + " data = json.load(f)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "91abb0ae", + "metadata": {}, + "outputs": [], + "source": [ + "#data['annotations']" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "7ada02ba", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "#data['categories']" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "a43e5d5b", + "metadata": {}, + "outputs": [], + "source": [ + "data['categories'] = modified_category\n", + "#data['categories'] " + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "1ea86c39", + "metadata": {}, + "outputs": [], + "source": [ + "S = set()\n", + "for cat in modified_category:\n", + " S.add(cat['id'])\n", + " #print(f'{cat[\"id\"]} is added to the set')" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "f3f16c20", + "metadata": {}, + "outputs": [], + "source": [ + "data['annotations'] = list(filter( lambda anno: S.__contains__(anno['category_id']) ,data['annotations']))\n", + "#data['annotations']" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "4e4c00f1", + "metadata": {}, + "outputs": [], + "source": [ + "with open(f\"{PREPROCESS_DATA_ROOT}/modified_train2017.json\", 'w') as f:\n", + " json.dump(data,f)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.10" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/binPicking/script/detectron/README.md b/binPicking/script/detectron/README.md new file mode 100644 index 0000000..e69de29 diff --git a/binPicking/script/detectron/Train_on_modify_COCO.ipynb b/binPicking/script/detectron/Train_on_modify_COCO.ipynb new file mode 100644 index 0000000..12150e7 --- /dev/null +++ b/binPicking/script/detectron/Train_on_modify_COCO.ipynb @@ -0,0 +1,281 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 4, + "id": "e1c917cf", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1.8.1+cpu False\n" + ] + } + ], + "source": [ + "import torch, torchvision\n", + "print(torch.__version__, torch.cuda.is_available())" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "24a28889", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "finish importing\n" + ] + } + ], + "source": [ + "from detectron2.utils.logger import setup_logger\n", + "setup_logger()\n", + "\n", + "# import some common libraries\n", + "import numpy as np\n", + "import os, json, cv2, random\n", + "from PIL import Image\n", + "from matplotlib.pyplot import imshow\n", + "import json\n", + "\n", + "#import some common detectron2 utilities\n", + "from detectron2 import model_zoo\n", + "from detectron2.engine import DefaultPredictor\n", + "from detectron2.config import get_cfg\n", + "from detectron2.utils.visualizer import Visualizer\n", + "from detectron2.data import MetadataCatalog, DatasetCatalog\n", + "from detectron2.data.datasets import register_coco_instances\n", + "\n", + "print(\"finish importing\")" + ] + }, + { + "cell_type": "markdown", + "id": "145a461f", + "metadata": {}, + "source": [ + "# Prepare Data" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "3f4c5afe", + "metadata": {}, + "outputs": [], + "source": [ + "'''\n", + "The file structure of the dataset\n", + "coco (DATA_ROOT)\n", + " L modified_train2017.json\n", + " L modified_val2017.json\n", + " L image/\n", + "'''\n", + "DATA_ROOT = './coco'" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "df4e5bc1", + "metadata": {}, + "outputs": [], + "source": [ + "register_coco_instances(\"modify_coco_train\", {}, f\"{DATA_ROOT}/modified_train2017.json\", f\"{DATA_ROOT}/image\")\n", + "#register_coco_instances(\"modify_coco_val\" , {}, f\"{DATA_ROOT}/jmodified_val2017.json\" , f\"{DATA_ROOT}/image\")" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "b02bdaa1", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "NUM_CLASSES = 7\n" + ] + } + ], + "source": [ + "with open('./modified_category.json', 'r') as f:\n", + " NUM_CLASSES = len(json.load(f))\n", + "print(f\"NUM_CLASSES = {NUM_CLASSES}\")" + ] + }, + { + "cell_type": "markdown", + "id": "9bcf4e37", + "metadata": {}, + "source": [ + "# Prepare for Training" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "6a6d43e2", + "metadata": {}, + "outputs": [], + "source": [ + "from detectron2.engine import DefaultTrainer\n", + "\n", + "cfg = get_cfg()\n", + "cfg.merge_from_file(model_zoo.get_config_file(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\"))\n", + "cfg.DATASETS.TRAIN = (\"modify_coco_train\",)\n", + "cfg.DATASETS.TEST = ()\n", + "cfg.DATALOADER.NUM_WORKERS = 2\n", + "cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\") # Let training initialize from model zoo\n", + "cfg.SOLVER.IMS_PER_BATCH = 2\n", + "cfg.SOLVER.BASE_LR = 0.00025 # pick a good LR\n", + "cfg.SOLVER.MAX_ITER = 300 # 300 iterations seems good enough for this toy dataset; you will need to train longer for a practical dataset\n", + "cfg.SOLVER.STEPS = [] # do not decay learning rate\n", + "cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 128 # faster, and good enough for this toy dataset (default: 512)\n", + "cfg.MODEL.ROI_HEADS.NUM_CLASSES = NUM_CLASSES \n", + "# NOTE: this config means the number of classes, but a few popular unofficial tutorials incorrect uses num_classes+1 here.\n", + "cfg.OUTPUT_DIR = './model'\n", + "\n", + "os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)\n", + "trainer = DefaultTrainer(cfg) \n", + "trainer.resume_or_load(resume=False)\n", + "trainer.train()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "05786fa4", + "metadata": {}, + "outputs": [], + "source": [ + "# Look at training curves in tensorboard:\n", + "%load_ext tensorboard\n", + "%tensorboard --logdir output" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "86aa2505", + "metadata": {}, + "outputs": [], + "source": [ + "import datetime\n", + "# Inference should use the config with parameters that are used in training\n", + "# cfg now already contains everything we've set previously. We changed it a little bit for inference:\n", + "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, f\"model_final_{datetime.datetime.now()}.pth\") # path to the model we just trained\n", + "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", + "predictor = DefaultPredictor(cfg)" + ] + }, + { + "cell_type": "markdown", + "id": "799d0b5b", + "metadata": {}, + "source": [ + "# Check the Model on Robosuite Example " + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "12c8e40a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(256, 256, 3)\n" + ] + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "pic = np.asarray(Image.open('name_0.png'))\n", + "imshow(pic)\n", + "#pic = pic.transpose((2,0,1))\n", + "print(pic.shape)" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "9df99e5e", + "metadata": {}, + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'predictor' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpredictor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpic\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;31m# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred_classes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred_boxes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'instances'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mNameError\u001b[0m: name 'predictor' is not defined" + ] + } + ], + "source": [ + "outputs = predictor(pic)\n", + "# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\n", + "print(outputs[\"instances\"].pred_classes)\n", + "print(outputs[\"instances\"].pred_boxes)\n", + "print(outputs['instances'])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f81b47bd", + "metadata": {}, + "outputs": [], + "source": [ + "# We can use `Visualizer` to draw the predictions on the image.\n", + "v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1)\n", + "out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n", + "imshow(out.get_image()[:, :, ::-1])" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.10" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From 46c47a15736734c4cd2cdac0cb033cce52d12ff3 Mon Sep 17 00:00:00 2001 From: dizzyi Date: Thu, 22 Jul 2021 10:43:44 +0800 Subject: [PATCH 3/8] added notebook and readme for downloading data, preprocessing data and training the model --- .../script/detectron/Preprocess_Data.ipynb | 76 ++++----- binPicking/script/detectron/README.md | 31 ++++ .../detectron/Train_on_modify_COCO.ipynb | 154 ++++++++---------- 3 files changed, 134 insertions(+), 127 deletions(-) diff --git a/binPicking/script/detectron/Preprocess_Data.ipynb b/binPicking/script/detectron/Preprocess_Data.ipynb index 8789dbc..5e14ca7 100644 --- a/binPicking/script/detectron/Preprocess_Data.ipynb +++ b/binPicking/script/detectron/Preprocess_Data.ipynb @@ -3,19 +3,15 @@ { "cell_type": "code", "execution_count": 1, - "id": "eb245494", - "metadata": {}, - "outputs": [], "source": [ "import json" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 2, - "id": "2f94b5da", - "metadata": {}, - "outputs": [], "source": [ "# The modified category, stripe all useless classes\n", "# TODO: change it to real modified category\n", @@ -30,98 +26,92 @@ "]\n", "with open(f\"./modified_category.json\", 'w') as f:\n", " json.dump(modified_category,f)" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 3, - "id": "bab1716b", - "metadata": {}, - "outputs": [], "source": [ "## data's directory\n", "DATA_ROOT = './coco'\n", "PREPROCESS_DATA_ROOT = './coco'" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 4, - "id": "2ee597db", - "metadata": {}, - "outputs": [], "source": [ "with open(f\"{DATA_ROOT}/instances_val2017.json\") as f:\n", " data = json.load(f)" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 5, - "id": "91abb0ae", - "metadata": {}, - "outputs": [], "source": [ "#data['annotations']" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 6, - "id": "7ada02ba", - "metadata": { - "scrolled": true - }, - "outputs": [], "source": [ "#data['categories']" - ] + ], + "outputs": [], + "metadata": { + "scrolled": true + } }, { "cell_type": "code", "execution_count": 7, - "id": "a43e5d5b", - "metadata": {}, - "outputs": [], "source": [ "data['categories'] = modified_category\n", "#data['categories'] " - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 8, - "id": "1ea86c39", - "metadata": {}, - "outputs": [], "source": [ "S = set()\n", "for cat in modified_category:\n", " S.add(cat['id'])\n", " #print(f'{cat[\"id\"]} is added to the set')" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 9, - "id": "f3f16c20", - "metadata": {}, - "outputs": [], "source": [ "data['annotations'] = list(filter( lambda anno: S.__contains__(anno['category_id']) ,data['annotations']))\n", "#data['annotations']" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 10, - "id": "4e4c00f1", - "metadata": {}, - "outputs": [], "source": [ "with open(f\"{PREPROCESS_DATA_ROOT}/modified_train2017.json\", 'w') as f:\n", " json.dump(data,f)" - ] + ], + "outputs": [], + "metadata": {} } ], "metadata": { @@ -145,4 +135,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file diff --git a/binPicking/script/detectron/README.md b/binPicking/script/detectron/README.md index e69de29..9fa7d06 100644 --- a/binPicking/script/detectron/README.md +++ b/binPicking/script/detectron/README.md @@ -0,0 +1,31 @@ +# Train Model for visual +## 1. Download the COCO dataset +Download dataset: https://cocodataset.org/#download + +download + +- 2017 Train images[118K/18GB] +- 2017 Train/Val annotations [241MB] + +save the file in structure + + coco (DATA_ROOT) + L instances_train2017.json + L instance_val2017.json + L image/ + +## 2. Preprocess the Data +open the 'Preprocess_Data.ipynb' notebook + +update the ```DATA_ROOT``` and ```PROPRECRESS_DATA_ROOT``` + +run the notebook and preprocess the COCO data to strip all useless classes' annotation. + +## 3. Train the pretrain Model on the modified COCO data +install ```cuda, torch, detectron2``` + +Detectron2: https://detectron2.readthedocs.io/en/latest/tutorials/install.html#install-pre-built-detectron2-linux-only + +update the ```DATA_ROOT``` + +run the notebook to train the model. \ No newline at end of file diff --git a/binPicking/script/detectron/Train_on_modify_COCO.ipynb b/binPicking/script/detectron/Train_on_modify_COCO.ipynb index 12150e7..e28465c 100644 --- a/binPicking/script/detectron/Train_on_modify_COCO.ipynb +++ b/binPicking/script/detectron/Train_on_modify_COCO.ipynb @@ -3,36 +3,24 @@ { "cell_type": "code", "execution_count": 4, - "id": "e1c917cf", - "metadata": {}, + "source": [ + "import torch, torchvision\n", + "print(torch.__version__, torch.cuda.is_available())" + ], "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "1.8.1+cpu False\n" ] } ], - "source": [ - "import torch, torchvision\n", - "print(torch.__version__, torch.cuda.is_available())" - ] + "metadata": {} }, { "cell_type": "code", "execution_count": 3, - "id": "24a28889", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "finish importing\n" - ] - } - ], "source": [ "from detectron2.utils.logger import setup_logger\n", "setup_logger()\n", @@ -53,22 +41,28 @@ "from detectron2.data.datasets import register_coco_instances\n", "\n", "print(\"finish importing\")" - ] + ], + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "finish importing\n" + ] + } + ], + "metadata": {} }, { "cell_type": "markdown", - "id": "145a461f", - "metadata": {}, "source": [ "# Prepare Data" - ] + ], + "metadata": {} }, { "cell_type": "code", "execution_count": 1, - "id": "3f4c5afe", - "metadata": {}, - "outputs": [], "source": [ "'''\n", "The file structure of the dataset\n", @@ -78,53 +72,49 @@ " L image/\n", "'''\n", "DATA_ROOT = './coco'" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": null, - "id": "df4e5bc1", - "metadata": {}, - "outputs": [], "source": [ "register_coco_instances(\"modify_coco_train\", {}, f\"{DATA_ROOT}/modified_train2017.json\", f\"{DATA_ROOT}/image\")\n", "#register_coco_instances(\"modify_coco_val\" , {}, f\"{DATA_ROOT}/jmodified_val2017.json\" , f\"{DATA_ROOT}/image\")" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": 8, - "id": "b02bdaa1", - "metadata": {}, + "source": [ + "with open('./modified_category.json', 'r') as f:\n", + " NUM_CLASSES = len(json.load(f))\n", + "print(f\"NUM_CLASSES = {NUM_CLASSES}\")" + ], "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "NUM_CLASSES = 7\n" ] } ], - "source": [ - "with open('./modified_category.json', 'r') as f:\n", - " NUM_CLASSES = len(json.load(f))\n", - "print(f\"NUM_CLASSES = {NUM_CLASSES}\")" - ] + "metadata": {} }, { "cell_type": "markdown", - "id": "9bcf4e37", - "metadata": {}, "source": [ "# Prepare for Training" - ] + ], + "metadata": {} }, { "cell_type": "code", "execution_count": null, - "id": "6a6d43e2", - "metadata": {}, - "outputs": [], "source": [ "from detectron2.engine import DefaultTrainer\n", "\n", @@ -147,26 +137,24 @@ "trainer = DefaultTrainer(cfg) \n", "trainer.resume_or_load(resume=False)\n", "trainer.train()" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": null, - "id": "05786fa4", - "metadata": {}, - "outputs": [], "source": [ "# Look at training curves in tensorboard:\n", "%load_ext tensorboard\n", "%tensorboard --logdir output" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "code", "execution_count": null, - "id": "86aa2505", - "metadata": {}, - "outputs": [], "source": [ "import datetime\n", "# Inference should use the config with parameters that are used in training\n", @@ -174,59 +162,64 @@ "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, f\"model_final_{datetime.datetime.now()}.pth\") # path to the model we just trained\n", "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", "predictor = DefaultPredictor(cfg)" - ] + ], + "outputs": [], + "metadata": {} }, { "cell_type": "markdown", - "id": "799d0b5b", - "metadata": {}, "source": [ "# Check the Model on Robosuite Example " - ] + ], + "metadata": {} }, { "cell_type": "code", "execution_count": 3, - "id": "12c8e40a", - "metadata": {}, + "source": [ + "pic = np.asarray(Image.open('name_0.png'))\n", + "imshow(pic)\n", + "#pic = pic.transpose((2,0,1))\n", + "print(pic.shape)" + ], "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "(256, 256, 3)\n" ] }, { + "output_type": "display_data", "data": { - "image/png": "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\n", "text/plain": [ "
" - ] + ], + "image/png": "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" }, "metadata": { "needs_background": "light" - }, - "output_type": "display_data" + } } ], - "source": [ - "pic = np.asarray(Image.open('name_0.png'))\n", - "imshow(pic)\n", - "#pic = pic.transpose((2,0,1))\n", - "print(pic.shape)" - ] + "metadata": {} }, { "cell_type": "code", "execution_count": 4, - "id": "9df99e5e", - "metadata": {}, + "source": [ + "outputs = predictor(pic)\n", + "# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\n", + "print(outputs[\"instances\"].pred_classes)\n", + "print(outputs[\"instances\"].pred_boxes)\n", + "print(outputs['instances'])" + ], "outputs": [ { + "output_type": "error", "ename": "NameError", "evalue": "name 'predictor' is not defined", - "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", @@ -235,26 +228,19 @@ ] } ], - "source": [ - "outputs = predictor(pic)\n", - "# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\n", - "print(outputs[\"instances\"].pred_classes)\n", - "print(outputs[\"instances\"].pred_boxes)\n", - "print(outputs['instances'])" - ] + "metadata": {} }, { "cell_type": "code", "execution_count": null, - "id": "f81b47bd", - "metadata": {}, - "outputs": [], "source": [ "# We can use `Visualizer` to draw the predictions on the image.\n", "v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1)\n", "out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n", "imshow(out.get_image()[:, :, ::-1])" - ] + ], + "outputs": [], + "metadata": {} } ], "metadata": { @@ -278,4 +264,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file From 0ab3d66aee5cf50e3c5ee3c08909fed37574ca08 Mon Sep 17 00:00:00 2001 From: dizzyi Date: Thu, 22 Jul 2021 12:22:18 +0800 Subject: [PATCH 4/8] added notebook and readme for downloading data, preprocessing data and training the model --- .../script/detectron/Preprocess_Data.ipynb | 139 +++++++++++------- 1 file changed, 87 insertions(+), 52 deletions(-) diff --git a/binPicking/script/detectron/Preprocess_Data.ipynb b/binPicking/script/detectron/Preprocess_Data.ipynb index 5e14ca7..5297564 100644 --- a/binPicking/script/detectron/Preprocess_Data.ipynb +++ b/binPicking/script/detectron/Preprocess_Data.ipynb @@ -3,115 +3,150 @@ { "cell_type": "code", "execution_count": 1, + "id": "eb245494", + "metadata": {}, + "outputs": [], "source": [ "import json" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", "execution_count": 2, + "id": "2f94b5da", + "metadata": {}, + "outputs": [], "source": [ "# The modified category, stripe all useless classes\n", "# TODO: change it to real modified category\n", - "modified_category = [\n", - " {'supercategory': 'indoor', 'id': 84, 'name': 'book'},\n", - " {'supercategory': 'indoor', 'id': 85, 'name': 'clock'},\n", - " {'supercategory': 'indoor', 'id': 86, 'name': 'vase'},\n", - " {'supercategory': 'indoor', 'id': 87, 'name': 'scissors'},\n", - " {'supercategory': 'indoor', 'id': 88, 'name': 'teddy bear'},\n", - " {'supercategory': 'indoor', 'id': 89, 'name': 'hair drier'},\n", - " {'supercategory': 'indoor', 'id': 90, 'name': 'toothbrush'}\n", - "]\n", - "with open(f\"./modified_category.json\", 'w') as f:\n", - " json.dump(modified_category,f)" - ], - "outputs": [], - "metadata": {} + "modified_categories = [\n", + " {\"supercategory\": \"vehicle\" ,\"id\": 5,\"name\": \"airplane\"},\n", + " {\"supercategory\": \"accessory\" ,\"id\": 32,\"name\": \"tie\"},\n", + " {\"supercategory\": \"sports\" ,\"id\": 37,\"name\": \"sports ball\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 44,\"name\": \"bottle\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 46,\"name\": \"wine glass\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 47,\"name\": \"cup\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 48,\"name\": \"fork\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 49,\"name\": \"knife\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 50,\"name\": \"spoon\"},\n", + " {\"supercategory\": \"kitchen\" ,\"id\": 51,\"name\": \"bowl\"},\n", + " {\"supercategory\": \"food\" ,\"id\": 52,\"name\": \"banana\"},\n", + " {\"supercategory\": \"food\" ,\"id\": 53,\"name\": \"apple\"},\n", + " {\"supercategory\": \"food\" ,\"id\": 55,\"name\": \"orange\"},\n", + " {\"supercategory\": \"food\" ,\"id\": 56,\"name\": \"broccoli\"},\n", + " {\"supercategory\": \"food\" ,\"id\": 57,\"name\": \"carrot\"},\n", + " {\"supercategory\": \"electronic\",\"id\": 74,\"name\": \"mouse\"},\n", + " {\"supercategory\": \"electronic\",\"id\": 75,\"name\": \"remote\"},\n", + " {\"supercategory\": \"indoor\" ,\"id\": 84,\"name\": \"book\"},\n", + " {\"supercategory\": \"indoor\" ,\"id\": 87,\"name\": \"scissors\"},\n", + " {\"supercategory\": \"indoor\" ,\"id\": 90,\"name\": \"toothbrush\"}\n", + " ]\n", + "with open(f\"./modified_categories.json\", 'w') as f:\n", + " json.dump(modified_categories,f)" + ] }, { "cell_type": "code", "execution_count": 3, + "id": "a779256c", + "metadata": {}, + "outputs": [], "source": [ "## data's directory\n", "DATA_ROOT = './coco'\n", "PREPROCESS_DATA_ROOT = './coco'" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", "execution_count": 4, + "id": "2ee597db", + "metadata": {}, + "outputs": [ + { + "ename": "FileNotFoundError", + "evalue": "[Errno 2] No such file or directory: './coco/instances_val2017.json'", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mFileNotFoundError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"{DATA_ROOT}/instances_val2017.json\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: './coco/instances_val2017.json'" + ] + } + ], "source": [ "with open(f\"{DATA_ROOT}/instances_val2017.json\") as f:\n", " data = json.load(f)" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, + "id": "91abb0ae", + "metadata": {}, + "outputs": [], "source": [ "#data['annotations']" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", - "execution_count": 6, - "source": [ - "#data['categories']" - ], - "outputs": [], + "execution_count": null, + "id": "7ada02ba", "metadata": { "scrolled": true - } + }, + "outputs": [], + "source": [ + "#data['categories']" + ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": null, + "id": "a43e5d5b", + "metadata": {}, + "outputs": [], "source": [ - "data['categories'] = modified_category\n", + "data['categories'] = modified_categories\n", "#data['categories'] " - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, + "id": "1ea86c39", + "metadata": {}, + "outputs": [], "source": [ "S = set()\n", - "for cat in modified_category:\n", + "for cat in modified_categories:\n", " S.add(cat['id'])\n", " #print(f'{cat[\"id\"]} is added to the set')" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, + "id": "f3f16c20", + "metadata": {}, + "outputs": [], "source": [ "data['annotations'] = list(filter( lambda anno: S.__contains__(anno['category_id']) ,data['annotations']))\n", "#data['annotations']" - ], - "outputs": [], - "metadata": {} + ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, + "id": "4e4c00f1", + "metadata": {}, + "outputs": [], "source": [ "with open(f\"{PREPROCESS_DATA_ROOT}/modified_train2017.json\", 'w') as f:\n", " json.dump(data,f)" - ], - "outputs": [], - "metadata": {} + ] } ], "metadata": { @@ -135,4 +170,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From 272bc8fc0635c1b945a07c13b3b9200bfd11bfad Mon Sep 17 00:00:00 2001 From: dizzyi Date: Wed, 28 Jul 2021 17:03:09 +0800 Subject: [PATCH 5/8] added 1 test photo compiled jupyter notebook to py file --- .../script/detectron/Preprocess_Data.ipynb | 91 +++++--- binPicking/script/detectron/Test.py | 66 ++++++ binPicking/script/detectron/Train.py | 163 +++++++++++++ .../detectron/Train_on_modify_COCO.ipynb | 214 +++++++++++------- binPicking/script/detectron/name_0.png | Bin 0 -> 89413 bytes 5 files changed, 417 insertions(+), 117 deletions(-) create mode 100644 binPicking/script/detectron/Test.py create mode 100644 binPicking/script/detectron/Train.py create mode 100644 binPicking/script/detectron/name_0.png diff --git a/binPicking/script/detectron/Preprocess_Data.ipynb b/binPicking/script/detectron/Preprocess_Data.ipynb index 5297564..0754da4 100644 --- a/binPicking/script/detectron/Preprocess_Data.ipynb +++ b/binPicking/script/detectron/Preprocess_Data.ipynb @@ -7,7 +7,7 @@ "metadata": {}, "outputs": [], "source": [ - "import json" + "import json, os" ] }, { @@ -40,7 +40,7 @@ " {\"supercategory\": \"indoor\" ,\"id\": 84,\"name\": \"book\"},\n", " {\"supercategory\": \"indoor\" ,\"id\": 87,\"name\": \"scissors\"},\n", " {\"supercategory\": \"indoor\" ,\"id\": 90,\"name\": \"toothbrush\"}\n", - " ]\n", + "]\n", "with open(f\"./modified_categories.json\", 'w') as f:\n", " json.dump(modified_categories,f)" ] @@ -48,13 +48,13 @@ { "cell_type": "code", "execution_count": 3, - "id": "a779256c", + "id": "9f9eb29c", "metadata": {}, "outputs": [], "source": [ "## data's directory\n", - "DATA_ROOT = './coco'\n", - "PREPROCESS_DATA_ROOT = './coco'" + "DATA_ROOT = './coco'\n", + "PREPROCESSED_DATA_ROOT = './coco'" ] }, { @@ -62,27 +62,15 @@ "execution_count": 4, "id": "2ee597db", "metadata": {}, - "outputs": [ - { - "ename": "FileNotFoundError", - "evalue": "[Errno 2] No such file or directory: './coco/instances_val2017.json'", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mFileNotFoundError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"{DATA_ROOT}/instances_val2017.json\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: './coco/instances_val2017.json'" - ] - } - ], + "outputs": [], "source": [ - "with open(f\"{DATA_ROOT}/instances_val2017.json\") as f:\n", + "with open(os.path.join(DATA_ROOT,\"instances_val2017.json\")) as f:\n", " data = json.load(f)" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "id": "91abb0ae", "metadata": {}, "outputs": [], @@ -92,7 +80,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "id": "7ada02ba", "metadata": { "scrolled": true @@ -104,7 +92,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "id": "a43e5d5b", "metadata": {}, "outputs": [], @@ -115,36 +103,77 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "id": "1ea86c39", "metadata": {}, "outputs": [], "source": [ - "S = set()\n", - "for cat in modified_categories:\n", - " S.add(cat['id'])\n", - " #print(f'{cat[\"id\"]} is added to the set')" + "cats = [cat['id'] for cat in modified_categories]" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "id": "f3f16c20", "metadata": {}, "outputs": [], "source": [ - "data['annotations'] = list(filter( lambda anno: S.__contains__(anno['category_id']) ,data['annotations']))\n", + "data['annotations'] = list(filter( lambda anno: anno['category_id'] in cats ,data['annotations']))\n", "#data['annotations']" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, + "id": "b945cd6a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "5 142 True\n", + "32 253 True\n", + "37 262 True\n", + "44 1024 True\n", + "46 342 True\n", + "47 898 True\n", + "48 214 True\n", + "49 325 True\n", + "50 252 True\n", + "51 625 True\n", + "52 378 True\n", + "53 238 True\n", + "55 286 True\n", + "56 315 True\n", + "57 370 True\n", + "74 105 True\n", + "75 282 True\n", + "84 1160 True\n", + "87 35 True\n", + "90 56 True\n" + ] + } + ], + "source": [ + "## verification and count \n", + "count = {}\n", + "for anno in data['annotations']:\n", + " cat_id = anno['category_id']\n", + " if cat_id in count: count[cat_id] += 1\n", + " else: count[cat_id] = 0\n", + " \n", + "for k, v in sorted(count.items()): print(k, v, cat_id in cats)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, "id": "4e4c00f1", "metadata": {}, "outputs": [], "source": [ - "with open(f\"{PREPROCESS_DATA_ROOT}/modified_train2017.json\", 'w') as f:\n", + "with open(os.path.join(PREPROCESSED_DATA_ROOT,\"modified_train2017.json\"), 'w') as f:\n", " json.dump(data,f)" ] } diff --git a/binPicking/script/detectron/Test.py b/binPicking/script/detectron/Test.py new file mode 100644 index 0000000..4ec85a9 --- /dev/null +++ b/binPicking/script/detectron/Test.py @@ -0,0 +1,66 @@ +# In[] +# import some common libraries +import numpy as np +import os, json, cv2, random, pickle +from PIL import Image +from matplotlib.pyplot import imshow +import json + +#import some common detectron2 utilities +from detectron2 import model_zoo +from detectron2.engine import DefaultPredictor +from detectron2.config import get_cfg +from detectron2.utils.visualizer import Visualizer +from detectron2.data import MetadataCatalog, DatasetCatalog +from detectron2.data.datasets import register_coco_instances + +# In[ ]: + + +## load the model and the weight +""" +MODEL_ROOT + L model_cfg.pickle + L {cfg.OUTPUT_DIR} + L model_final.pth +""" +MODEL_ROOT = './' +cfg = {} +with open('model_cfg.pickle' , 'rb') as f: + cfg = pickle.load(f) + +print(cfg.OUTPUT_DIR) + +cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth") # path to the model we just trained +cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold +predictor = DefaultPredictor(cfg) + + +# In[ ]: + + +pic = np.asarray(Image.open('name_0.png')) +imshow(pic) +#pic = pic.transpose((2,0,1)) +print(pic.shape) + + +# In[ ]: + + +outputs = predictor(pic) +# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification +print(outputs["instances"].pred_classes) +print(outputs["instances"].pred_boxes) +print(outputs['instances']) + + +# In[1]: + + +# We can use `Visualizer` to draw the predictions on the image. +v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1) +out = v.draw_instance_predictions(outputs["instances"].to("cpu")) +imshow(out.get_image()[:, :, ::-1]) +im = Image.fromarray(out) +out.save('output_name_0.jpg') diff --git a/binPicking/script/detectron/Train.py b/binPicking/script/detectron/Train.py new file mode 100644 index 0000000..7def0ed --- /dev/null +++ b/binPicking/script/detectron/Train.py @@ -0,0 +1,163 @@ +#!/usr/bin/env python +# coding: utf-8 + +# In[ ]: + + +import torch, torchvision +print(torch.__version__, torch.cuda.is_available()) + + +# In[ ]: + + +from detectron2.utils.logger import setup_logger +setup_logger() + +# import some common libraries +import numpy as np +import os, json, cv2, random +from PIL import Image +from matplotlib.pyplot import imshow +import json + +#import some common detectron2 utilities +from detectron2 import model_zoo +from detectron2.engine import DefaultPredictor +from detectron2.config import get_cfg +from detectron2.utils.visualizer import Visualizer +from detectron2.data import MetadataCatalog, DatasetCatalog +from detectron2.data.datasets import register_coco_instances + +print("finish importing") + + +# # Prepare Data + +# In[ ]: + + +''' +The file structure of the dataset +coco (DATA_ROOT) + L modified_train2017.json + L modified_val2017.json + L image/ +''' +DATA_ROOT = './coco' + + +# In[ ]: + + +register_coco_instances( + "modify_coco_train", + {}, + os.path.join( DATA_ROOT, "modified_train2017.json"), + os.path.join( DATA_ROOT, "image") +) +#register_coco_instances("modify_coco_val" , {}, f"{DATA_ROOT}/jmodified_val2017.json" , f"{DATA_ROOT}/image") + + +# In[ ]: + + +with open('./modified_category.json', 'r') as f: + NUM_CLASSES = len(json.load(f)) +print(f"NUM_CLASSES = {NUM_CLASSES}") + + +# # Prepare for Training + +# In[ ]: + + +from detectron2.engine import DefaultTrainer + +cfg = get_cfg() +cfg.merge_from_file( model_zoo.get_config_file( "COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")) +cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml") +""" +Model Zoo Link: +https://github.com/facebookresearch/detectron2/blob/master/ +MODEL_ZOO.md#coco-instance-segmentation-baselines-with-mask-r-cnn +""" +cfg.DATASETS.TRAIN = ("modify_coco_train",) +cfg.DATASETS.TEST = () +# Detectron default 4 +cfg.DATALOADER.NUM_WORKERS = 4 +# Detectron default 40000 +cfg.SOLVER.MAX_ITER = 160_000 +''' +Detectron default +Base Learning rate 0.001 +GAMMA 0.1 +STEP (30000,) + GAMMA : Learning rate decay factor + STEPS: num of iter for learning rate decay by gamma + +MASK RCNN PAPER : https://arxiv.org/pdf/1703.06870.pdf + Base LR 0.02 + decay by 10 @ 120k/160k + + Cityscapes finetuning + Base LR 0.001 + decay by 10 @ 18k/24k + + update baseline + Base LR 0.001 + decay by 10 @ 120k,160k/180k + + Benefit form deeper model +''' +cfg.SOLVER.BASE_LR = 0.001 +cfg.SOLVER.GAMMA = 0.1 +cfg.SOLVER.STEPS = (120_000,) +cfg.SOLVER.WEIGHT_DECAY = 0.000_1 + +# Detectron default 16 +cfg.SOLVER.IMS_PER_BATCH = 32 +# Detectron default 512 +cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 2048 + +# Number of classes +cfg.MODEL.ROI_HEADS.NUM_CLASSES = NUM_CLASSES + +# Confident Level +cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold + +cfg.OUTPUT_DIR = './model' +os.makedirs(cfg.OUTPUT_DIR, exist_ok=True) +#cfg.dump() + + +# In[ ]: + + +trainer = DefaultTrainer(cfg) +trainer.resume_or_load(resume=False) +trainer.train() + + +# In[ ]: +from IPython import get_ipython + +# Look at training curves in tensorboard: +get_ipython().run_line_magic('load_ext', 'tensorboard') +get_ipython().run_line_magic('tensorboard', '--logdir output') + + +# In[ ]: + + +from detectron2.modeling import build_model +from detectron2.checkpoint import DetectionCheckpointer +final_model = build_model(cfg) + +checkpointer = DetectionCheckpointer(final_model, save_dir="model") +checkpointer.save("save_final_model") + +# secondary save cfg as pickle +import pickle +with open('model_cfg.pickle' , 'wb') as f: + pickle.dump(cfg,f) diff --git a/binPicking/script/detectron/Train_on_modify_COCO.ipynb b/binPicking/script/detectron/Train_on_modify_COCO.ipynb index e28465c..785ffec 100644 --- a/binPicking/script/detectron/Train_on_modify_COCO.ipynb +++ b/binPicking/script/detectron/Train_on_modify_COCO.ipynb @@ -2,25 +2,17 @@ "cells": [ { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "source": [ "import torch, torchvision\n", "print(torch.__version__, torch.cuda.is_available())" ], - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "1.8.1+cpu False\n" - ] - } - ], + "outputs": [], "metadata": {} }, { "cell_type": "code", - "execution_count": 3, + "execution_count": null, "source": [ "from detectron2.utils.logger import setup_logger\n", "setup_logger()\n", @@ -42,15 +34,7 @@ "\n", "print(\"finish importing\")" ], - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "finish importing\n" - ] - } - ], + "outputs": [], "metadata": {} }, { @@ -62,7 +46,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "source": [ "'''\n", "The file structure of the dataset\n", @@ -80,7 +64,12 @@ "cell_type": "code", "execution_count": null, "source": [ - "register_coco_instances(\"modify_coco_train\", {}, f\"{DATA_ROOT}/modified_train2017.json\", f\"{DATA_ROOT}/image\")\n", + "register_coco_instances(\n", + " \"modify_coco_train\", \n", + " {}, \n", + " os.path.join( DATA_ROOT, \"modified_train2017.json\"), \n", + " os.path.join( DATA_ROOT, \"image\")\n", + ")\n", "#register_coco_instances(\"modify_coco_val\" , {}, f\"{DATA_ROOT}/jmodified_val2017.json\" , f\"{DATA_ROOT}/image\")" ], "outputs": [], @@ -88,21 +77,13 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, "source": [ "with open('./modified_category.json', 'r') as f:\n", " NUM_CLASSES = len(json.load(f))\n", "print(f\"NUM_CLASSES = {NUM_CLASSES}\")" ], - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "NUM_CLASSES = 7\n" - ] - } - ], + "outputs": [], "metadata": {} }, { @@ -119,21 +100,70 @@ "from detectron2.engine import DefaultTrainer\n", "\n", "cfg = get_cfg()\n", - "cfg.merge_from_file(model_zoo.get_config_file(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\"))\n", + "cfg.merge_from_file( model_zoo.get_config_file( \"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\"))\n", + "cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\")\n", + "\"\"\"\n", + "Model Zoo Link: \n", + "https://github.com/facebookresearch/detectron2/blob/master/\n", + "MODEL_ZOO.md#coco-instance-segmentation-baselines-with-mask-r-cnn\n", + "\"\"\"\n", "cfg.DATASETS.TRAIN = (\"modify_coco_train\",)\n", "cfg.DATASETS.TEST = ()\n", - "cfg.DATALOADER.NUM_WORKERS = 2\n", - "cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\") # Let training initialize from model zoo\n", - "cfg.SOLVER.IMS_PER_BATCH = 2\n", - "cfg.SOLVER.BASE_LR = 0.00025 # pick a good LR\n", - "cfg.SOLVER.MAX_ITER = 300 # 300 iterations seems good enough for this toy dataset; you will need to train longer for a practical dataset\n", - "cfg.SOLVER.STEPS = [] # do not decay learning rate\n", - "cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 128 # faster, and good enough for this toy dataset (default: 512)\n", + "# Detectron default 4\n", + "cfg.DATALOADER.NUM_WORKERS = 4\n", + "# Detectron default 40000\n", + "cfg.SOLVER.MAX_ITER = 160_000\n", + "'''\n", + "Detectron default \n", + "Base Learning rate 0.001\n", + "GAMMA 0.1 \n", + "STEP (30000,)\n", + " GAMMA : Learning rate decay factor\n", + " STEPS: num of iter for learning rate decay by gamma\n", + " \n", + "MASK RCNN PAPER : https://arxiv.org/pdf/1703.06870.pdf\n", + " Base LR 0.02\n", + " decay by 10 @ 120k/160k\n", + " \n", + " Cityscapes finetuning \n", + " Base LR 0.001\n", + " decay by 10 @ 18k/24k\n", + " \n", + " update baseline\n", + " Base LR 0.001\n", + " decay by 10 @ 120k,160k/180k\n", + " \n", + " Benefit form deeper model\n", + "''' \n", + "cfg.SOLVER.BASE_LR = 0.001 \n", + "cfg.SOLVER.GAMMA = 0.1 \n", + "cfg.SOLVER.STEPS = (120_000,)\n", + "cfg.SOLVER.WEIGHT_DECAY = 0.000_1\n", + "\n", + "# Detectron default 16\n", + "cfg.SOLVER.IMS_PER_BATCH = 32\n", + "# Detectron default 512\n", + "cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 2048\n", + "\n", + "# Number of classes \n", "cfg.MODEL.ROI_HEADS.NUM_CLASSES = NUM_CLASSES \n", - "# NOTE: this config means the number of classes, but a few popular unofficial tutorials incorrect uses num_classes+1 here.\n", - "cfg.OUTPUT_DIR = './model'\n", "\n", + "# Confident Level\n", + "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", + "\n", + "cfg.OUTPUT_DIR = './model'\n", "os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)\n", + "#cfg.dump()" + ], + "outputs": [], + "metadata": { + "scrolled": true + } + }, + { + "cell_type": "code", + "execution_count": null, + "source": [ "trainer = DefaultTrainer(cfg) \n", "trainer.resume_or_load(resume=False)\n", "trainer.train()" @@ -156,12 +186,17 @@ "cell_type": "code", "execution_count": null, "source": [ - "import datetime\n", - "# Inference should use the config with parameters that are used in training\n", - "# cfg now already contains everything we've set previously. We changed it a little bit for inference:\n", - "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, f\"model_final_{datetime.datetime.now()}.pth\") # path to the model we just trained\n", - "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", - "predictor = DefaultPredictor(cfg)" + "from detectron2.modeling import build_model\n", + "from detectron2.checkpoint import DetectionCheckpointer\n", + "final_model = build_model(cfg)\n", + "\n", + "checkpointer = DetectionCheckpointer(final_model, save_dir=\"model\")\n", + "checkpointer.save(\"save_final_model\") \n", + "\n", + "# secondary save cfg as pickle\n", + "import pickle\n", + "with open('model_cfg.pickle' , 'wb') as f:\n", + " pickle.dump(cfg,f)\n" ], "outputs": [], "metadata": {} @@ -175,39 +210,44 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": null, + "source": [ + "## load the model and the weight\n", + "\"\"\"\n", + "MODEL_ROOT\n", + " L model_cfg.pickle\n", + " L {cfg.OUTPUT_DIR}\n", + " L model_final.pth\n", + "\"\"\"\n", + "MODEL_ROOT = './'\n", + "cfg = {}\n", + "with open('model_cfg.pickle' , 'rb') as f:\n", + " cfg = pickle.load(f)\n", + "\n", + "print(cfg.OUTPUT_DIR)\n", + "\n", + "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, \"model_final.pth\") # path to the model we just trained\n", + "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", + "predictor = DefaultPredictor(cfg)" + ], + "outputs": [], + "metadata": {} + }, + { + "cell_type": "code", + "execution_count": null, "source": [ "pic = np.asarray(Image.open('name_0.png'))\n", "imshow(pic)\n", "#pic = pic.transpose((2,0,1))\n", "print(pic.shape)" ], - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "(256, 256, 3)\n" - ] - }, - { - "output_type": "display_data", - "data": { - "text/plain": [ - "
" - ], - "image/png": "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" - }, - "metadata": { - "needs_background": "light" - } - } - ], + "outputs": [], "metadata": {} }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "source": [ "outputs = predictor(pic)\n", "# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\n", @@ -215,32 +255,34 @@ "print(outputs[\"instances\"].pred_boxes)\n", "print(outputs['instances'])" ], + "outputs": [], + "metadata": {} + }, + { + "cell_type": "code", + "execution_count": 1, + "source": [ + "# We can use `Visualizer` to draw the predictions on the image.\n", + "v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1)\n", + "out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n", + "imshow(out.get_image()[:, :, ::-1])\n", + "im = Image.fromarray(out)\n", + "out.save('output_name_0.jpg')" + ], "outputs": [ { "output_type": "error", "ename": "NameError", - "evalue": "name 'predictor' is not defined", + "evalue": "name 'Visualizer' is not defined", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpredictor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpic\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0;31m# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred_classes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred_boxes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'instances'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mNameError\u001b[0m: name 'predictor' is not defined" + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# We can use `Visualizer` to draw the predictions on the image.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mVisualizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpic\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mMetadataCatalog\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcfg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDATASETS\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTRAIN\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscale\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdraw_instance_predictions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"cpu\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_image\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mim\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfromarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mNameError\u001b[0m: name 'Visualizer' is not defined" ] } ], "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "# We can use `Visualizer` to draw the predictions on the image.\n", - "v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1)\n", - "out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n", - "imshow(out.get_image()[:, :, ::-1])" - ], - "outputs": [], - "metadata": {} } ], "metadata": { diff --git a/binPicking/script/detectron/name_0.png b/binPicking/script/detectron/name_0.png new file mode 100644 index 0000000000000000000000000000000000000000..be2aad732b69ec27e5f90292b353944b9765c89f GIT binary patch literal 89413 zcmV)XK&`)tP)goSI})B*Cg;6_5ZD2_OgL9CQ5Rd-oo9*TUV;y%CuS zph)UfR^X36-n;MKbA0&x^Wo6{{D1a80YC;200IF9gAp-I2>@Uq5CI?p%-^m=#Kfu! zGhznCKbk54Vj_^K8JS-M0Fi%>U(L+SOrddY{|yKP=E4B`?Eojxln+RY;1sn^ZrT~bT09B=aeagonj<1KAuSWmgybUge zkYDQ8jmLqQ7A}61u>ofJ9sqdzr|<2?c6<8l_cx#3*wB?V4hGsQ*snAc6avSlK`s6h z|4*$MnxZL2MN>>_lVTK%ib=5(8wH1z4O z2t-WU8i&AgaFTM&>)`@5uMDe-f}1?+|hV zk!uYTH1D3-zjKy%hN(Zi`2(AoD~HPffk0*kX1^cqqT}zeRLFb`W#Z+-r~GLy904#g zvJ_IJ>MDmwn&By3F#`EY5tC``DG?cr;D7qclc~+0^hGRr9kW=;t|LVA{-#RiJ;XB< zDz4QqBPPIaGIaLiPwtJA)LOqOpHwC$5Qu@oU;+~g`)b0>h(K!nyTAlBa97y)e*YgP zFwwyF)#D%E*?;igJ-WXC;x_@hapT&_$r&@(T5GKUN-2~tFdLQ^3}tp5Fc=p$lQe>C zu}mq5A$=NV5DckQI)h>!=W-h~)8%pG6zmmmD zg<|zBvbgEAgxc+YXt4i3v9XfDd<%(*N?LcLh0_ReHZv%UT{is_v(;oK;!u{9uuw}F zN?}xknJMO_H3Fqu=+6%j`_iNa5|~Kwlnz*Yf|zox3)o=l>PW=79yWWJ5CO|o=B{QS z5DKNG;8M(%PD(AL5Jp5il+=TG90f=Ls97oH!;e0B@bGvThBBV2g|TM>5jcdH3iD>O zCL%Qx(I!eF5QC}#aE;}s)+O{g{-Z*8bnouhw|7tP;`HVlH*Q>KuIu%BwOW-@hG8hB z09dcr<2XKke0=lfjg%@)J3>s1nl97&-YJ1ql}eevKk!0WK7*0lpRgPT(-h^tTGYkE zCClx!Td4FvzR&?JKrmCH(iJsuM|=r!DSngHOex_9Q$Y%Xx!~k7`CDe3699#XKvc{W z>K;u}N76iWw}Gyw1T*jq3EWhKQdu_%8pfEBdOq7)7?_Kxxd<{Bm|93}SiXf(4-7?N zJY0NjVWBB34TB6YR=4hP2VLwFr=&w8W;0{oqkJ4URNS#l+8{IE>iD|Ee2%Qb_L~_I z73O!|dH?b8*?PTZ=1Y27xI5PP8YXPk>%&Wjdz-yF48%N*)5UgsezDz+yKxc|QBfu` zvw>j-bOV|iK?G5}aHSj_>&eMmdi>+vcy0aM8vur3a6hnGty*itFc8t>$HyJK8jOhn zBQTLmv#Kb;h&i{zm>C8$B`#sPX15$VFyod)fZatCxVvW{ws^LH-9Yybr7a<1gXOYy z+r>oIaV7y30MvEBO?2D|6T4pq^9-aE#*T+!FfwRbXak_CW~Sm&&!rO2!A0C8!Au-n z(X0}_0qXS5XT9`q;{|_^nsi}cLr37M7HOKZdWbeoa zP&JsUxIJSo#3?@HcH5U-0WxEn!zfyKO=TB#GwN{HOyP>Asm}_k?h*T)JU%{i8_dj? zwOTci)n=~@r+iinTLVNx&&e>bu|W)#V?SJp87MEU4&T}} zGXwpV&7=ROt^w3qnVFeg^qJYhHd)?hh#rwYQ{XxaW6Lz7K}W&nu(}I703tROn4u(; zboc}c$3@{*h&a{5R8_;&QmHkhkthIJE;A7aNN|Xe0MJwoaNVd>DltPlI6z1784;(5 z5OJX%e{}hZ$d$+pR4SLc@E7hD)H>c*1r=siQ&RzmN|_;|nyH}FIgm&QVWMCnE^fq` zYkWWF$^u|!W>ZbvvrvNh(I~{c6v}X-KT4{p0U+keW%d%Tjv5cnL~8LY232VQp@er& zNGzZMa6AU6L~Ns~G>7uUCH=G;YP|dYhqQPdOvf0=v9_AINgTy!655T!dc9Ix539Ys zy52M{6;uptxDbUlAu>uUyJm>N4B9~lo9s1W=ODVhE4oCrU z4Y$-@HKn$Amb5({)mh+puEort5|%oZJ%TFEeF2CIbs3w3XQbyhV&=ky280q5U{FDi zoS+8c!e%g0M?4fzD*P04^$o3V1Hr{iNWR~k~1FWS~15E6zZ3L)7#L;4R4=_`i z{8=&tAvY;t3L69X9b2+ni_pq_j77Xo9=67ga>NV_B9FqEhK_es6~kGxvv&YikQl^F3}QPK&7d{DWmUio zY9YoSoZIy^4h9f`87~b_9@BNV`phgM{xg6_kDd@0)!Y|V-+4#;AR=A_925*7Q_0AO zVM?l&acjh~W~$A%HEz?*xv4KHC4J;TS!~%g1_YwQ%nOdhxD1psR^vNB zAU4(vy#dt>7yTyS`e)2Z7O|*2hRny4w0>pFuhymJV2;w$D5V^x$O!9WCs>#~}th?#I5g0@mq zQzGzFY9G>zqnX9+`*QJR}>^{5&iw!5?H#?8WIGO@uPKR(`UHvWXu*%OPkhbSsh zQ-@73QQJ+R9hYSdis&bcIk@rNYZirY^Q1gYt~lEvT+8SpICWN63iM$VEHCAQvhI3@zSf z@yKB>TnZxyQ&AO%vlND(12+;3CaqDTwQe!pYZ@(=Iz9q*L~Ow_EznjnJJzFK>#O3C zTwXh9K{IA1=DVNWMb}p|x_Dw?&PfbJnADkPk7LBJGr)*NF&)ZkwK9;0uXUJeYhx2K zgNtLDK^MYRA=de(T`⪙o__30lNRWSkkz2Gcoh%r$0#Gb}oegQFSQnJcOxpL~az_ z_7y-)HP+lT<-j=u5V5o=TqOX`aI~*aDP~GN{xe07tpPCi&CIsx2QcP;iVOH^dU7GKm$6$=qr9E&yb{b2F`0~Fr5PC=;R%yiS?A@bBkJDl}BLrzs!=xyTW(FmhUxzt)~69c(qEH(om;H^jcH(oD~E-asRPF$vfVwxrq@!zORn5sdE zSfv@We~5^jJKz$5slh;ODi#_O(X3l(c7zEAX@{ywu>h(dn0ji`F>Xm0LS||}#vYai zlTFmr$N?=Ue?;kH%|rzuu1v5>8Pi0rDyh|=NBrYPFzPUZ0HVkRxW{1U1T8{RERc!y zFcxM8=0ap3I21vGned(-o@Nx0EU%dhe#D3Y!OYZ}AuUt5>=5RTB17-O2oMu$S2d_I zin}|Qsz^jhHd~r7X_0;G`tJs-fIRgP3Bc~Ug353_6rY(vY|v&#>~C=On>8`$H2aEwFs@!EL!^UxPDSdo#RWG5r4^RMtaN1B2_9q3=lP8_Kd^gA<6&O%=aLC zpa6xqENII3%FZpORao3UGoiQ|n3;QXm*!GxAenqS5%Ypg%rBU~3e52!xgy2@F+_u( z3~h~EOMH%CD#T9gI)w^kJb_NIE%wdhndNiLo-TA$iJa6zDJ4w+mI~n46tku@DF93g zLaoX1($xuBe)mp~TLOq$qSr#>+?S(OxBpf*5z~K`F&ez6G~o zXJ=>qmJo}~3tY-6e4UjxMA`|=#h|dPec&#ueeu|1E#HlJ>_nI;#YUMu1u;Nsjj0k$ zVPui_uf<#EFqzi}8mwr){;X2@NnG|9uY?IC^ zX=KVMN=l&vYLj|cpgIggWmZ!a<-#&Ghvy0>C_+3Q@9ZioQR!~awRW{y?RL8b(zbp- zzGBoL#C6%@$a&$qt4!mce1OM<0s3=+(l;@)lX$b#C#}?N8F_LEFeY|rNoJQNGZ2kIC01%+YDJ7oen~70!jTT-2xwye# zx|vHe)3qfYM6ExFdCdv;TXb7WJ-kVN%=-2c0l0hbeqcEZP-2Bu3$Jw9#YMHecaDCfsHX<@jQ;&VSKE!)1 zHF{WoT3@J)&>YNqHWeTvW=ECo3IoJ56${oXq&IGZh$3cAJ?Ta(9wdY+ z-w$-l9hF4^f#SY668fRC6rQ@190_iMzb9JCZ!hKv@1t37;!>di=8SUOW~`K|>zgfM z=lt3J8&d{ZkARcVekH5oE`QH9&X}G97Uy8RS-35_)zm?m@l> zDk0)73VEONql7La)^E1w9hMkYtMxdZbsbPA{1paF8aC4yrYPif2o|Xl4)#RUTGKid zCskDyL`@w;h7Q2+cmMv`_9(;G%_wH@C+t%7NP3reyR!W)SL*U)q3!rm@ zGor}4oe)8w6QeU#AJBBh5&3a|dcGX#;rrGGkff*@?jKB%R85)hAIE|b3O`u$);X(% zT{ztDC_K>am*|nE%XGk`*}%Dur~y#o4g^1MROY`Dsz5OXQ{H)uN$U;1m z_wOr^4m0c+v1R2s0goR)S+CcsDk99B`797HNj#4kz%YgZ3>0@dN@}V`#Y|MIYQ};V zIG{TLgBdo3e(zHOyk!Nb^TjUj+wB&>(b3`JF(|fjE{^Qmgoa}~3hU22{CJxz-rem@ZG(~z|SX$u|X5RrBopiSczwQ%~A*N8b?p4FtT_agx6fju>nWDM@B$(Hrx+_2r35=OJT^Il{Q;j0Wj~@sV^>`rH zBJAzn-d<}>MBEE)Hk*r!iyoRM%YocA|7$|T*bFxoeLFgJLE(MDPi8!nJ z_tFmM*KZJb&Dw zPKX7_^{3x!)XfGDJBL;aFv7&aAo$;5?*1%9)^m-$CTAAi8OP*?Sw00w+46ZUpTgG^ zAgGr(Bf&}f^}L!_F)W{gy2G7YH4>r`ni%fF6 z;nt9ewCKn(mM}=&)z`d_B-m7`)LFJc3G_pNye1(05;LVabSWe+B#4jFQ`^t^{9Ty* z8<`pG3P4GbJBNY}3)1R#cwlB~GJl1apk>A$X55=Fx@S=9PhoQy5?oBzh49aDWp-B< zOTmCDY~v z!Az+}G#qeLDx5!;d`IS-T&tQc^O$2y7T}90Heag-wuIP)vMw}CAPvJU(;K< zCa40$CCrdN7lDxMJ9V^gp5^uBDDk%ZS6|}9gJeC>cH7tANwX=*%RBt!Qp;>GgK?T= z&#XrjRHI!rFW9tI<|sIe{BAsAZsW{=cbkA2vCY_!m^y4rQyG6{X6p@af$?T^3Cs;) zfWRyMp{B}QD5@>G4PY)Dh%>OZsS0!HneZ_DF8K?Po%hynjTd18b!-)YDt?u=gTRzM zmc$;o$Nms2LOIhoXv(0v(et}W-v$sDu0(ipe6o~&A}V{A+G7)?QEimiDz|B2`z^xr z@{|IADLbby_nW-+NdNjb>Syo&^Ok(5CPf*K3;gVnozr=*|8WKIW&HLxU-LTlsAn*Z zr$}R*#qgmXDpI(YyZ81N3u2zUvQ2x)mnc5UFH@q@VQ>~|#nvyxDg6A*2&IWKWDqu6 z)DRM7=IZ2e@0F%5#(qQ?Bu}`W%W<|JWz(8j92%nA?qR1o+XhZ7>hcwWrdN#l;ROP5Va7;?%)Wx768LhCoYUshB`9#Gu>yt_i6H zo`cKK!fGOl*gJ2IF;f#Gq9B-6h}j}V8?`#r6GItY&UOY^{60&WW{u9WpZw+&Eq@Oq zx*>B(1D8uO$6inrrp@JxjJn$(iVAzvW?paOYMizn#6-*)%&7@Xy(4kf&)HneY8&z1 zLirO2SO;mCuh6|?FU5kX1~wyN>J88#XZIgGT-yFw0alkpPj?ER^1!)Hp{D3C*;`Yo z1Y(QwW|uDle(!45$fygV!UFm_!?$LA{4#SizKuh|AnH;>kMp&Ph!wCvS-6ngX*RW}_<8|G^wr zgo3!3dBdjn&XXsey56|tG0)-JET;kLTsacLgI7{wSD zTc~q!H5iMnOTQ_pC{4#HY4!A0sQ#UW6swS2zef$0F$6P zwlRXj)u+RE|G_BA$FgrDc)L4q?7e3)o|lGgP_`_-iy9+t@9-hM8~< zy@SUR_gVMh0Xm2v>Nz+<4nhExI+zKe5Y=oSOKd=dB!+XrEH=Uy%-7S)`hqB9cP0ro zkS2-}B!{4=YV^!7OpQzN4Xy!f4GK+#>#$<(-2#^m(|o-X4jU24IPy>tH8$>LGzyB+ zxI0b2RMd3Qs@#nxC7b47t#fWD9_m;Wf>nVS*kcWc?*t_e2I7YZQpkLG1QEUa#I`>;{{{O^&FEhYqM-v_0UnggFtpb4 z7x0$IZ+`PNBKQ=eL=9Q4rW!e)UiD_CQFt7in-$XmZeervXaHTnq`_+|EG%jG3ct1) z!diGC$lMcxu12xk3&K=;>=(c0T983tWP&sN^SAmdIz>z^x(ZyosiQsXS`UJWBtwx- zfWbUC@*^TpaVaH2HfKac6jj}EoBo|?!c#fIW`<;Vwyz}y1?Hd3M2L|cNwEg;IPnM} zUfsPq0x=aLGU5CurzpIc6|{J8-_&lJyy7(<0pa!39W5pEmQiNb5>)SH@RP1ith;^)F2L$KRh=QvuTOFBBjKU?^|_ z*~t!zS1IN4MegQ{t3JlbsDqWEoD9BEdCBaqrqY6DILr>)XU^;l)za$9%a-NUIorQm{1wm zM|AitN^`X|vmDxGNu6Y#!A&i(S)?J*;~&jV>cpL$OQKY-yZ7!hm-`PNA|D|@pv7eY zd+_k_as<^gYT&7WQ7~zT^=3Bb+9Mjt(I>tpE^l`}(tq-)ND966(Q)QwcD6)8YB*L= z(jEhNEzs2~mlx*N(WSYUHZ^94E|I@e6}KI3H^bT_BKL+TE|sW{Arlqv=p!?WCh3e$ ziE>I!R>dJ~r~46@iUT2oB>keDYo76^r59er#5GlrqNxX+UJcI$H*sd!@Ma;+erp^p zLSD%m5ne`|1kymj?(DiXNK8*go>{XlrOox2!ZZ4+sG+AckqFlF9R|uAhZ1ErJlBp= zQD^qek+bvDh~XLPfnBz@nk@@X5q!vE=`Do-T&ni|%&=NnkOzX{CvX4!-S<9NUf~O0 zrIh{s{bd%cOaE|ic@(TmRD&{sb0(F`ZsED=C=;U~!qp zfxRq4T89T7@DY(Wlapl{h;qQ7BCwTt#W})35awb)fB;D**aHs@sN_>5U{OS5-JGXK zbsZpfR~u>fc;GNPUVUx|Cm+x6?+QymL|hhdCG3Q|DJl#l0+q-svDeGQoiGf95J$>i zQw%C#{?U)${`ouachinpe4c&zAgX7uPyZs@JF<47hEc6Di6zTv#zdlle>#Ijz$?Eq z{K>yOWuihf&fq6X>zZQHkp4+$E~qI=X}d%~W;RXJYp;IY7bGx8D8FV|6bv!W#H&=2 z^NlSqM3OjUngU)#MCXOZToRm;|AV{V1c*c+reb{y%zK~0Zkx0&9qCCrQ5$z-=A9&1 zcm;2Xw4@*jrebE^f9SgF={Jq?8wYZky`};~$~|+24X})UO?%N_bWSjnmAEp+Y#Qsi z%?QZd*|AT}3m!XsPdpVsu;><@U5c4OO}yEnWc8aPa_ggQRU^9ZZA^%z@;YiS@{4RC zCFITRL`;vMzPeslk!4QZPXI7`_~d;r1^zc&Jy)WDcB*mmik)c>d?_BeQw7gt%yDi`^2Lh9C8=BVrGn{MreWi-TN>stH8GWi~8{ zIOfWwWcPxfWk;Y4DhY`Q#6J7VH>^i1Tq+gj;#o5?lYkVeUEG%iEh6sCs{AD6{sO!}gii1-+`aM{MZNyEn6I=)`1lxvf!$q?T|3NScc7k2QuAwQ~)34WGekWBo8d46g(u22m99d486=Fi>ZH(i-*+ z=(Pgj?t$F2u%!(^jWqIt9--13eCH2vCM?ZEKJq80oDx8QkG@G|rCym5Tkkd25xK>J zcBSwI(7frz?OP1YaEYRRw+{Fy2I<6KeW6PUe}goMl({qbjGq|yG8f_}E;7pj4qVuq zwy0D6asR-Q3QMaOWhZHU002My(NDVVf9BO%%f&A1_1f{zG5|2M^Yim7SFTV7@sb$5 zr*El0A4RK(fB9egO8LM3>*;WqVZFcKTPwY!!ug*TF@SkimYcX&r{hzv7TU-8&ijRGCN3S}(~Of!BUDh#6zC!Bdq!)+S6 z;b@uwpHt!wgaCc9x*_{6k)};8M{pYQs)SCMxFd{3G}VbPaQ5jMqR+^zgCbfAQ!a_O z*zv+z^cx8kh<24U?y4<6n(qvM5#)xNs)~vb(8n#tBlAK@0A#0ct<`%Dy9ZdZp8&>j zWab`*64C15@>U=PD?(E%L}F2!=RS)O#QIT#_iuzpiCHOD{sS4?U> z9#G&!+l0x|`+n=2ucdtnZ%qa>F_lt>pzN4({k<}`c(G7Yjg~#7V36e;fZ(hXONOya znQ{?Q04kU{Mr2ewzf4tI^Knm1Z*~ zm=EX_z(BykWFh_SY3UyYJMA#Z9tZ)uTeofi_|4yVBi1FoTFR}HA548LhIz@1 z27x1Dtfj8H%xcciwk8N5AQa-z5fRbCAQ|G2ht617{>y@u>D!h7K6E8XjtJz)GU9f* zau-CPNF2m@2{F>whzhf&2r@jIrlxEhd*4!8qTCGW5v3v#1s(?hoa{c^Ab(_ z(xy^YUQl3asuJ+gFQgJdk#~O0-@}wspaEib#-`J_FywSMz*IoV2qmWNyamt5BKAHP zeb(62njj)T02fn%sfv{t3x#|49#A%wyXXfDQ+0&n?9lb=*Do$EC|~F2=gadckf1hd z*An^{iatpv9A3Ok|M@rS-~L(q%Arp-2)S?Z=Pde>{8BWvrZG^jsjaTu`sO#kee=$9 zUfXJBQBmSRU`C`qvS2a5n8?d8!RdFIdwEO1w{4@m?KP^ z7eAv(^a^>AQYu2$)bBPijzR_O#c!qK>Ie!E;u5o9&4+u1@s34?DOAj)hx_O#p0f{q z4)gkWAB(6O{AY9A`sSeGp{Kr-&;OGqIUQA8?u@1hwT=lIGODM+=l{ z2^Cdw9y)I-o*@gI0wDS5moO?8^G{+0G!`m3Z07mqsR#XD?g)!Qrdq2@zl(lrZ5+q# zcI!1Z`}_MRCnrn($7>KoaykUeuXr$dK1yhP{g85r5}r%~yO;U^p{4>8GK(VBs0;7`In6~e1jDSFikZsT z2wp_#2t^b|AdyDWh>FiK1(AvnGiN~Wj)<8tc~fhShsj1+ zZ4*FTD+j(u&Xj~@MV;lrFzz$}^|A|J&DKWuyVJP!lVen^l%mp7UXg_YX6exl78O3W zU_BhVdhqA*wt5kK^oGtD&2L`CFqN|U^s{@%C#Su(*;}hb#O2!sy_9l(e!gC>yV+Q; z*WI)4?d>f+fM)lf0iviJ7V|GYU*12_kIt+IADv(5{{eHVtN)S^rzoA)qxI&>FFyF> z=kNUDr59iL<{PiZFfssSDgt6_tskA{D^CV(&BvnUU7D%5=~ihb&YMU2)@}cyA`2Zk zgza!nsFw~~kQ{(1)H?@M1%^_Fbj=3p?MG;B{1Dqj?!}+X~f$9*e?75*f)?rQA!|sL|#AZIFJomsSBC^E+ zTppBrVB*FmilZVW2r@OFRLki_dnukFnlo_!!NVWD^>degAc#euu-RqZ9n}8*zD5x| z0dOqjYrem~zw`imm#&;E#8A_;Pl~_(7lyz4Psdw(0a7}2qx=skXh==)mj}?5uoX-S z2K+3HUwrUMt-~8{yqdVx5S2D2vMXf($XnL{m}x1M7b{P`@^zJ;>WP8~2r1|3o2d^5 z?^JzMmFsLOkwI^y=haL!XTK#M5sm&4fR$2-OP`#E$d-zzmNEd*cLy?4DW&#C*aXTH ziU_k>>&%UE4p8Cf*G!&~V$yd~1?cFW=E1y{K?zX|2!9BWOfoidgXMgj`N1KGyg-6!F4RReV5IkPTBpAd4xBq6?JI+R7EMT;G?p^dVFh@9D#ZXr>M9iLz<^4Nx&Glc;-s({QpI%j zId=5>x9l&XQsG>InwdJmYCsfDF;~usF`l?q5wEpUQ5E7MvCt43nF^OWtkwoIkxZDozGFXbo8orNF{m&G+uM%1Lq}HUj$z5&nA*=Bbd6`{lySKYr_H9^C@) z`Dzyz7qQJ|wbu1|?E&tUD_2@;bKSqlg9i_8-n{AG`{tW(b`<0vf5eyn#_iKd?UOV7 zlV8YQ2BIBOEG!VnP`px_sE|od%1{=SDN&4+v)yht41?cQt@Y!NKf8AI^7GH#0Zc`# zPv`a+RIHSNxiA-lsWu{_QZrz)2n~A}W@Z*-M%PUvaH&WccD-D%G>AekZO$xpw+ZrM zvzX-uv%#M=mSzSFclxe zp(@hmq>ayDD2v`-%1jj@#kYKZ*%9nySXoDy$*Z80s^(_`B@e+oFFO%bL#1=m!5{paIEBmGdL=-V=W8@WiCcWh*LA7m?9z$A)4Fv{fGmm z=uIueSwrtK>q3_VR{tUSN&sfkqC(KS=Uu`wWX>Y$Q1K+jya2&m$NXSmuZd0AO&xTF znudd6@1X_}m%?>mt^jeVy&EG0lp@y_0gx%8vI^N1;q<`dO6HH2>YWj}8nP=q2_J7S z_Ig1jcEYr?y&)3InO3d?X0~3h{p8){%RRtCi>ODy0IpxZz7&1``QFKPU=X1aDxpxE zZjfUj=CgzgqcSREV5}JHf=$7u(yCC6Y$YNHoa`W^e!WU7z#-CUBuz zOQ}RmTq0`EDYL%8$c&bS@QW{d@X8>S$ux%PyMIJf$OY}=QbDExcwEzopc=V$=FU4K z%C&mm8#1jF-I7!)6((jbg$g_D@uC=4Ujl_I4XBPVrTu*%NNXysmtit-se_{+E_EIU z5Fx)iff$`Yu1793+9&fxT<$WSkL06%vRoE`nM(5^0v=D%Y^crUkuulq9lB=VQoPdw zfV4@aX&hBX#9Zs(OhK$g9^a%A;b}T!qC8#-?()4}*fVn?KA4-o=+58z$`Mux!y)O6t{bu&Z_TsfQ5u5q!lX#b!=6_J#2rqII+G1$?OpEI5?JXK!F)DsIx#w21CR}3E{ zX5DW`KF^(IPT+_#66;)}hO==TqJCx(84;D5l^q-tU(C;K4tq!b5f}5too4{=F3#)W zVwz3*%IG-?u1GC9sEHPLu9i_j{L?cD-Dh5oyg&W))0;PMb_n$9tFPX^efz~1Uwre; zH=i;801gfgT5JDc`x5V81uztM0x=P&xH1r!{b#B~15;Ue1`75HqvFHq;1Lc180_2M zdgIp38@3pi1VBVS`Q+1(F4Yh?it}$=aO*Gxt(UcfsyZhRgOPAn-fN|@?cK{6!r_Fx z&Dbt%r6R|!Bik?nz+`c52!x77?=OcEMeF~39XJ&Z07`^f2YRY8C&hn(!pY-TjxnLh zBGDqxmtl*E;KPVAvT?{X*VlAL4|Br}<&2{g zNi7&X{Ip*hc{3>k<@H&G{c#0Z5%Nt2wN<2me^|&IUl(=)(-|*l>wApNyBK3u@ zKm5Z#ymsxHhKpc^?!MK?Qyw!X+`{u^ykktoWqQc zyxy}C?zi>Z?b>~bg1qORyJITf{?==?)};dH^1=bdEEX}*D5NE2y>n^?;%@R0@CZm! z#LWCmG<0PE1K=p2fUms23`WVEhM*&mYT8^yoYOHg7U{+UT>1?Lh>icP{_e`9sP=>f zQU${w0ol9{cb(ZMOl+EM2p-Nltr=x=Ff*v46!ys7^3eJWdX0jcfT=W1&FPE>^%X;K zjs8bKPeOc!WouCqTKYiGaSB;s1~E&U;(jbPk%^7ehk>Xn zX*Q@6A#&%^Wys?JPj!7>?pLc-KY8xWH{U!uI(qQn!7vQhuU}uU*IfseYH)OP^uY%o zIBJ^qZ~U7b|8W21=Qij2_uqQuGVNEKwkni~wbyr$#PPA7b{+kCwK%Dv;qWmpLb zQJbd8xit`d{K;p2EVhejoCyF+l50s- z5tR}%ES&HoKoq7Ug2Hhq;&7tYWWu~P9)?Uit7_rm2iij+VSqqc7Z;&Y1j?Doq~I+C zv|82GY8Bt8a}kk6$F{>ZCXVSQeZde0$E?Yeylns{&kS%4)8v&Qo~FrJ(I-!yeE!Y; z+H5wKXZO}xFD@?LC8oV^|C>i=>-GBgzWwU=|L}W9D;!?BRBPQTOWphVyYJon%F9pA zp9qOujKg~4nK)JTRs?eUVH9{atkDhG&7E5iiD$@N66Dy*WK zM0gg{fw@*H95t2EALJwHW1e;JjE`cDXq0htk!CI zD0eAkv)TBl@y{zjXZ=el(=0q zaOu`#;L*;o^Xs2)F5h_a+-@4cfn-Tm;r_a06syF5ia`_V)MXZN`H ztUfR1G*j`~=A;Y!Y!@c)o`#O&BrPUxn3yqHsaClR%np*(^4`r8lz}D)X2vmgLsJRN z%*xVy_-LT0dZr0g%ZlmJ{@xx4N@~*tBQC7kWQ;R#oXak&!INS=gs47Mvb)T=h!c*~ zUHpteew?c({?!qVnu+tJ66}{DVR7PZ7z}J)V>l}S5GjWU`E$6Nsu>k4G*pI~dhJI% z0{2rq6BIc`>7(!aJOnpCNP@^yz93K&ZB6T7MHPi98^}wHk)OD}pTYB{A8BnOCb-DZ$52(=<)f#~**Z z*=%a9`}_Nw&Bo7+=^o(8lPCUW%K5c@RiF8Dan#aWQ0Q3y{{8!^`tWQzYu-m18}I&M znx=2Rcq7w)Im0UeyIEo&WV?qCNC2qT0zY#E8YMh!8kO02C|=Z{1hP~BN;4Sn8s__n zNbIEShl^N@VfE||!7A_0w-r=|N~IbX(*`3dg^HT^a0W6W74-uFy`P_eN(87`&v#`E zT1#cmL#gG$IYUM2DOXbkMp1|KT|LtT`(H>96O3d1W3}>;EFvR-l4gSunQ1bQESH4I9jeXVl}`WimF+3X=3q(9>ipj;j2t!MrPdoY{WaH{ca72ELn&_ zVXp!1GZqzUcKy=+mA$jmsR8KiFaD?NLx4In0EdT%{pTJ!cjw@z5}lr&`lYtp?a|SZ zh`3^y+5Z0iZ~fM9eeZkUOFrZ&4rd9vy444;zrX+5Yp=ce=9{W|czC${Z1)mgw-=S? z#!(7vARlQU6hxm9kxzfLpgN3+jM)S^a~Wi6)1*+9Ode1ez(oaXGSD7MxxhUA3=t3~ zRx&f6w53oPchgYIDh75FfxQilnBlovSer~ucwlCVS+UGDNeozw_hL15qTIJ$40Fv3 z$jC&E7`0L%iX(UCXmu(5vmpxMvrs$F1d#PdH}w*F*MN$;Dzy&GMGb&*7#lOmie>oy zJ1<4v>!RZ48}ycAUm}8AdZ)>f*Ga(THMhhBB9mko^8dVzG=>NeYaY!BFUXFtgB0{}#HadC0u#tjj9=bd*P zuk*710qEc2o3PnznAwkeIXOA`_~VaPt5tUh{eH}>7p|)6Z@+RolA{chXvU3#sEEnj zc|z2DyllE7{QmBcdPrnoADDoU5zo9TufMw4T7pj#_y>O{#$ngM$Od<>%+;o{RNAH=B)1Ei<2=pVQN?d-v|0pP#$D{{HX({`bE3J%8tB zvw8C5$&DK~_V)H%XMXXEUzpkT>({4gdh+Cn6RE4!%Ihq)+pT|N*S(JZfAggqi5@eJ zFXqRP5s|^G8K@RAfJHiG>}P)&piuEzFTjkNS~adrS_D49H5zMt^CRuUIUm{iZ_V@z zXfY|o(oTh$ZLl)#rrpIjtOjY;V9x&VwqH^P`E%{I;s4^{0UO3i7JGKq%qH63=yNNKu5i4$SPRz**PC zf2L3)CJhH%3~;+0F$em3s)cd_0Q*NII6BWpSVx~kTnfk-7R{2F0d7#ROk`AooQ_&W zVrDNrV~}C+G%QoC)sb`$I(_nwhXgLfr>Cc?_>$K&O}pLh{QP`> zfB&ETvw!9?eQsU1IO#Z9sim$;kn!P7tb;)ny(v|*1z%6jpwgl@-f>{Q>sRU!VrUz62Md}W_2+;x9+dVxv)A( z%^|3j3)jMx>dG^YCan>*1kWKo9xXkc2?~oJBUsHq-uxZjzzo_{4b4bZWm2wHpe9OY z%vKm4sGMKOZnR-lh80aFT!?^ray}9fGD+GO9|$e=jIvMbGG;Ihm8aH6;2=b}KL zTA7O#EFJ*If$aPhhKXzCT9ueo)Jq$gGVYS_M}gldf1KMkJpeHEzELmz&>Wk}R40CL zyz5QCmK2TafM4wyK{qpA3_-ynoc-lG;3Bu%?U?!f_uoG|JM$8{v$HeTfs2a^Guv*r z$H&Ja;!JCe*~V~*?>8wTFWx+S?S(6~^2v7Bt!A=ACgMC^q0GcbT~HDa);%zpifB5& zn6<~ug)asdjfe-$GJnI)B7m8=6f=o*k7+4|3Y&;m)*+Hk^Vgb^nN6zWxH}tlJ=7*9 zq9E%9Akw5YYYjHGI!vW-tqhCUXQD{L9qa-34A}~y@hCm=O zCU~#(YtJ9T?EcC4;AC8i^)r%vadF{_;VcmM(%N683_ddG88vVz?-%iLLyoMTaU-2# zV&*qqIDkP-SJ9FY6 z!c=7(;Z07_#N_@rsrux$q{(hntF8>C73A8rM0Pv)!udP7Ynkh&PhpE2%{*h?W+Ln+MB zOp1AZI~W5}bh19S~KEO+{k1C3KdM`87oybSgkx%UR>3a<&mhPY)y29E=*4= zw4)r(NE|>!C*;7- ztPjj%oqPcGzV>w z7hM4A2ci-A89qMDP$V`thJ2-)TCfosxy|148tpz=`Q^J`Zd@(^yhzYrUkLR7onM_$ z`jzjbtZe&2O&}s^%>wrl8${i@K*cD}_KVITHD}WiV?8{s5DqwM-ZxB$Bl-}yBJCs@ zV3kWPRn$aKn5lq`VvJ5oKMW#f);M}-4Gb=9D8&q9UV_7sej<>Qqf|?QLd;ln9A&7r zRstC21$u4~)B?SFPZL$^hJ+UA%eg-f2aKmSwn>IUZg4$7O7<#zomabc`pT0O!7gaW zqRHsR$ zSt(izRZmLvAr~p<&=~$smRgcQv_-WocjbyIdGd-XS8a`S-$FqD14=)VBhtk*p12I_>nwpwwDOL;B!tNq0cf`o`#G9E+pfRN^Y(J1&;DMlHd1ni1 zDa<^zsSs0doX=X4fKooiPz|IsM|;7-QE*4F7|0NwTbq`lz?}fo*mP*(D9*y$c!6D1 z`+4I5V8Mt{Lbi_PfrvrGRGJi~x+-oQU1(DO4CHEM+0W1g*hI87=`%&<(3J>4EnEt> zNzAhTr*fHR?fXarqrUcW&mMGld_hRwiJ7ynK2RiO{>}%I^~PpV1{o{ZJZ?c(s+1b{ z#R-yJ<>4b^XuiW-CD(_-#BMJgsxBr)|H@Yfo}ahY&s0YEh@nzSPq#f&H}UzG$H(jS z8o=`}ym;Jn*{RG5+ak{>J6}g=K|r_zuaLYEOLYAiGu}j z3T-9N7S>XRTE?k`P|h@s)94tOKlqpn=YbgU2pQ7Pg z=`GQ-#p;8&5T?d+91(r9F@aDVD`xqmPrO56F4fcM?g?&OTHQN->bS)}?Td)E+wIq0 zdbO5PYyH}buYT>NS4HIi{9pa=kc!c>$i38}3w>w$dr>M8z5CAF|EvGae+k2fAHMVM z&wuh~zx~@kdh_4@;?p<3a`Vs)L>&l0YzHbp`d;nvRtc04h9>b#z<=&TX?vhWo~NwC zFa+o*#Ac=tWM05kE?QIE^towC^IC)??WiFRq%@+K&R&P2Vnsz%#?cC6Z#{Gn>*Yza zGXp@VREG@X@^-rd5H;22OfylG4oC8JNhL)XK+M@`U_kf@|26p-j3{|ai|GR-MMO;$ zKBlfU(WI^x=GfnFFsRNXDo08MK1LqN+*@$E0d8FZJO@w;`$>KTWXu+@%KVt@95i65 z#4wHeDWdR03-%L42J(u)NPb(ESucW;Kds#R57RXLum89I$8o#eKR7hA^Yb$R|LDK_ zyDs&=8eB+}5h;jDDW$g7M2%xeGvyp3gntjtg0k|_ zjgoOKtkqPB5iy7bF+@z#LhyX^O?5Xmj~O#OoneE;?&mqdf$BS|h=O1iX-%XSDy4X# zfdOpX?`Opn!9fGuk&Z^Q2>^7c7b% zkmNBz4S`o-(GS0kf#4E`~*|N6iB-~ZzMpMCfD|I%x(f5(6M^Pm0r^VB|)-T&oq$kPJPf91v9 zZoAuU|Lq_Ciy!~+kId});qagS-d7_Ih?7zbEkbRVN?&*}^SHG8eP1PCc(N==w;(67 zOEE^@c*az%d4nJ22(`#0FoBeSxlDvBybHutidZx%@F^lOs2K#!43NT92$f6-t;tl% zYA7KvgibR#yTUm}kkrg+^_b&A%v_0C4MaSYQnJ*~lO=uA&C|3F>H2;Xn|FaZ1L?dC z6SLFpQ8(x-$mzeKJdfiU1z4?CZ~f?x`fps99FBGzkU|{SiRU4}8yuMcZ;0|nzOZ`1 zX1htIP`1`$Vw8(3m$F(7wN?YEQ!2$92_n;MMtN+KA^Sg)xGBXm@sX%kPuth5-jC6i z1Rn{*fV9a(7&@>OvSc}p$ruLJ>etT+HylsTL`>clF%+u3>cR))_YpmD^n8|=p(A!{ zO{BG$lzJgePG<00KBZzz z#D%*iBO(Q6=Xa9W$+|883?C&OVVy@nCDtihVJX}OA;%_Xko@n8faX^z#VhJ+t&bkv zKRCGb^Pj%?){p+^^GvAD?DG8TB@aEqX6C)Uy^lV4_jiB)FTL`OZ~ou@_J8efxP575 z;|M^E)W~?CQWu^to`<|!J-X6Ciw4pXOj3-Dhy-HqfC=+rfE6ya)YhhG%k#u@WwbaB z!dWGc9euU)&|wON~5gV|8IR1QS%Bi0<-sUHmB_R}+aW;srfprz2va3N?YD)U zGbB!stNHjXQ_o`*O3LWpP3jk~Wv=(i)oZ)$?%usm%q99D4OSrLf)WUco-^?Cy8U2g8`}_HU_C#A%yw> zQxg-lp?LkMn*gMNBcP7E5jjBuUZ@y>BZ*llR0yWcBsFObVZ@ytpmLZeG4k{@N+5DSK zNzEpaNwjJ+$(%RE)n-CEb<8bAy1h9h?QY21z#)nj->GgFOIU%JNcTnQr$6nh+wUL# z;13J){d=GIWoxaTUvpQ$GkyFlk?LF!GcN<;cYgNfXPW}GtvkVjUJOT)WTBE4K+jmj9R`*Y|ymKN{Ph>}!?;j@y#gV5q1qz5<8WR)u zjGA*b!Pt~SL4EqEkLpKKGm6g`m4FKl`BI5YwL1Nr1aM~0HOtQuUgStp3XJZCEm^T{ zbfcUh$~?tGS|5=Z)34lF5``$oUUe~dt;;~)f0q+8W;uxJs!-)_mRLkGDXW4a^{ACN)-3FSho%ZVFQs z5|DAJ#Dy}Y8@1AEQi{hvKr<+eNiD>w0)v{2KB}=NRlDB;90G;#M;|Q!pT0$&2z1dm zu}CrGFvTtw9#3$>b+hmls!-&BMpae#(H_nnJ2CHyzw#I+iKwV@3a|z9$^+`NnmYw# z-SZ|fjQ)7!9dJsSY@vav7OpW$JJc;@-G9?TRc>GO2U|9>d2NvsNJzzz9({QjjF_8< zJ6HDJ{$w8b{7YWX2yTCQv*PEA-U8L-6q9F|pY8;_2Q6+qtDyU+MiH5PLZ{mR_yM|o z`dbf($xnM;>Og#~o2Pl!Tfq5FA25ZCc`Att(A-wI3HLsmuxX(zY~+^SR885!f~Foq zO(D4&UMt6`a12tQG79t5w6Y9^dypF23#)#ZOb$EpCKnQ7i}T-1M5PI-z{g7i#1@&M z9v}9gVnLi}jw4Zl^bTqQv9kx(#F|7%u;k5Mw-yqiuGzUqvGV#9Vz-Wxdi9AviyDb? zILp`(ieeuaX=HJH#G-kW4Fg9hj2{|lgnlTDMNO+irT8+9Sz2a&7`uvn(LlIA)yqZp znU|{iz=);Ve4an`AD*FxE-o(mu>-&RjT=Aw@YLTtwbpTr8vA%=13_n5_wl&CMU?G@ z8$!0sKnzD3?wnF z2qMm?k}0TCCf~wH8Ic;Rnke}(<%JkvKf%QA4*uGs|Rg@HKS=JabVVkBRyiNYOrQ{g;aO7jyQ{ zc!U0)r7eHvBR^#fOc&z|fQjUuq*ywD_mQU}7BRjnun#VW!ra z7N_)>%ph4^ZbU|vnF`4$#uEm=Tfm&iCaSB0Jz!C-V0LzXF(yR#+Tgpr^Ds?r~)qG$*C6&`dEiaHlM5X@K zo={-)B%domP#N)ii()Nxhf~NnsY~Ayw=)0`@e>dunS9MHtF7HGn3?$jguczhGa-C> ze$mb^3W^^-Sql@FVfDrJ`oI3mPZ0B;T>*NZF%fx}<7%~XRMP`Tq~Y%`%05gY@<`zN z(cZ9k8TiyMesJw8h~Y>EA4q0qYW+xtP8@g`=T-x7PnhX3%Dk5h?3}dliiv0#yyx1X&MawQXp?D_COPGR7&aIPkkXl1LM>pJh_$AZ&ou;UDXnsA z5FS>uq`@bk)KV6K9&8$=jcs6F)si*V8P#QGyO((wY2m1tw^)mq2Syn}avWD+26!8? z+jIv=>s8sT>uRVm#0owtiPE(M$XEXc3#P6BCMqqRlKUgGs8q^Gn_{phyw8sdmAKWo z6ksuEAmeTVIq@G0Vlx6rR2x4=63%bW%&TDwQz<3Y)L_4?WdGNB^+LBU{XP73iC5KU z3OxI3`AcsFCL&KPwAPpanMXB#^vV4@*Dpn3f~lGMxe{FbOwGKY%YC*$`y!DMm*|GK z?gHSzl5(B>92-lD$N*I4>K!8RvQy_1Vq3#ZGMETMAy}IPUD81F(2sy+fG6$ATG&G` z>9e6^@Xa*T5+lRJLq2nbq9a@k#m9#^can415?U>W6Pqf8sSp*;@w7ew4hH4jB!wsn z8DRigld5g1H5zAjqmJX`WeDaYvRrxtCTI*3A^;>J+Pr7bDQ$(EobENHsr9N-RDH>1dWMG;O=NUdLPdN#9X&FZt;HP`o z`P6o7e$GvhNuejsv_~V7qY-B#%?I)Ow_EOz#h?UZkTc%eX%VZWLOiuLv^vxhlW*AA zJM+DmY}`%T?ZhyKkF`*<7JO0s=g-Z_O!+xbF}#FKJ(AIYOrR(`$nr=cKg0(p0I!F- zUe!|B|Eq;dp<1a_DwVl_tDpPLRHziHt5OE81DC?pCsf5b3{kx62Rm^z2ZZ_)w_>FDcPl>xXwAl|s*5zvRF9?9r)LfDy+Y-FtHW{dey*ktUKy(^ArW zrefNBVST=p&3@awWHD%@2o60W?Q&#pSFJD=U>ao_CuvQlm|s1U>Uqf+7VBT(xh6A9 zDkds=(A~HP5|+N53foYep%iwJ8wvfDOVltEQDF_j$i&Pf8wnf{6=xngQ3?a$?(y!29~|F39s&IE zJC8j1`k(&Rb2pC$BKUbu3}Qwp==p@N-aG`3{~hGsUw#nA@;6I@s1NsCiu;$5p)UH} z2@ui84^IB@?GO9qK7DvnnC?Eg_=kV=?o*dlWuo7B@mi!lBDQvl$$G}aeQx29_4M_9 zV>(})aK$Y)p@3@Tsrg77gp=^983F-fCL#qO_SuqTMoh&xSa_R?K{z6KOlp;lnb=^_ za1eO76!PG&Em=Re z3kI0C+GgD#%zCGyk(#PSK(E%~lnKjrc^CcM8bVYmc!_7}xd)&4z)On8%T9@sxiN^D zOid@58ic^!a7NaLI)di>-cf9LL5 zYwaf=KlzVdyLt0yFY*Ky1~^yG9BxhlUzHFX`CR#nh^n+t- zpKD}pSi{5u)|yJwT0K5j)tD35W84Upy>pf#jo~dt$g{~jvKLSjFAC+<^MLq1g=n#d z2$zw6p~66TYU^SSFT}NgSRgi5-zO1M@!b?7Z<(CqjQHy8pGhZ>b74FQL!Um`?(OX@ zt*`6az0>x$|LF05@w-<+nctoR2Dn{V3e%tJwTyuMwFEz3HW1}kzLrjai0&P4^O5j_ z_wW7q7Z2x{EAt8d!)~0^)_zP@m?JlMAt1!CJbl^chV(o&Ex;rJ6b+E^1T<7GeiW%t zWmZk*BNb*X=CfMWKu`h|Fte#LgT>k+D4oQ1lTsD0z^iq$ z4`6JnXqy!et^tJJi{libM-_ffoRx$~msh$vqt1%Cd+e|xxL z0-Bi-)~jJP3|>YbVX`lKM(BzD!i-Q7LY}Ar!N+q&bZR+i-Va|0GFt%}Hmg+}o`(`; z`f;{1EEqjAb*j4~4)-VwBc_9cgZ@kZ!CCh0b}J%v|0^;!p0>4QO?VivjQ19@`|E4M zmq|YWpKbUnFV!W_%zDYgGftrA)INE5^0hlxM5GT+eCDgqwQLUdnIqqiSatRGnczF4 zwHnE@+1!8u5Y!u%TukR$!=|HdGcgq~$T=8y^QgsCd~yVnp|E6AU?^7cs3Ngvh7N9j z3gEL72{sgJf~M9cT@8f@b=co;O~&zjwJPgX9ST{ujRuf0vw=#r;)N~_>Cl(Jh_X2X zAlT;c^7_){P8Rv_m!JOh9iM;#1T$I-SBF_SW)O~ql5(+Rf;>S3&pQUOYBBoSn}8ki z`(Q60SRBWXXdTwW-kww06xC4bTUcVqv-$}fj#J}YS0gpirXc(1-tja|^mKEHi-?bN zTkRhR@9Iu!v0Aw>C7yzsJ~Q<5*LMid5S0BRdwYAYzyA7C=)oR&e+6e^bTO#oHg`?i+=%%_D z0I}XGK*-H7=SbcD!;Il`RWk&0zSBq*D!u!j3ZqA0p1z|(rK}3bUV8qj{NXPE>~_2T z{rx2s^E z^5e_D4#RNg&Ykak=R1!cJ+j3)8cQwu?U!%eJ-+yZw?EnKc0FUj%x7n3tJUhc>ql|) zwYOXJv;YyTfX(YEy)l)+R6u_IWCqP<;Sj<}x}9KR;?=625}N3$5W59&%F&HFQLVM} zqr_$qh}*5GV>2?9px{fX!>V{qrVnvr0~LZYfHgr>rxr6F>-OyY{OqE@3Y(Ao;M_Ro z5;cAvYC5s%196ea+-vz_-nU(S?qxF+3V@V_0(kqUKXoTH#b93;ti>sr5%a3oqWg&i zX*1M_oown8fn1bpC>jA-+|z9_@jaD?RE+`#2^KQv9?a3!-}rMc zeSH7)U;XIgaU8w+z$N&(E1S040%+sT)Tl;XMe$s+88LmH$@xM7{lYi+cP)SEmCQus zl|R4aZzrvNlPYk$o^YHoGM}Pdzz20u`wWz=Qt*?Fc_SNTZ9KpS( zspnEsJ0l3TOx}?}W?e)OXsz468y zue|cg&wlo^FZ=8l>c*WrcigMI_S$RyS%2H_zIF$|b2sh~8I%3?H*Wsm-G_lE&4}oC zUcFOFC20X_;p25I)xqgVch3g5j=>KZ{CF!LsBMBYq^3SPmMo63KsCFVRH|AbiYDzS zm?Mq}6iz{A!yPrzEW)?g@Yd2Hdz9znd%;9d0gCwuDKavfL<#-G8UcJ*Mmad@p1ZmQ!rZ7;Uxql}^e{^Reyw%d&`<=~zcNz~3I z0dK$m*}ES<=sNJVuYK+G^mH7@|Iu&1j7}FXFrRq5shF|)@cuVH+SLK4$4^X}7+M^xEJSQZlg~jWdg0Ek z%hxZDIeqE+?dNt{oeikQ1ZrYL^|O0V%!&bzFL>B&cH2osL}WXTyKx++iHx|Gs76lP zu^xD{DvYSOY$D^-c4M1FSE~x*X_DjPlgCevuU4cwsdj7M!+rz`d2M->6^wCEE zUVriS;iXH#xlwRnNUq>7{`TcO%CTS5Qn8oHW(j{RtC_RNB%Qazf~32t|6AT{{~x=Q zRYd65^>68ay1Mou9&;@hdJULs)wDfp;{}7Mkf<>+75ITnB27hBg?7&D7og;M7OH^a zCm$Vu_}Pg_yMA~8FTU_ZDOVpt?S#x~z24ufx4WI@Zr!@`yxWSxTnkT4RPCL2ej5GG zhV$JhqNY-6F|a4nO;Kv8r6PtZ6wkh!19bLU4`>cgErn|xM2$!lwqCE+s}+~Zj6#6n z^5OpeW(C!$wcT##wb0%R?RhrOU_=E76Gci9Ir_tUVgM?oP!oZl3XyVU25+KptIbYo za}km4I0gOcy>ea?Ot^mea68J+-uvLu<0r#rpX;iNIF~r6GbKt5&=^B&Fni@|w_bhm zHkkjH|K;BW;NpAb_SZ^{3Y4Dz#=;3$_ahN~GJwOw!w)}x~U2<#Fl3g8hH^gGtoRKJ^fTLZdrZhgiS=nS=bO)4qM`)5SdE$3I#8& zm=mUcae}I-$mY_uCwD(U5PvE{Y-DNzjN=3VrmI5ZIM&U&4mBrx1En%bskOE?-GB0A zJIT@B282m;D8&qFHnm2jOuF4QxqJT+$VBw=np_b}IeG^&LvbR|kKvY0vs$*-VtWM8PFCEq0ZaX#; zsjbhOYsVnsZGRXIz793#<& z;rR5d2Q&!O12ZwTPwziDJ>SljPviWl;0}q1>CW||Pan*M6@T!f_pcr8#b2`iFDLLU zj3^9-pFZF_cka0UIAcMCcRzZNb~GrQg-1b}Ya;RqIu7DpKJ_V|U_hR6oA(^a`ll>3 z=4S$sDnM~~M5-?geb`N2dFDpasdy1U+$6w&O|56}7O#7*i9qCL&F) z-@1M4&h1PffTL|PcInd(-=B8dNwgN4nljgfe(-UT>iKt70yUK;BGVLS5~>@50#Gh5tqX3;%4{Dlg@CCUV( zEu2eF0dx)1J~^EdQT15{Sj0{5efZh)*AKg&S*+cAfr!Y|i0Sm~EErRw7r*}XusAWE$cS+7{(Z!VxNxyP zDqTfZdy=;r zWRr+qT%37f&*9`{!rp zAAkH2wihw#8OgX>FCl`V*IxjDXO!oyduHeSBIp`GT&rch@aH!5^Se&BuU-1p@BiXI{+(|CpVQ~hU*k{xvS%;wWe?Rk&iL4uYQQu9 z(6a$gS&yeb@uPP?{mVDr@RGB#f4H~a-;Ebq9U*d)$;<-{W1tX-iOry31H|mjcOOe@ zyX{3AC&13O;iJdaJTD8teEtkWRY5ei$>L~Broy}JmP1zY_S|J+t)7LzIJr+C{WPiUuf~?v#*_UFK1fd=H&;!;IFiJ^PGTu_LXBG zp8YiddgYA|uHQMk@%#ln^DQ&W={cC02wLEuFfz}$faH6geU6VldjR0xm7_~vd->~~8*pCMD_iI}Rbi&pFqmx}M-mdNfepq? z8EjEuYN}iq+J<2`J3FbXqOFNEKVD}N1*o+))CxePc}h;i&$k1@gk+xqk%|zLpS+1^ zD2=sVuU7z6RYVF=i?g`rouEW(oqQpGr-X{|1wRja zM>u&j|NaZWf&q9k%_U6d4?cMIE%-96S^gU>Zu+T!Lr1~-g@gT zcNw5?7hdA(DP)ZU9m3c@6$W@8~tc!a7dqT@78Z4yx?Zi@^Q`VO?4qKTO3(v3SOczeJ?9Mg%UUN4BL7r3|GG zr4CtG@@u`?MQ}k6PcELmp{20I&zm9u?%cUEogXi;&%%O!4wrwyYhIeq3F@;hfPgF4 z@2&Sv%8KiHGP*cFJ3l#p^pkt^i--FkKE5=6)qj8hV19PDZ$EtZ<5Fw!psUoce)WZ` zRSs<0wCy3cLu!XKUZU|bjs8RR6G_$d?T^o=X)1*)^YLy=PrK%^wN>WmRJD}aDAG5> zYW?WZBM_viCzuNH{rmTvF!5?4H7lOdH)|?uu&55T647d?O*D^hnzNb2)I>$BRsu*= zm#5({7m5^l0N~;8mtK8Q1hu%q1S-E~_zMux_2(X5zJAXXT#G2wtgh|x?#d?*uYgue z{LbBjuU$L)`pq-_g4^xP(()E}x- znwg(*<`w(3)C$8pcSkB*j0>|*=*lk$fp``LIlH1t%%6Pv>E)xN&W3qwX_l>0VYFl2G7O~&j7#wXfmp5WIPL8ju@gbE zB!a?4_KnR;nVG6F&z+zCaufu_v@=&BigqCXtKV)0FGLFuK#RQuGu2vwR=gqxSglsW zuqrvC&@2bDJ|hOd>UHb(t#)zp%Qa}t!UqQcri&-Ysj9zjusZ`4&@VC9Pyfqj$Df^^ zPjy4LZewcv@Sc3~=-R)1>o)PqdnR`~efR#M?S19;!6PHvvriTTX#Z+^+K8mtdI}B_ zU?1Oogg8(j?(v83d~)aJwYW3G{?X<0<0ocX-1-nRnHXR;HLFuIRXp0-rNe{WZZ~eX zAi5YM0O(Tl3>V~~uuhGLKltE-%fIoPF!9xmfh$CN2m8D21u>h;V$5JPV(gmi&0#?G zFhF}4VBQ60$+ZKZ5^*p$l{yTEwNB$WZZ7~_jBO!vM$Fn;6LFX2Lz>hC==FncQVD@FJGU-^3RI*)6wwfB4FdKOzn6oE8j~-p@j!w7N z-@Es`8Xs(GsrBS^T&-3IhpSKSpKr$_W;#3FJ1U>OeBCU4?=+TDl8 z_a2?zx_;!hck`<+zxU%Ga2*U*)NP_i0;Zysf%hJbCp%(h6#?OV_r>F!o<>2!rGj9l zPo6wxZ)+jIyl50mbnDjbU%dC8u}@MVF)b7)C{itE>L`V|1Q6-MXE9UP%h()zbMf+E z_9E(GShZ>LbMLzJ@5Ywd25n8&t5sn(6DBl4hcJ}RL`^>Y@B=d|(L(J_VUY{lZUbRE zmUF`&OCMn&hk-D75KJ8-EoH($g`=5}iIgrbF2?B~N@BA3_Di<=i~nu==%Z3ZSGFI0e>=ZndGTkzEGRBFzfURqmxA$UzE5yLv{IeSSk$)Liw{1T&Zf3motv)D z&PGvu`;D8=-?{RQS8n{n|NiYCz4hrd$%Aq@UF^Mn>(Pxh-?J^yq3OH`9Wg%h`smXK zaU<5>KtK4|2Y>a(Z}_kGE?wR`I6QguC`JlVv_y#~bZ z(9>a}Uc$<|);6nwfbDjxa;74M`E0>f!7JvA6G1UGzcuMxJ`=K)A0YbhgAcFYyrrsz zS=|S~N-DsU$4~nD2Lzfc`=fC_wbnhr(qX6JBpS!<_WWE$O@+B=Yu<4(tk)@D&yII~ zO2oc|;*_Vr&&)kMC6Q14W%29J!RTkcm}aLm6B}O?3Owrqn%HW8J#ME`%Q&`GZMDdB za>B;Y?b&Yg^_QO8Z7=TK*FXG|56&(`6bxM5=o>fXjiG*}9S=`F`swqpynXRGR+(sd z=f3PGXBS9JYu2X;;_idvVOV*n2BYU+dHuZ~e6I-%6n!(~Cp3YFVg2@}=h9kXTCX<= zex6c)f6XXJm55BWa500UAg?gtNbeE|hnEib_Ya(hF}QLVGn*R03Ks|8o()(Ian;Ny z6M0!gKnqnZOkz+GQ7MJjn@uV8soOO8WkrRUN@1HTWtJOHL=crU!*Im4r27sI4kXlu zLe!ty`B^OT_HvlkGb=~r9VA1otJT0<_V@Sq_V%b3Ied90j28d@x%d1s=<_T1m0xy+ z&m*IF)@M2S#lOcViB3n}ZjH+dFi>eyh9cG{8OsUUXQ!vfkDr`fKH8j~ZpX2$i@tGv z`ughPt2kRd`Gk+Iyn3+t&%XKEYtO&%+U?s|+WyaZfRnSW%e`S90pv>X+5N{zw0m&n z+QGG3!+JHW>S|qwbuFt}R>QEj;-B0bCo!&dvs$lKt4AlN0DaloE;C}P-WvroefaR< zq8lYWDWM0ne(>Y+(70U_r!<*qwTeb;4Xe5!-3aQZ?RhIYV2Y+-D(ls1H4M9PTHftA zO&&ss_-t2(!PU%WE+L9%RrLJ)yi4*D?=JAAxdPiVK7c>h0#tY+>={69AV3WQ0`V{m zt6FQ}^?E(js;Vlgq9*by!rm|0j;CD2eD`1YZu^4YKCO3O^f6of`D@+X8GrNSgiEcp z)?p}QD6Bp+R?U10;NwRRKK=N^AN=60Cr=(Q@cMQ6`qlE%je~mehz|DUxBr5@^af4S z8=t-Z%@2Qa`{R#3XZ+Ds;iJzU#=~2@p>Oe@y!BqR?1-7j!Ii62D-D%~>SxMRVJckT z`tY<4tG$7Tq1=6Z+8@cD@_31YF1D+wnsrIuxqz;Q;oiLm2bT`v9c~myL|(sfn34vb3g8rgN1B!}-MyWvIijF1#9w z5?$y3(;>DZdL={r}`M0Qg< zKi^vPK;itNb3ZndPXq;YyVZR=K060Mph76@eIIB|YORH-4v{1Ot*`DMtC_YWSLoHR z-4?kXfFdEU`8O>X691POcDQF-x~*km*3EWDq6_)St3GS8;$P|My$>P)33v$oJ8hwT zz^9MokKbNh*?j%_YWI(H_3}4l(>mX|p!F~~9ln^1X#n;%d*^2- zM@L6~lT0y6|M2j@^9+88#Z|o+!F*gpAsW+8M&!-cwh!449tM3C{dr_ z;tBEi92=CVmg;$EWV}&gurse$t5SFfu9g1egY&=g-4{&7e6P&NM5bNHDREzLA6oHr zxdEky*i(pW_E~+`Y*vW*vIaQ6{DqgbSp(+Xx5eqt$f{?5BP|O(wds>bCwniva`(X} zUw_eRZTknjUV-EM7t~4@7wznP5;Zleq}S{A#%tgB*-zj4>O0@Z>%VnpSmDwY`sf(< zKOx!uqs>=EMW#4ub*X_ag5Q4W_T7h1kfz`nSMmL~-~X$B?viV$nH?M+ou5871!`p` z5ivuB*-&UTlqYA~`;R(P=1c%BgR0GH2*Nd|Kq0y)d4F$jsHG0YPtl8+sQ?$geCg8p z*;!>YQxPlV2YttxVm{1Mkf&7IeE-a)l)_|WC{`3AVzm@6kyw>l3U9Y(^9{oQ)B>M} zQJB4*!vB#=%;kt`4YNFZ0oM1F=Y#Q(TkFSjQw$q}d2y?k#P-2EfD8H%bNUrda1|n8 zurf`;X0;SQVN*4GjNGCVbE%~c`58dhhIi@jrfP z0Y0CCX$0K7c8Makblbh~M`oWrIG)r0OxJIJ<^1FrCIG3=)@M|Lu7*NfPS3Wp6``kV zOAH`eud1oWu7!CqfDZTfFJIbMX>GS%t=GLJ+$HGFbI-l^-g{M_>^*m%T8wfQMRFLnVaWb}?ZF6Svk-OQ|bQ zF*5t4q7H5dL|h7x2js-Hjo`!Z@@=uGE~!}*wshC|S3(}9W0uR3sKqbKdDu!GUg|Ld1;T*&(4hq_=~Z>-uG1=FP#;!QtWd^w{x>h_+F9#k9W~YAJUgJ?^U* zkJ*LoN3oJziRcy;YUgfQWs4bO)J_MbEm!s6eJbTv)7E=9=?P zW})iO#rX;9)zBRlU4|;TJw;OwDPLz^N{`7eZ;G&pCX8^ zu@?VXsD`CA&L867H8m4Iq6{(1J|D%8V7Hu$Izyy>DIiu#XO-!1P|1;4eSOH6H-_1& z+5X{P(N}Kr9=46HeRS{U-G@gfr$ed5&*`m0DTSwI1X|Pd-9P^te((qO(rdW+9PK8% z`zdy3#`0je`Mr-O;%QaKSc*Ahu{;GAK8oVo-+2CM@-X_!5~kx#-k zPE@g!3SufmI6FI?<-ZF$XTLSTOl(ACAkvOPXx0*P(K5s-Gyp+BT%2hY8K-8gDPi1= zrKX|*Fhlgt`zbXBuZfU8aH^YV%=&0P8LI!>%VwwtS$+&l{GH~r>e7JqX`P6)5IWLN z^knFD2fa#pX$WTT;8H^L(Y@`Td^mmQo7b)$Y+l`7yz>A%e*W?4xVO4kudFGv;?lty zz&Ht${_eML9*w88+hH8>_yO94ufF)`fcK+tpCvFqeemS72TyKYJ94!j)_Y}D$cSnfok}|GM*8^PL+a;kA(m|!3h%8}eUw`F zGOD(>+3c;>enOt^w$pCA-fUoVaif|PQcp#7decpw7b z5Eu>*56{oe%!h>`Vo0IHNYKmDs0zsJp_TEo^d|0Q;`zyYDlQ&nrceW!L06i(%SJ7`jy3UuP zWF1C%IyIvL4Hf&@XYH*|+Xs(7`SX9_)f?9~_m0ob&sSo1#C)emhQ$HaoiZ{G+=co5bX`sw$s7c+%zm3lZhPW(^B7xLgqs_VzZ! z#Sf4dC@4xQ`be+}e*U@Ve){uw9J}VI4TFJ+npzWWt;O}pO*Ao4Eq-DMn0cH=Rb!$y zP0UtEo3tj+lx)!807Xs7T5BS$))j{ZwD4{OA{3AhxK*{Y)6@OK!w!wGFur{e9htcp z&f#M&mBk=v)|e9JOaSMZd}gXBFcmR4H>$14xa;QDa`YEc;!V4`z|7@D5qcj*yRiBT zu>b&~`yYRJv}%W&p^#oYcu=mbwqx606(05$A(O$wx(I&y<7IrE5d_lsA%ETNnV;vd zUe$?L_s``2^N)Z2`pcL1D%OO0`MHOmPOGx5rKs^x`23;`m0y4H;QF0wII-Q{t6Lj> z{@$aHAD=%N`O>A$(Uq#Zd;15EzxCqRt=}a2t%O$tK4C6F~8J;U2=hZZ${mrWBf(Ziv0LVZ$zlhmpwcg*`bArNg z{K?}->-8FxXXAm)Z13Q}VNt(Be~1DaTN|gz$FoGYKw+nP0zzjJsuWU{*4nz3)lj$1 z&d#4|iXG~r$(SH2>_^sRDL*C7$U~XJlmWA3(^JVUJ3Ku6?6bS3hMM=CdNBhv0G`rz z^&=BA74`BBuW?W_0++&Vl8dn&0%ocr2%p0~f!MtTv03E)ER1b7n9qP7F}-^Z$}$}H zzkKxmJHz(=!MezJerl`F9-W6t4B5YONj z;XcjC{>57t*RJjll^;LZzWs~S!wp}$v{!a#SMGe}@!bcdwCk5v>vh?z%U+?w4Ie)` zz5Bo(pY0x9>>i)jOGkUp-B=%NCX&0S58vHR&9JVu<_TOLuTb>bBS)_RFg$nj>h0@S z?mj$z8vXz=GTJ`9|M=F`BO@@dcW|_S>B_~~v6Jofv&U!0=es13o_hJ$9&A?rbqLhz zS2hssAMEe#AFyLsBIYuVBl@vO@CHc0^x$Cs^yC!O{|^KrhU`R|GGSc^r5cx^RNwlT zhX<%yDGaO9#LU*KVOX!u$MbEI*%$-sT5GMuT-bReu%FW5)1o~I0AsLEm^V`A~S4V9`1QBSV~Mu*mm>Xyn{`vu>H-91e5!(`{jXsv0S1$OD_W)?pZS({8<4=O-%Q zXITN=yGGR8?X7ENoHpfF-(qO}yibfAYpe>wNAJBe-v4lKqjm47pOOmG&?1O=r8IFf*@LRe=wllt1|Sec@s6VDI|1gBNd_?(XUK!(zw#oA%ksDR}kW z*S_jwXgkb{^9DSxAHN7q{myG&{fGbhhiRm2zA?Z|fB4R4cYgmWf=;EQ8@Er7pO{yw z$0XDlhZCL(g%fS&TR69fL5KVMwKBxWj0f`ZyVE$X*XwTU9W~y$efyoWGf^kD8L+*9 z-L4lUPwVoq1{D#BqJ`irjfpnHurr$4REUP5NRv}ffpy6wP+_mIueJK=Z6%6;hzQ0Y zKg*AZh$0IVO$^#Ega{hLV?;DunCAe=`a~xSV%-SD%ylRnhc}dB?A*PDX8===)!lwE z4>caa|JmW80!9o!^E|OOfv8XXuY}h)yr)>ZM!Fr?|yXW+M}CS z&iAfMDf;ACx#(|v{aZ1#Tg^Cz=64VpKUt!F(Cr&n5}qYDzOZ2T9v$C*a^_t!01hr) zSsxr7A3q_rdc1A09${b+3Ws1FY}Q@Z3D8uIgM+<;{e1=*ld1T@zlAZLpX}{zsArdY ze@xnnaN}GHQ!xyN)xbk7rSPuFZX9c&p_J+%nk!UH8#763bttQ0Xd>GdhKce{5B4^t zFw9m%QL{=U>==rAw5MB`I@1D;yaREffO!$KkLtn=*YGE7!eDt%Q z-Z|RbdwlW2&0$12zL-e$)`9JD66D7w=QcPR{5#CvH_mnjfab4f)dau%jI-r-haFhUh5kQ z(D#4(!C(EoZ(1gHuiv_Je)`x`ViW^QHaGq^63>yc$X~yBNjCkFvvE_G#^Y1R+%>Qypq8 zTuP};o-Fhn050wCcL3H}^Dsw6O0B`6FV(pF&fX;-iC|iafUjhr)*^bo=Yu3wURjM1TXFmK*v#1x(N8MdSeG;A?kY`(I` z?e^=x`}J>p27`Gw2JKf8P5#@>&f z?9R`zDveAkkJ}v&6=L}R(e`i6vSdekAm*2~A|f;QeV_ZO8&Ig~0t%=C&{uMrWYc(Q z$v6~8W*Rd-=uPvW7nw|wncnm_^eP)0Gs#Rw8&TAdGfi3L+T;uVx%O|IoH;;B5Bh42(J_-b5SkW7!|ASM9Mf(FFl%Ee zrPTZv<|5wy>-}F+ATma-gFd{SJe5+4%-W`D8wFYo*Cvq8yoX(l*=r5V^-i5AFz*Gc zYaacu{`3%Gxms3Uc3Smg87+YA@9y-gD+6!thmAR?lRc`5iIHb*!y%+ph>?IAu6>T^ z&2EGgs0kgwfQUpKLY&+*B6?|%hm@|=XsTl@0w`!m<7{3Y^HezUK)83~?q{E#o|+xM z`KxCDzxt2=PyOk`CqMbom%j1JM}PQVBG_IdS9y89=1)#mC(Eo-o0GTV@KFP%QX;g8 zfNjba!1NhytTZvRb$-&dWFQ96G?vxUKt)*+fBNK3U{IhR{NTZtb^@9iatt=4u9UbQ zx=$W!T`ypk?nV66E=CxQ~3V?Qj zcH1y(BNHHDWOh;Sen>@hJ!B$mx|RvWY`N~Uh?x#U62rG%yFX+d(!eIfw3yFB3$^j$e7p{tVQ_#8dT!rQlQ{l?$xAWRNPt5N9 z!~Y}qXFvRB|JP#n-ji~+F0bCdHA>EmCJF$+Cr^zFT{p;kyCvAB^Quk%YPt@QQl_(u zZaM6n4M+3c`QH2h1=nX6B@YMt?J$%l8deetHn(o_)iOMKvQ%q`G;2u91|gGO9`5tq z!-aWum;rs~qksM0gLm(}bokOs@o(MRS*?y9Ji)V4y_|3<*U|SYNq&b9J=;*~2F>gscWI000q(wwcm3 z-o;dJa3;U;bKlH)l51yXH=Lv@9SEGy5R;OKmS@t~xw^d=&y6>~#Qeb~#*J4hU>OUC4p(?$NyNE`vWi zm-{cjqD_<^{%Ce|bmJfYr{DQ6{_i{g^}l-gYhQ`4zWm{L-?R6=|I+Wii^O2??whYX zJUQQlPIa6Lnkf)HdUo;jVtxB~&#xym-D>YyFtZS%nIaNEMg$_>UCi4MhLl{j8%>>6 zbWjl@6JhEF4FgdVD2s{;BAY_a89g5niU-5QID0Ku^~i4 z8pZj2HW+6qV1|G(FuFQx{W%0;Q$r>!g~0PLC`Gen@RoltGvje693g%0y$?^%&JK1L zaWVHNGn>s;tDd>3hP0Np$z)NTC4g}Qk6_#*(K@EnYO1yxxr1A;;Aq?c!3=NTzIo@) zP5k;hVDR7kH^291)}CBk`Rc!a=LR5Gs5kxRhH)xabVET1bt0i(5RT?RD+dLZ6J}_e zMcixpo!O7u#){zORebn(=i$X(;9^z~fDw^!XMxP~g5~nEzgT7iyLDr6d@wsbUtccM z!LH@?`lZ|Z-jiWy2L}h#r=!39JKcA_|M&+#=>Exn`7-kpF!_UjyB>z!5PlHxXs@09 z)|V0C*75G`KhY&*O*Tb+}^~-gSfW+wKP=r_~ z|Ckk#$TaU-MgX*aDDzp<_k*YsHy0O|5SqSES!8cM1HvY>3LpTFPtIeE2qw8!hOXCr zkus#*r_6+mm_$^;R7=s6v#9RHn2R8xrL3wPLJ<)aqM8RJFIHn3QW?jNk(r5zC?Hx9 zWo9!a4rv&~fRs}zC#UD23S^>YqON%7OQj7IV-o@ck{E~r0z7^CboSMk0RfRrtj6Ti zuyGrQT)H=?)ZfTV*4mhH(klP}9NfBJs{(M78{W$87!qD~cb-1@L0~qs-+1?}v69u_ zJhlqhxEaSFy;4p<6REdZTMn84&Q?zsVHtupEyuv7paq6?e0J8wKnCC)e`?xkmP?Jh zN7l}c8D2f6@BDcA$){%=crkBwW)?xd{wsH3WDCFbCqLk4+JE|>`|e-v?d{$A;3u}b zGxUS@>kF`BAWHcjhP{uT9r5K=D5e39PWh`n)L&}s( z*Fnk!5!QVQfk}Xw{O$(vIW{UPxqtyQOaPX0aS5qZgpE@vt3LVQ&RG`$U_S2-4-Zn& z2PfwW7@2Zj#~2o~E)_{Ri)bz)2$4y|cdm)4s1>iBFaQxmDy1OvkV~K{TIw6X0D@9h zDMg78#1O4G*hlad3Ns_~c{6H=S#rw7?^RJS6$BJj5B>{#LYPAg2M7CE;3tnxXY+ZD zxxt7Tfpbn;#73gQ)Hajx7xXJB@4>y#XPq1oB_;}Bpz4^0=-plQ%j=;4$3s*iiSp&w*t z-mLpP6nS-#F4z5%DD2!kJ-d4M-J>_(y8G|`^wS3qFHcX)q8*sBsd)Dv06h5gSsvh5 zE|)v2e)j0e0#MJ+pMZS?x(2uf@G{`*0{!m($=F)4sV3R%QKj*Q0KlLB=%c@Xa~}~H z832e~NRUWf^%k9Y)%DQB1R8w6mKg%bIDsq%GF9+Fh=?45svrPUaI5J;Xic5!Jlz6>u>oRuxfi$fsc^^t4E%04Vt-70hpSKnT);6#Z*qOP7ZFg2gkFf>!c_%a10Ml zc2CZCm?>p7gZVs)m}Kc>eRnRfT18#+Z0C>v;z?^I z6_vB|^&!A83{Svh-kt>{fY*WkW;6Td?%u=w#lt%{e*D(!mw)5yINF0d$NRUBD;7n+d2HIyboGBkSc(LUBElg!PlCi3mshurk4%LSWZV|cl!+QEgi#l&h;#@c z#54@Vftf{$ic(-iY?`p@2dKhW2-h^wy*jDr5Ca85@D_nBY*M!zQ;2cI7KQ-<;uPkBLB%{{j4*DI8XHU-VDk7b zXzDKagZDpp_Ust|{MN61?YF-9_I5U7i^}x14seY(sCkBwFm*RRm4^TrwzDP#pMGr| zI14_0);xSNR|svOltj&jB*ReH#H8Fmu$A;bdUSgG<-0%mxcT5gdh>Yw_TBi}(eWRA z@9gMkpV_Sped~{Z$f~S>UAvlf5B7G~PfzDfxM%`UP+~mX`}WS_q1iVspWI#?PLJT7 zH(q}D?%X`|rxo0AR#7K2LMp0+xLmFA2B`^o)ecs5OGKRX1S3;1 zL&PG=gv1P^c#fzkVd^vi^tdy|?rM2Od$hl^5GnhM z?r3iwn)&(FPf`&=6c3$MOA#6?L=ed*`QUiHTOL3|!%(V-r#l6$m&kc<=r9pPf8gVE^l1`;wM4R{lU`^F=ll;Z8SAmona; zQD4VDPz{$o-tVye`t9F*wEFbX`eDxS=u!CO)7jN}wzFVBDy7`MwHStyi#AlUVl2aJ zHy39QhxO&<%P$}Q=<&(^EZoA?M;|^wDTn*Jk~5e?bcr$At1qR;kB30blc%y=op#Os zxn`AR46D5UWZA89KKk`v&uKHj{KjrKK6@OUl ztZLeI-F(&}!jSXn`2_-$Qq0V!T9i_!0ZdN7ht*Cypn4vWT8ZEqjDE?V+X3U^@ecay zQYcgY6|MpDNx29%mbo6zR1s`Gn{f=Hx#Tp(ktSjyCdEY1fa=X+q=2<7564BSFqS9P z+b5nTQ&4^Q;NgG%pZ^#C?4SKt-~RLOeDLF+Oy~EC3IIDfd1e4#{py#A897V?X}tEx z2`(JjGBY))&0jSm0bUzuRY`yl87Yj97MY0e?!5But=~8d_ul_-adNQ}g3H2{!^7Fm zqV;Rd0=qjcdSWJfN8+z(uiG0bglo_3;nTARk5Bx~V^dX) zXf$v{1Xbg}7nfIiI|~uTx`s?d>ri+zRSVV2SB6=j8dl|v8j|2cNS)4oyxNd?}EovGI0ojMUtO_nM7rriDX_Xu48wB z#{kGUEpE?#sQ377gKPg_Bix^szgyY5B{uWA!3_uS^U`iu$sJ26wOt3G{gA7^g0D`= zTS0|6<768H?hnv#tUsL&JR%?e3o|Yd;o0fw>FJpP{NW$|o0ngH>G=5gYhU}Cd)$Be zPyZPJ{MOgLXceW6-VD5q4@ady8+`^&$zZ_I?>%>QXgE5l$%cIC&MW`>|E2xGchA4| z{%XjUvmESq2Ego4Q-`D?J(aSz(|&Pz_4MT8=*H2fkDB*R(x=a|0={;8;hyHt{^$pR z2(GTe{k!qAM@?2XF@O>|0ibA`O34KS9p65}sm&BCH~h{UFa5Xw{)fJ?VM78`s7dC>+d+ndTz7lv>-22z6AT(YOJOFb2w^6AJPyJrMzzh=dpk zv1#I}PY7sg$Ri4>Ph$*Q%@BZzQYupz5Mv{v21a1aVd(p_)ykiH%B5`$(dWSwFTJ-> zTV*VCm|XdGDGQn@BD>C9x!zPWl?)MXp85psM@7l$k=IqBK0)KUl+tuS&mmJ2N|B58 z+Mfg=R?!H6+AF{eRU25xMmfhkggntM2!LaATJ2u}qY;(Y=l$`=pM3oBC*S_|p926h z`($jYHBkmOjk`lcMBXqb^_XA?JW8I7>4xdnT)RB*B&=VC=Zya5SNHzr+q-}87nd>e zr&nP}#ot?g5)-~ng5CxH>2mowBpsGSX`lm4raKX^gn)(IWwFg_t zk9tE~&U$3m{Yc{~oMc4|Y7jh`$ZNI|k(@m-v&ez#fMHZskz7hCrAUz}8HQs$be&^j z3Z`aiDxeA?AYL^krlMN3Wbl}0T$0rmj@R_);iK`B0HG0hqlLMgmzlTE3#fpqNiiv= zs#=_mRLNGvL{-F@8Pj4?v}8YZjYZJ+j#VNH>~DN^@mp_pU%mrhz0W#!vFD;1J6K(D zH`|}{7n;kbPcLVo%$foKEsAL|EvE9FKYPDzVpy+dmshha3xLzEdECWw#vw*U6sFY1 zul(v8E;s#?DG`n~a%juqvbCu4x{+2a$D5>k;;Bu1{GFdzUBQW2%V7gv|Z zM+YiGpYmJo}ms0I#b#^la~ zMGHcmaoIIZ8(ES5>cdZ_i^xSBE!H@gssbVsX>l?1as4|HgeamuqQXYl^SlD9uN_0L z&A>WZ=302`pMQIut7?7IvzTTK6@nd3_Jk%9DpKDH$}!g3{7MvWTN?OwiVd|Vb#G^we2{VMvi~l@f#1IGED{c z1?NQ2#x-pq(AXWiG4|WLq<*tfvD=dYuD@4BaxPAw#1NVgVGEMhMK}pV0Bk7#I?tNC zpcU%=_Ch2tboC!n5A7C5yR{d#KFLcrcPrbw;r?r4vm-fwWu7lD*RneO+uwZ2 zwepep@#72H*@3-X+}rJjw0p5Ev(66p$}9Ks{d=psw^mIkB1gxIIx*XCk%y{Om;T)w zdp=Hcik3zaq27~6C+81O&I1CbQgG{*nQ90g3M(;RUS7>_9-9ggh#{DX>d5n$b3v(O zWX`3uO%p;abqp{&K~6|*6VGi$zG6N{h9aVvPGgjYoVzxAjUB_Y>`!;O9kf)B=P8W=r*Gqck~kBv~(I7%bb%IdAbi=?*28ao67%1D{R>C<#iOeq{rH?2 z4C4O5>i{48_|e~e=iWPSy<7u9fKMKt;%bdAy(9o-NNc2cdQKOYHl!c|h-Kcc4tBnH zIIr`8YEU$Uo0z|9@ zo5c5Ddimb{m+s!XfBNjCna$U$)#cfmj&uf6#jU;hF?-CSnSwB2m^fS#V% zjl<>FUcd3+1TL01WKtavAhH#ifBNL@|Mahoi=p;a*UvvTSl_=rw&vmnV{SbgdvtOh z{Pr%_!~V{k5Q|B`=+ft+VyZ-VetEUGvoKJ%4+n?GZ@vBY+1c5}a&_D6_{NR(dX0dw zoqg%;uiv=y_*;MSM*zl5C7X!GCSt+1iOY2lV@lYz$xy^1BHqZj2?MBUciI4Evu-}?`t^!&Ga}klwPb+N7b0y#~Mw2y~;^L z77+o63=|X*6v;*X5Jqqe255G)&$buoD{s8DTV3o2QBX_V(kDP)TxY9lH_#H$78%@So zTn+ob^7_{1t%=58{na-wzVlc1B}5{SoEWi)4$I4ni}O3TZ$Dfw7yE~Ozv_oH?;0WwOnFEm(neB+Fosl! z5JHe5U=}FGNV6t1k(~uv51G2fvuDrtW})5NdH>Ug^&^jH0|5^Y_5smMZyg_&q375X z5fy17cY)Qk6jTru6~f@M95Gr=oUs}d)j-rXm9%97Mr=Y*&@Ki*@bZd4m_<~<3kRLR z3QP(Ni1T^3ySqcE#XCDq+a@+KFaoQr1x`+%?ax9)%4rzUXodR3M4=&OpfE0e1rM+$ z^p6gCc;nA&HM6;?swQFl&SaNsux|E+mv32hQ#!$v$pcsVoCv)xdgOq{g}F6+1rX2} z2{9Ph7!y=rwE;t_wPte5gKU5P35nqNX%Q!40f_1E{%8N_yZ^U;0iQlvfA#IF#q2z` zu?&lkK8ee+%B3fG=XbttV-SK9>Em7eTkHMqt;4%F4j!IdJO_VZGzVcy7g;UvAy1>$ z;h}(wt~)x|3o$6t-IrcP4sZO*o9nAfL%dvF9_$~)ruoBv{jVd38#izK@gM#1-oc@l zPlG9P2uxy{Q_8t8Qqu$xMF1vhV?Yi`lbRqvWNu<;LO?WTUJiQt+1xjlIS0?%WhY;L?{Q3J6UMK2LxU10lzz)^Z{?F*YG| zq2WMDv#4BLUOsvJxCy8NPfniYGJ*Me9U+7!a4%(fwTyxDkcMGsV&mE*fuTsX5$KaL zT((iww?HK+%*ZMBO7aR}vly$!eRk6K zYboX8;%c=X00Q_d%CUOF5hslCE|^t`AOp}*zXH8zd|dGc0BE2~3ELD$-4}j2W>24- z0KjW^kLp(Pw>ahML`{u@5zq|Lm!zTMhP-ib@28|qG5cuVM*`Qj{s8bqM%`a^4y;y^ zvYySRVz22#)#c}JQB%D2cYgJz*%!<%pDoL)i!|JWl5gBO92o;J+Zs-Y2#NqzV7YFn zYPM`z{W#J_jyHn=+&bK|@!bd(0(*@o69o?YySpmNfz@<*b@?lAzJY*D7GlFl`+Iu? zaQoKH+1}BEhYwFre|+}rDHxtzt@ig8h8CEz$hsehm}3A@H3@-ZY=)FwJTMZwhJ%{q zBxcYLLmJkBad&5D$a=Lt|Lo~0R93S-?OR7jfM_OnZrzy8W+zV`pPrp{-3*yALuBj0 zmoJJ~&F3QJqF~z1x)7S8+KC?Q&Q+v{M%55FFvl1#o;|S%H;)`bT_;d$`mx$CU?eHR zOcAkw0T=}`!88mPmzSVr)^$kX;_7M`dWbs!peiN>z5d7Q&|0XCfab!9W&m2YY#ads za{wj;^x9e%Z2SDnHVDS8q)a{RBTpkER*}kZ*2w@!9Ds3D(ioc`ahw-hBan$o)*1|} zZo&*-`z}9L`7D6q5f>1LI)?!!nuw}__VUdUEfUTGuw$^WBIDn#5!imZ6qN) zo576eZKHVmc=yin!GkBK8`I&7NDbT?-8k4$QIFD@5Xk%VK-K2E?P4(_4rr!DPM$ov zdFO7on3uG6rUjW0Z0P%sKK*`9>EitCum0-27&&rJx z5CR4SV}uZinT9@_*~R6h6xrR`iHsE3Jp1NmZ(NOSU{PbD#bSpUXY={x@~UYX8C|ZI zGXj#Tm|Cq`Q6ZwQfAiN44vs{`BWxX;(wUHnfAm*BynOc5We(AG0YG^Pj(CVvCp{K~8U{P&-2 z?#sv!06@5XxW_~Ym_29zA-Til_<^#~4KP z$=Su>!LFzTjw03%gHP3-w{g)0CM^nSNU>>Dg<~)iz`%h4A&cES+%eNKtcM}twR&I! zz&F3}5{CeYfoZi`Ap#T4x;B@hW`QXe5kw4(-gRj;SzkmO#NFLJ5z&Rkz+KzqoJAyZ z2$6G<#e8n2f!SrB&6J1~fe?e6eN~4oW@CSI$+7F)W%!J4Gn;iCG0od9HqF!ZQbn&) zjtwHB=A=Wv#Uc_n9k^jaG$SA~S4J7z%{+XhD(OrkCZi2Kx%H5%JtPExzxB=6Yv!vi zRIJPZ(P&3UPf)89ag2ECa`$41P2JMiWC8}0O}23ek9iQIWN0>?qV5fJwN|Q^Q$ejG z7Y3?TuTT?z=3}15leoFpYj?@id={LsjTFUraH!cs06@HR^YGD=v+a%R?+7=K4opF6 zL9YRkDX1wT?k?tSj7$Wogc!gKQA=8{SC1a{s&;;Uo^x?L$Q()3JL^FR5t%~C8Hhx! zloCQ9(GZzJbJ=e+8!~~E+{Jld78Ni{x$G|HJB!8D)ddo&$SQ4(gCD7cBEo=3s48vK zFkuKG7cm7R^dVhDBw}E|L_VdIhymb-Km4Jpsg~Jn*2ZX}3bt5u1Xham>s1JpOAaBR zSs)hBTCKEY`J4tE%h~`y5Q$BR8BB>$RGS!@7`iSRn$|*b0087bfydfSGXN=rR*FW! zpkiDbi>I9a5mZ7aE1kF0!T`-Rlf2~to;-f?+wZ(8c^K)M5iY~EQ^15%5CKsqJ`rn4 z6x3B&L*aU&iD@#Zfdqg}L5odT%V09yBbhKJt?HJjM%gcNOgs1o8vTH6;-k%`ola5F z)0m$m6F&ujRjTakZ`}XAr)L|NGH&7kYDgHoDU3)BrWv{idwWe2I1mElQU}B^7Ab`> z5>v{#sCYdnF-h&9MKxoN!>|@Lj-ixnq9F#YxWba;lnNsT4l33*E=NLFSAAdtz@y{i z*<$|m=@SDMk&ERTu9-Htd3@;iUPVl`2?2=FTn1>2T8YtAMMR4mIgvt$ufOqT&c$)v z&<`M`h%_-=U1V3a`qOWH>v(?wfRTA9N<>wU8HvfE4uYAOfFT43E(_`)LOq#i0YOSx z%;)WFjzngps#E_ z>3{Y=dwaClW=K}~6ROekXwIu2nbh~E|07M+E9lAO8v2Z+iSGqZsF?vqRU0>$jDB*% zq>UVCHT~5>6K;KYtXX;|F1)zZCT7dbXtQt4U9po<)k>Jx@{RcucguHg9^N_Ld*Hfp zb+;fQ24X@4Q$S`j8?uOM46$v(?VCq!(}WnL2oa4zWI@Rgf{7qN!%RfJ!~rpeFu=fw z3Q9!JF4i}W_6^Ahhn!5UiJ^!{63Llk1VjXFm`xQ4hat7QJ7OA}hB>bKfr&4sK^Wub zHh`mp{j>9Pky1+0B7rG{1pumOV9F``!}JQK@lK+z-T8cR^VUtLS=uHvkxd~49Mk&> zfTmX4jJ&@A9R+|By9TCN&48JhsB60zV+cV-IFN}Ek(N@#LeqGH36P2?OdbG?$yoI7@B(G%11-^5TEfA);ClOU4w;_Vv;58MgV4T%620<)PKqk?HsB#JRG(edH_ ztck=VQUW5zpx*jK)&x|7z=RlxXzfz{NQ9K7_`Dqs0SQEP7;@8uCU8zkq%eo|dW}eC ziePHmFzzg7#B3^NmUGI~c0lWOUy8I&zuz3KDgrMHBq(V%=h&a@o z2BL~c(XMT{2ofn;KwI^BhZ{r{wMbOFr^z8N5)f(0W@2C_GR4H>gHr%RU>cuhrA@%< z907v`4)ZWT004NJ7Xa`y&nFsDO-;(^5o{_!ANj=!w~zzYQ(%3o+w~2`_ka3Ez;PMY zwAvV5SFMdrqzML4(*eMMKJ2vO5*UGfJ<$~1RxzUrn{wcJZ?Rv=F z&OF3$`s^8@mPh$;f6IcW?WLJ(0AWiT-FIr1gvjD%<+3V<<^fiW5ryO7S1G6FPB z;}4vOLTuY^K2qzLOX`>BRs&cXi(5uRI!1fbvaA5wMuZF3uuli61vt26l9Q^8sAID- zU1@qNg*=9owO(ihKE8hHX|~PK79D97lZ*q4pt1wCRjnfbiOtta7EK1+k2B7F+P;v@ zCNcm}6$G-e!gah??(oN0WV~0Keyh5@>V|*kOZOf;em31(1Pr2JNd%Nb^ofD0azbvcEWjavC?YkHF8ZMqA;yx2ekcH7YAFp( zY*0zn0(!Lx8OG)6Y|(_Sje@3P+y#}wF?ECwjz zaEJ*Cm`M|YSL!>k9dc1X0tLplVP74AxoP5%2ET&!daY_4+tH2zx$b(k$_7XioUd77 zt$5kS2Rf}3+)yQ8CV*3zQ87(reWpbV1_B=U8jeh?R+`JkMLg|+EqRQ}JJ^U)a2m5Z z9%chp#9%BXsw}8gfun0dgKu~kBi^xu+s4Tt(}nnt#u?t5Pia6f76{ZU`P_v}*zwl> zVx+U&J5dmoV#ds3C@P{7LO43y1Ar!mJfxH|G-SE7fkN9w#=Wj%rVlTL2-_HlxQG&y z6^t>ayzbhzPeUovwjD}A0A@mirfG8OIk2j#WQ?I{8fMN#7--hDO%pFtf}&T;rIjQ= zk58|zuYLo>mchj|5BB!wJM)ywa(QXy@+}YcIJ ziCP+pYC*G73J|VVtMl`76{+EAHC!5-4z5{QPspf>x9-<9JFViXtA$f1S1rz{i-w3e zAn}x%0)VNbY_y0Yv@jJH)}Ay&>&^B|mTx$m{vL@-U<$x-gheAx*$72Ufv(lm9+(z` zYo`FSjUs}LMjglOG!XiQY~uL3Gss}ffm^*@XsS}s!1UUM_~!EqxqYzp36WygmSGKuDGfgNZt`&eaJaijS!2_{FzoH` z0+IpdA^9jqQ|lUD4_-=%-h#+YMC867m_y(9W;&bAODQpM6DY(m8U%<$hF}21G^9iv zvk0;4KQVxjG1ejr7bh)&gR90O1Y*`9VL>n{gd&JY90LmgfEFRZlarGeBY?S@1JuCJ z%w$B<1T`%>hI&;}+m}wA48SptYYfJ2m1=`Wvk#_aT@DvdRwGWWE6az`fIi>3u{(HR%Ag!~<$gAu^8%wJIYqp09eC(LDPH8~vast?S)z zcT%er0P5|s5f;_?%zhSUL~NXM3xMDF@~gl1@4jyyW*7toBS3Es9EQw+5BB$B49x6i zCWfez1JR<3v!>Zw%(LpllM5_Lfhk5XbDh+{Bt>DIiHk@f@GS6Ztu$8tHBA#Dr!){k z4D2{05*Cq^(ja6=rO5f&scWyVw^LvM0MNz|n`X5hI5b`p+h4ARH0;b5fk`C;peU%A znv_xyk%`BwiD?mv)UVdDjb73dD5#mbi6uhB*?dlA1u*Hgh={pRin9&oZN5coy2v}q z0SGY$FR-=96qu1Hi{|y(Ok>*uSk=8#SC@7?fF_H=+@6%6}98PK7(Q-@GnU-vc^g@LD}RUI$J`^hIpCG16vZ zTmBSnkGl)2YjVqulO|vdk!s1CC1>IC|PCRb$4zY#=r~!svM9wfP(XbT^o_?*|V#Ic>|!QSwj<7>p{ed))KLH z)}|pLk!hJnO1bzbiVy;tHi1!9i-f?g2N3`W6oFDo$PE&=Fq zOKIC?Hk+5r6^4*h!Cs9U2TAmaY+ z9^w;!W+Fu-*VX1O8>hk>_3keliOX6IQRhbD6eb#wmk6SPr&K3`e5BdI3`~V6AlKar z==0JP2t746;uixLg`aA0UOA_!dVE^Ij?u9;vDLLs%t%N;<{kP5hGTQHcY}}R7*NyC zV`dZpjH+sm)TKJc);Q)NoSiTv9xE!mJJgKPP|V1SCa3o5$!TZ>{ndZoI^2Eu^xVG@ z0I2VC(?-BRWQY5EUDr0uF`(qqU@!nP3J5UPkc}&M)WlIRHvZT?1KE)KrQmMgnmY;&OS}H0_XjW@P5R?{{_^ zm*T860ugKq)Qu=yT=g6|#8`^ul%cF0Kh!ruyO6z%kU}T$OvHO>Qq|S;+Qh4K>{Ldt_7aI7Mq+x zjqfY7StAfcRHT3gp37I!?G-=%0b^o&^cr3hVTOQrj`zO%UVYpF037h7TsVXqM+cF) zX#<*N&=^80Y9>suPUW&6Bd+MMxQ*GRl$Uf z&1@)=ix5E9HUJt!m^FdD?2O5XX7hPkFBgl2YFYN1QIJy@aD1@4?h^-2{YnywXiDkE zjpI}jfD%&5t^k68NdTBwt)yX?&t@vp#yJ=a%gfV~lf#3ZSzwMarv!+B$%jC#E|;}} z(vTtpIzF*LLDf_W6GJUDtN@6aiBP04kpg@7;{Ra z*f23+ok+J8TiEEy)SO!(0B@+8%A}`jfepJeQbty7{N%fkY3f+ozLw9O#egaNl(S5i zWA@y&+`Dyf=Xm$w$prw0FTeQ(Lh1qnXxH%G&aRpE_B3oTv>a1hHVKo52 z4?j5((W3D9<#t(k-(ZT*sh*1nLFHYxO0|2wO9g1|b#t@d5>qQ${ zrX0hgXJ^i1JSce>im2@@7EjO5Vr;rLN|6{s-wz>##Vi1tsqW1<$k4?IT9)gzA_Znq z8?qW`6R2%tQAI7LVx@=*hgfoU4}*lS+`IGgy*rEsXennfOGD0P4A?{hfW<7{B~-+R z19Kqpx&Q#Edgy}zl(Ys!FdBPSbiyDuir#EE?JW-g8y=JZ(bTgCRD%jo*dqToyQoqr zH4j>C_@r4jaV$=~k8BG7dao5oRx_+oLG5xJm)wS`m}baNOB9`lBh>#N$M4Le!<{qE zD0fFf#+^;Z;ZCx5$fmQCl`_wWvqxl>y$J~+D+jyysNWew%X_~hX>D^pO$Rf-@ z6X0qsz?oHfK=ZSD(^%6mfi#jE*Eiq-H?DD^K&2HD_-!q4#sGx=X24Tx0p+~WC+u-{ zX6#TkeqzB`FW+@d^_#`QuA&=Q8hu&<$ro z@$raa@_4^Sr=flZz$$(w&+TK)`uJIl`72{{=+*nrK{c?+@^@CZ_-yf71Zj)k(EkEK zb#iSw3y$YQ(u6>Z%qbebM%ZpJWhEamA|w;_ds4=|{r9KmHP^V2B?$#u?Okm$Obo&Z zK}0#*BMT{b@2Pw57u@UdFI9dwqWs+D8NK`(sUonJ+A)6ZJHQFIH76Zj5>^MsiyUj! z#YN>dxN4R#0C7f~-Y>@%PFT8ZTP&n>HpRU?rmA07=_#EYc>A63Tzr`&6H?-{Ar8zQ zIRt;NHoCb!EVmC-uO%Dkg*Lb)t|j>MJOfDkG8I~%_O(~NIx<_V+LZsXEkCQKd5Zl) zj)(+mO|kD;kt>FTHOWp6xeRe@vff{<%zhIBgUVJdw?alj5IC1iU3!#v%0S4GQz zJ<)lJulzDLhNYs`dCvPc#jZCh-IwhxeM7^57FG9?a9DL5HLF(;yz#lkVc-G1Dd@p) zcd7LLq+fhNcmG}w!KJ)l|DfNR&#J+SfCYCnlZXQa`Bry|^QLV_&Jxcap-#(Q*LrzX z{w|$c$hpnp`j_4fkPuHY+5Z%ZepqAo)geSidZ41i^RIjkLz#)At((ke-G^dP604Af zer-Lm*=={Wf%Q8Lxqj`|T%TR13v9yX$1b~F7t5+M?hm6VY^|mqlPWnWp}fc(ddve& zlRCUX$U`5V(IVD{f^~ zL+|?X#~Nc{xQ2x|GX@s74^jt?N=lDJV&10Vw8b6izxFnj6J0bpMlvZ87&XzG=p4JA zVx^k*ro{kY6g--&4-+|B`J1xKtIt%tMTH1(P^=3VFDx5r2>2^cN{tF?`Q1RuzPRvz z6uiw=SRwO4OSj``DeI5*NoZ%HUQ%XCk(+}Fashhtet*Ao@^Xo?TZhfGdn&$q@(mYZ zkz!Oo2jTW_e1ycIR=(7qplJAz)gy8n?ds9Z|3+%MBGMyXRJ~0p`{%ES(+f!qmX8t5 zJaohO&ht*X(67(zN;L1b!umZVsjp0|ZNEM(vdR4$@0uTl8{od7prjJ<=KHiOoh$n= z@}Pa~Lz@Jw?fr)j#*`LX&w{f5gXGK&&$~8?qR4T{z?jWgm~{=9JU6Cw6o&$SMi;no z0nBP)#D9X(T^CagA&LSB19Jv7am!b=r%L|&WN({j^oNFqCJ=WyAtD`2gQIC)I>1(ws+=-^!qc?{6B?<9`_?f}QSX5LfjS zAyNivWA>)tG11yLQTAX#R>3|&cZ|vNWHK|#yG0j25}rkRZDqM;Y#FGF)1gjoZSs1a z{_<(dkFL3+3O&IXeSU+B1slbkTC%@j`qTIfIw~L5%8`)!ajsPy$bp69J%FE7=CMZq z{3DRLb+mj`)QFVA%5PqE;W~J~mHOL1FeZ6@+q4{g>QHEZf%t=t^)H>SzuMg%z++z4 z00@?uQ+?0$iW3(PrlbdXng)_mP zqOx&ZqaA%1#dmquW~M&}3cZOiAqnB7#9sB50qFC&6_<=YX0KE5`$Srj!K zV)o_?4!1Jv_!%_wX11L(a;`cx==2)~Rgxw;dvEmZm#KzpKOWaxHmkRoX;!4A;p4ER zuSYH#bbuMPR6|2LP;zYc$pFP@OB(9_u-|N>+7JnD7MEt9?A;+rH{x)w2x<==o9>x* zMW-vZ1seSMk*oR3yu9Rdsv)z4kiDy3-m4m?SIho2Ut}kGf5Mnh2H@UE!QqoNadv8f z%gC&4G38_To3PwD;(-^^Q)7hwlNX=dr^@J4UVp=>q2;ZY_4eKR&3s5(Z;Ejd#yMQi z%8$8RZ~2w|s!)sMBHD7E53v-}MV{Eb*?~A@6jimV9Y8cp<@W z!mrX&K7J986Y+nj}Y46=kN{%&BYGz(IcOsR-T48QMR>?=FKh>nX2B{8Jh`Lpo+S1mNC^XJ)v(BK_~ z&hr=>-T;-}3AfPl(R-Q`g(Nov8x* zpps$;MS%*uBsEv%$)4F)*je?X>=zFk6a1SrKXaZa|318yz4C|BFR%3JE?y7_Mki4dxQVM{-Rh)A^eO8A2pE3?>5L_VK(@Kx=>V zSAc?0nrQiuU%pnz7Xu5SWi5MAM0ugt%1ogJ7J^3rDlX8w;0BpaqhOHnThT~qgA!vF;|-u^G*Hs?A#jesZRSS-!28lu)_vsa z<&<_W%0K-jLka9}&2W?EcGq^Ki0{AK>`2%A^4Rw&B@sV6_NB?~0}M_o|5fO5eU*BA zv+Tp(Y10Y5ApRn^!7@&_%;!S&{l*EeVrVb>>m7XF+83qMtR=$1-3MpeWv#Hs`&TD< zdh5$BjY7XQNI!JD@`of5t)c(N4J{cYP{pD#a`gYX73^iYl|LbUGT)okhrlQ@D?LT7 zuh26+&uSsNF1Qv|RZ!V+?jHdFe$Iqbf|ua#BZs5x5&AbSN zn9{;EkAJ1|d(ZRi&bVfC@S%j`J4(^)tulIlWuraPg(^7YBjFk*VFDSdkx2Eb0(4Gy zWVk&j%!P&asVH-~9kE>3VU?Tu#?E`1zbDybeXtf+qrRL=82i1jTP4| zHr=;B55M;wVlGl^ZNf87@QfjgsvXChdH+;Gnw$c>0DB*czU@b!RXnW?Pt4=E{NZ!; zhL&UAO8G%w5qZQPH*KjUC*Brf?ZvYC%AByG;rc>H;H$BcZ_YJNO6BqICpR+n_1{?(t38n=WV3%hH59Ay027FE<(=~k2YDR-0bidPlISDB$Fi| zMzOc`fo}9I%~=7VitMyWxQX@BM4bKwjl(Ori^HL()d6p+WZlNgd@apzMRqFpju7vF zlhyOzKU}30w^oXJ9>^rVQT{BLbLM07FT6Hna`~8}^B|6Ui~9a??!7lv+fF>Yef2*b zX+j88`pdRO6R{mXA1lMcpB+nOhgJ{8Pt%ndj4bln{4r%@fNTL+WbyNC!Cj%NM%cGeT3%E)Wo|nGm7e%+lYfK!m>(oM24Zop7x@Ht_{5k$%T7ZDy_` z{RT4Adi;69(fNK0x>hf3TnF^}L)_I^p^EX z_=ac2DOC5X{On=d!BqRGwCViUXphY+xmBrlF59g?tv3KcjKSTexpl#Fu9vr7C)s`; zedpExDHAR&MM@dQ$k9a9|etN>A zcr;!7=&8uZfXi>4;LB%|XSUaNui~$LRY-M<+B{|4XHNidLKRtF=K|nnj^0e7g@hqX zw?#<%G(>^r^|5}rDciS}wK9irDqIc6$pW9%fl;1u+qV_pZBS-dWYnB6s z)%vcA6!s`02A$Xbddzi}aGx-tOtJyH?Y;^=`5$c(MK2<`s^L{$TP` zo|UDwHq*~g&0_f*D=o8e;otJ=9j!lz+Biptx&j{8)U0K+ zNCOaaC_1x?`nh2UoK3oxQ)6D6jj#FHBz0P1-Pz@7^`N1XW%p`|@vp?#nGprx>39Fv zckhvZVh))>AL>;x+2uIp-`7T!0T9eD0Z|6nb=<;XL8D|j%byv@@`EMB#-W<^!Ol|KM z9TsQ%oAr~@iOPpSJ*h|{h(B8s$%sPXKggMw;QiWffB{hc?WcJ>WvJfbN=%frx<0;6 zn3FQg!W5Vv^3%bYs8I~5&pH(J(VH55e4=2-RJx8rOLyw0N`s6o@Zeszc3p{6P(zTzm2y4YPl^T41iSOQ)@Cc6ka22u@*z($OcDp%{_WVGk zZ24~H^QYV29&9Hy$w2|{JD%%CS9Fvqz55LWtNRtzz!D%WOMSn}zRz0cAiq?9hfp9< zAAgx7j-FRs4^z=3+v}`qYF}1OE1z!Gw#|~?APW!I>o^814L!e*Qu2`YGd{M9b0kge zJ6``>5YyPI@3-mJ(7y+Rn_}A^PX0oX0C(p0J{dOe+0h{9TA3$JRxAcSI))|&UN5%V zp|(>_FS>;NB(I%HNY$a9LrA0;+-IlG4dMD2B}k9Mjg-`enU`1;di%_`N$Ac#+LM$V z^0Y%@*TTfS)-)p{BXqMS;*(ZK9lmoX?^W*k*!`p@|DK#7t*k^sH>zYPj096tqY96X zh?LwnxW9d3w{^yAX)aa0NL3;}pqZ(i#hemT-|n12Ru*WhjC-pNoj)gmw8!5tKVL5i z^e1Pqir=ra1lP1b^h+9b@6(T0Zfen>UyAhyw_2WQll+4){4)HkkD4j?H@HNMF009D1CP(Ez9Y(FL-=;rOme(l)oAELA5E z^Gi!DN_KRBBQo5X*{tYt7fT~F9gxA`v$hJ$XGs8r%%3dIaX83HNlS-LK*{_ zZdj+&c9N@sUfjB6*zjk*jZv=a@jOhp-#NYXc4j>#OI#ahCC?QEVj5*=RyQZw257hX@7C?;X;WpP~}s3Q`BGh9AgRmNKKh@m%cr zmh8XU0cG=Nc(=5EdmHd;=i}sWjp~edn!#exZ4D3(fF|@2X#^WNVr+QNJXTYatbLOG z?tyWA>`zQ_EDl?(kLt~~wzh(H64!v% zO7v%TwuEJX)$GN-^2x-mvF^Rxhw3^i|DLy9HEa#fU8v<|drjJ2%BtCvrR zohcL_{?+uMNQaOpvoqdjoBuZOoAk5r9&B->LvG#Jx^ZKRmLCIz(crLzSRH51bpTaV zp(z+_5}R(1y~xoB$ZYfRgY6G=k?_7Tu6%I~i z;+xe?Jo0C=EiLrCu@-vuo&h;RknpEaK0i1Z4GiQ4+~(6-S4^tRpUPE2Qx2v!G>O#T z2B=Hi&bQE{1B%?HNRCMj*^avW)1at~PjT<_@MrMSlm-dE_{JnIkprTkL8~J4A!;}A zxAnE88L)VrerGyXVDZzy_snfVk*L;>tg!+}s1aW|lFHWfG3dcNGR>ikiWn(0gL7Sg zQ*dNvkQ`pEsfGSqf=y|kyTMxnf=a94L2n9^f9CPMidH8RLk6 z7&rt_(|@5fI9s0rqakT%J9ov$lLu$i4k~n*GMM3xtWVi^8R#Y&A5Rf7k)}EY=~)Z9VrU=1Oh-3p>`87dvenrGgZiD{8^z?g^Gd|pXMw{I0UZ;q_{}ujfWo4 z-SbL94Y-pEAr3Irgr6qHKK!a;1GklCso->SwhSRU2EV^OKww(cwVnowh(Hm>gc{y#8+meo z0dzX9A1PuvQ&W>_ap91*QkORV$iBXk4bd-6i%Wpe?d^=6MMWWeY^oioxKbQ^v*cZ~ zXp(M*M&=7Zc-lws$oKgM3Gz~5U|}WGe_mbZ-pUY0{n%oIR;_pF>401`bQbjEQ}gq~ zgivXyxV}WDt2=G7H~DbJ9JF$YU*Bz9SQ&sxCP2QcZ=dqGuV2UbX#UcYDRNd(RimIb za7xYMrF@&kCdf@ukmY*aEMC;V5P$_(yo;47y!>;vadvj5GxhX+r={rC&V7Q?x9ceb zy&Qb}Z$HvBH8`)V1-u&i5I2JOSR#-aWjA2X(lvAMtU7w6Ty+&FxwyGiq@DK3}^!>@)=p0h0n-zU%ygNtHbiCeRL)~ z@hN3GW1wm!AN+kM=?OZSSKq~ypKToSG6Dg+OP#c)A%8EEtg;O}zI9=6I>FF~M1F5F z=D=!Jb*Z;QmQ5^%g@2+;|FgD_sG%lw(9rt^zy*^d)=FrxWT?bJK*abQHpp33iCIdB zM?CrX%W=k55?GFx76ZwZ!u6GXG%>A`t9I=;Nl2>W%h_)FDsCPgYmr(F&DDI0lJP>r zYuY0X2-8?15Czm?dv1S=eM0gsP7>U+{lH^N$|TK-A%?XVKvK1((Z4Qif7?(SJc$3H zfdi&ZY8vpV;Agv(%F?QH6z|X1N!} z?O&qWl)fY2BP3tSkr}FAa1X6lQ0!`EG>qwx2J(;9%BelB~D;I9fzP^UbXKW^EyV)jjjMXwB<1!9ti z0>~_k#h%3o<%^l-EMqlvl9m9G0MK(lBc{jrDqwEz0TrdISpkNhi8S$QG`FMZ)t)@* zC03M|C}Ae6-Rgl7F;>D_gM~mBh|hDND#$VP@IH^YJ*Z$ccj^n6PfZk6nvudLAx1lV z5#aatASRZv1rDI8uJE$O?X~DyagFFt$HIX`@bjdFx!3{kmY}0;8(J8G#l#2+chhGn zUKI!bWEEveFwyA(8^akO+Cn0(i0s3os~_JV&Ua4x;fyu!30hT;Me|}QC=cyyMsECb z6de1yApGBnp@co!czUj`f*QpL797@hpJ~)Blt6fkxZLRq%W{Zsn0Hso+uhnapHe__@USKX+K(@e#Fq1!_jcm*wC}@edK>Sj>)x< zmV>Uvio0pws?PFlIcsuwBM;96))*MeQJ(2PjUdJ_V=&iP^i`-Bs|yFZ=BNyb2=a%q z#5E)Ymgy=;qqLo)&7vKXZ&OSD+%~*%Lz!%{e)4%7;K)~mTyv>M{aD&zyvxgL@{oY& z=}K|&um7nfuiVK!Qw1q!xmNIBKs+I~RG;>ii_UBJTPD{j9?ED%(JM`U73$PtgJ;!| zD$KC-RQ>TFwPm*$Bu>ih%S%C#*_#<({&7;$BmxQnRUv!t>a`RCQcOf2H;#mPw1|$6 z)#?|JpsEYfrNaF7O7IMT$mh(l+YBRB*Yz^IgGPGLA1;Pe#`=nh5=@a~ml=&Ky^tYa zocy{BN}bmT$P+YB3;;u2RfwYNajyuUdY04vo1UIlK5xtcIct-ge>7Bu(T!wRT1ULU zd%kLKRf_?&grf99AK@d+>O!ejzZ-Mouk9>VpjExW2xZEDUp%6ckLYW^kko=YtjY9mEmDS^(gPlI+ z09<34KxC*^6H})RC;8zi1r{cQlqP)as+Y|?x&FRAoy*8VQgK>`cZS;3)Bji@ECq>Yya+0aG!(JU ztaskMQP(~SBTh`W{y6{8qBc6Nqgs5G;(oGy0e_)jR+g6f*SHIZ_Uo!YTgP5?b)LsX zckZuol}Z(5k7jDXEu$B1iZC3K;-hl)qm75# z*9tls)nJ_+7uQ6$e!8e60%=!hF@wT9n@uDH)uIZC>M5`2v)5ga;bGV;{(;lC2D4FV z_Dvk3ZlEC$DOpPwAJEQE&#OiYVu_HSy{RI za!GLg{zGY3ljsjSdU|9Ro#CWX`kRD8Q_ZmoDgu==bVA=m^gg8|z+rZBm1^m1P_1)O zXvWDsbG_l^ityTqPP1igpE$5$RRK)mf0N=F;FtBKsiX^$ZGyi>OZ8DDPLMen{w9h$ zlfytKE}Z08FWvua;X3!Wjg?jGPa&6~dE=Xz#VU9oj=OgQ|1)QpMbiP%E~%t_qVK* zZ@MM33sb|!(GS83IqFU1f(}%rUKDC*AeHWz?~eTaN(Aw3?D2#g%f=}_NMC$~FX+&n z#{fW;_+ms>Ma{xpyN&Wqd4Gz~U*P9*whx96la|k7VtR9GkAFL@PgkB#+62uBFAd7UCIfjCFQ^*ClGCCL=>d)y z=tIUE1H2>$0N~oli0^CO#oAwvNfC z*VfU}XvToGg<|lL%;jL+ZkbCkCUUxKQlGVkGw!= zeim13`FXX)F}kr|R8-WyIPdo9PWYd1|E!kJCc!a4-o|t+>*Snxp^G#WE#=4)T*VF5 zWHS8FW}eRC;LM=`m+B>Hpt3Tt+ISOYXLYm*1XBx()~mt1X^umav&MO!8PYDE%5@$ zP>T472nVM8N(%X}nM7E@PGex_{m61QDTzLVzdQi)n)}bzH^~3rj9Z}nHEKy!duuC( z;kfCYOE5Uv?-4r#w8R+Twe&EqCugMWmICPk_Vfts2-r$)bk4tBWi zFm*PEc$>ViZw));*b70AJg9j@n++7)>O&3HB-7*F z9<{6un1XYA1bq5zma#{;R5c|2v74-NgV*ZI}l+DkVi{+TY>rQXrL$i@g6T zNMY~DGeq&?UwPIdNsUNB`jG_h_Ok#6smc^0S$sXEYJLoC(yRh;^I^1Wx$O9?|Ul~aAt~f7QE@yOVAfnUSb4UHO1ckeOT!D z+EE{&;=1sNaUT-yG}G}^z++0hlGzT3eW?1Uew*K1LjVAxd;-)?>H7EOalJ(kc!5aN z4xU`gmr43ZIdj5G_+8U%HH#(|^Hp+xt66B(p$*rnXOLu!k`NKfQqeII@uQR-$3*5=1{OSCV6GP`zm@39@y3BA74=%~NUjtFL8*}d z<(f3b{Yu`3hR?2Wt#5qRYj)47hT|-xM!~)O>pLZtQJGKgwvnK=w&*E-eZkbX5Q3ii&Unin}FU*3)iS8VrC|f>MN|iZOMVr$x@u zcwf+#rWDn|PiQFpRY=;}-r>H%W5|bv*7l?cn_Mg^5N@K3{7faO^7;&tbCJ9zj_dqQ zcpSbxWx$C7iV?y>Z@>K|JFiJu6(+bIKWse{QT<{aY;R1jiP1+CDy_-Gg&-JTD+wFl zwmYv#X~uK7Xca?itD4ggDX`W(bDXiEZt%T`)!gWURh)KWQ;JmCJ8LE~P-uGE1nZ8< z=BLD<8F9|>{mvF|V~9_mNk@E#9A(LEeErD^v3CdGOLygGPl!?Ng>=D`NCf_ueU$o) z!d+3_vIq%Ie7N$}p8L>cX`1!Yz0QmD2e$oG+PIOh`Ol4S0F!qk&gMAXP~zd0`2JEl zeDFx!b0xLhQN~1ejhE5LCa9uwdg^JUBNmQ=g=>avBwk!b>j;*}htV>19Q}Dw8}hw0 ztMC7iLtz+mFF$?dkbXWu(n7p))tWAa?!SpRcycBkMn+{A$US%WBYO-k|q1*L!1P;im|%NW*V0u z+Z4r;S{WPm1q~7h*xu!%JGJb^Y0^wis8eM&zB>Fb@D(FPivJlyq`HPH^~ebBEkGi? zJPWm^P(9lo*3XYf)ksZk6P3oeikhbfDOF~Xo;koO*K;$w&bBVkcyfIMZE(FEyP;>( z?YZ;@y>pFJBzvEEThf&p0?UflcWxD+4g=EHHMc|}P{F2>H}zG^Cg|qpl6gzgPCN366mL%L zrib&p>DLW=O+|$%XAdEiB?66tEd_s zzu@tl8#f#ZXul-GT%W0PWRWKkJFx5mbL35e_fK6G9G(G$AhH#kv-2o?$d>U6g}RvR zEp7ASXt4JET^hY9)5)3H*hs8p+!6NAHrLI_3CU4+X>!NNu5zntG1)vXZAj#xQ;ACh z=WBw6N{Y3YZdIf3)nmi@Qz-gKswE{iM|#g~#>BU;-)x=j2CM34YD385Jyrw)ZN%G+ zg=Ye8?~V`&2ssHD^kubN94V^;qzy}U)s3$k;{cmI0AVL70Tj434V)v^unRCM!>rF0 zgl?kwd?=6kZ}GYIv?u9!`{;&M(0MT2O%5uIc*n54JF+1%{@`_^LekgdXEtzHSbn*( z9ABaAo3p`$&sh%%+z2$rfP|y!U4?wbS|P2cFE~*!aVYE)p(8Vg&H^LiYHQZl*){U6 zoDJ=a+CaKYmew=~v_F&bF262$v-ahE0w)M)tcCQ|xo$^+lcdHPQ&OEznzO44lx!I? z*=Emx;1tNP?6q7IaI&VUNZ;LdlnO`#*J#$xLd8eOU=uTs`RFdapD$2yP{Nz^6O{77 zf{`P}WlL6C@3#aOEkC2fD|C3#~NOCiZ20GTxF_OEwoL{9#?uhrOLP@I~ALTq#)Khj=}*U)LR3L=Iiw0UoXThk$sY;j%}24p1g$$`lcO7BYfM}8DiRk)%}oUN zlB19%DC!X$Ud3ljFd`dLe#_ouB5smHg*p|0aKuZ|<8C=QS+Vn~^i1HhsjpG-gdBWN zdynspPP$=jB@%oRtI^ppmo-(bG8K=~Ar7qjK1OVUTDvfa7hz^S6Vfcaf#2G!Ys(mH z3D$4|7GwKqg6b!VdiPiwtCc}0F+=dLNNrp{FPE!WOg z`HiV{TDJ&o1+=e2AcMG+o|C@k!%M3L)o)G}R#D6pVPT|Ljlohw|MJ+_f!*Lun5W^d zvycoL$?e*ga!v@tMoTK=Zk!a8ns{S&0G3acU&;h!%xXlPZ;4<%8(dinbo%Tkb=GE4 zvlQ%6fTzGPKf@ag4tcB+wO1Qwtfx3K)kqZ8QKVH^LgLXzl}fDZ)5A+7t*XSkct`__ zI#omJOmShj5X+q@;YuK0ONhK0lJ4K}SO|Fbl}TAq<{gT|0H_g%ggZk4#&iZ z9)7r5K8;8+k@M}eD-W0xtRA6TwZbk@5wjZ&Zu4W)EEIO zhLGTf{gb{wT#stVdSY`UOdZwA#3VK(tzPk*zH9@|W(UAVI#5B-g?9v0{2`HT_|Vh& zLl>IUFQ3Fx&|xHMV#?rX&2CS4dWs~6Jry##&0O!G_Vs>|7*p6o^N>G}f_87&9K%5Z zEZ9ViNPQusx>Y7fe6%M_B`wCKw~O13ao|&t>bISlsHz{Hd5>`KgokTu|J_%FdQ)vl!>iKb2n!;*hC1x5Ck5 z_zjfiv(L|%i%eT`%Vv)PtTm|giPR>Bt0z@g%{Ih{nBr8S7&%0Zf8hQ*E6?CXcbl`O z!}f&9>qwFS1=A7sJ+4I!O`q{V~8WT>OvS7pYpj(&_Q^{my^&<>GdN zk=z5%a2_xib)yheozvE_nDW-&Wd&m>hr`$HJaJxNF%u8NS4;y7V9!_0pKIoU|7MRc z^I{U0!Huwt@O(%(U1sRv+CQK8dXI&?^YyHltlm)|ghD*L3yn-Fp#Mc0i=eazq-dg4 zit(p4Ep2+wIZ|{t%!?sV3`$&cdxeNEuF@7*qtHrAexhUd|ZXItEPVMcJMAqZl9pFO4Ldcv<{AZTDUq&SfR z0EhJ;Q_5vAmBdB zH-G{ZtUNnUw;DVa3bS$~95!vEI*-!#eVkaJsF14{(--ptUARgUV`3N5JQcxWl^L0m zb*x_pVjch|Nt`0}KBv~rK{tU?gG7gh(!D@1 z=Q4tCf0K}i$vP&|J8{RTN5(5WQP?5eP_hU-TVKqPvFVx|pt;9~W4ZfQP>`rWGI>2; zmGyn^h0QW~=*XBcw$j!N9A6ZK^MAZ6%^;qP24K3rGJ79DA90N3j*M$?*%7{jkEilJ z1fKj|tH0oOIA8xzvhYXvdI7g;q{8KUKrty2w|6dsw~LQ-q@jot{s^DX?o`-hpG|l< zg&Xlc)y&kG0PS8HX~b46psQxkysh*0U!&fx1RCsz*Lj!TD(8 zK4OSj<~PwKwMgQ8go0gWT>fJ)DXcFwTKIyKR`;^^x=e~%vo{hN0ReL#yCs@RiEUn|}>-yILgsxbYa{nY&Dy8&YtLvbw;Z}g@F zSYe6+k@i!-i7@h*m)bu6T)gY*;V=cEwJp$j@|WbSg^2i9CMOmN`WJ!g^(|CN6xVQR zF(WL%&VX2#Z1MdkZ5OvXuM9ZF^C39dDk>E6CLb~#h43w(dpVJ!c-ZF4RCK z#2|_qqK`d#eNU4_Qr+S$qAlUk%It~zw=letcl0Zhiu{gT?q6*jUFu!6T~Q;SF!7fg zD*X2m1Vk_V6x!;UdB6^G#Q8%N-H$X_zRg55VhIiBi%h@tlvk#Hkt?42Tm)n2No5cJ zVl-1Cu|stlG_)E$B$uhC#w+D*BTclX@hWj!&S|A2WtK4r>yG z8t$diNIcz>5GqTno)>v*&ddy@2$Y5B|3})P0r$oSzh@c%pftudlz;x=07T|Tj9-Ty z9UXDhk8DJL@?U}q@LmUsw?AOElYMs`ga#z;EVyaMY2eOh9^zHwiIDtvO9-#Z_!~pa z06s7%xf}QR_~#vf{B5@aZN(=-SLfrQ&y@0#^y0I&&n}rRze&OB*t9j>gny*ap7(z> z`>Dagbt4C^js8BzKwur)TBkC@)i|Z7?f_`W<(g@b%y0Na4o+x=ZY=pqKBvLNr)cSS z^jJ~p>0QfX+B}8J!&{!fd+h58uZCgE?|d2A`%ye);F=!C34f) z1WNJTuWoB=Q@$Fkl~|Nu#=-&rDSo6Wjd<|B0swSQAyFQ#sDf-Gcan=3KIMau`cOW1 zk1aPY0^A@?ASQ)cBRyu|RVK%O*A+0xhUT#EV!OJHuJ4sCc|uQriD6k|{}x*e1M_61 z3zS0|uHQ*m>4a@8R=gISrcih+8-TbcI8=?QoccJ$S82(TH>(2Lt%2<70icCo+X7kNn zw~K3JD=XWLWRuOMi)&<*Eqg_jk`dWLMv|gjmlQX$_vYvKFMJ;7{dt^oUeEJ-Er5p- z4FOI|akZ&32OZz7GohdO8=J^7sxPP!2$vrr*lBLR;rkAOY4Xfe0|1C=5&1Y!nWy+t zW*~TDlp=hSJ_9&v7|Z*s?_g?_8c55oW9DQC&A>Rb!1jjz{HG1!DqVhU6jzt0OFc|a z$RPCKHT%JXnsq176F~qL&`{uD%KXWT*^!GVzObvEfAM9@Wol1fM=;-bshF!Eilt&= zKsu+Kp!|#@_m@beQ^nRYS>K}wJA@RTwcNjYky23y&1sOv31uLHzhm4$5XGyEf$$1M zzja2L#HzeNA@Fic@T~^-dWtGqzF#z3cr-#hj-tCS(U$|{q(SEV1xv&`L|iDzW{K6B z)v{%owX)O)1nnLsM+;q?`+SIaO)s7$b9jD$_UOwmFbyF-LT^pTs*;H6U&GquZneq@ z07iGuTIx&Yj~mqO6hH}q!M4-q&CPxFfdpr4DSiEaa>+2VfZW$2xq<{8Ab4BKcrzBcK>m3Yrl_FOCN8rAt?I%Q#t2a_`~`%^{`-MH({cqX#)lx3QXUSj^SA0SRf5> zxF8=imo!?Jc&AUhjylYW1s=LG&Y1 z+g5vmI)EGYki!+ABa2>q zQ>0^PSP>Z5GNq;_?i>=*cB|2^!)r zUN!X`XKiIcn&nO>#4)AK^ZkGkb%g+EGc5@mBo@BXCr(IA4!ea@t5-=!#kT_=RIJl{ zqW&f;!;FN={$^@BUc^%V-gp~e&ky_L@BAdL-crS`LYr?X(1 z3^CtSmoR^d&(jKQ)O*8Nmjz@jU^e|aC*9)}NI!toF=WIbzb^com9lAjEc-j7|E|u! zLIzFNBG>B$AKqP#*tqS$mb2}9rwQ|3K6789h&lmcsuIjHP&tYwL$*f`yr2zKaGAY!Kn0TO$I67 zkLt6bim&E8+n_!#Sb9PUP_1SXj|C5NxU^4@%#uMB?flMw7uf<48|tZB?21gdI}DGb zcazsw9ik&c_@;u+pJ{yH9=&ykU-{L?o;>PnluuJJ`L`aJLK?RWN8~YPi@Y0}Db#xo zytW#*4G9sx90EV;s;zPN!N|85YzQL(htyMU@Ym$ew7Efh*^^+cA1ak}{wMn4+=&E{ z>ru67yqOvZ(F7o`-luB3OPVWr@11w#VI?h{GMtbkoBV@d^B_dJDwsX4|BATDvKHBv zR$g4_q{qPCsz)^brPFgoQ2mysqmz3b)|y{L^rU45x5d#)%a%j5oDnj7FOi@_=r@nw zI2LD1{whQEDjU~Vl&~=3yu*n#8cDjv*`d&4!7m8+=c*?i1bMoU_?h5StE}*zw=Hus zijqZ|oG^`xx{i4e%(Nyq*LPEZpKD(0(PtkApU3Lrvk(DVNeU^UGaDU&0 zgQvr6JXjYgL?8P{{qZMygt#^{TmTIh5a5Q7)!)}K^T7(VaZUtCahW?I7~v^p*Zpvh zWz{%iWDQDf+=!=&*C+CTqH4c>En@+LvM-mNlCA|9cVBLdtG{n=57)Se`un*^>ko~g zYvM*xOkboV+AvXi6!xsBu=9m}FSSc%x@H_8j}l-C&WwX38uawmDTm$P!#M$mUNG}y zr47F4R!_&l6SZh8MW1<2ye4grd`+|cvJ!amk$91#af_sK60!pT^|eM)|p7jM2xCtA>!JOf#01B6~=M5I(T?@&6+Sj+*^uBVhggC zT4w{1ytimz=V$j0)bZ9%$LT9H*?Zi0}ea6=ahtqZP)ljOtnurT;4BEtI z3^~k{^^jzoXCL|S2spN;!RcL%Fn$%ndA2&1st2F2IB+rA=5u_9M#w z<73!>P+x%a0CEGqDUWshhD7sL@K&BmqZ+WQGnzl^cUTh}4trRB(df>n-AA5=CNp{@73u%M?nymH-1H}dBb*aWAN)&+g1I}dxvwkR^M%O&aIt1+FTFG z*fEI~=(Wi|-k*w)7BdxPGa0EFrgTH(Y2Sz?WV_X%zC3CnH8fQx-1}0JM--|2P>kye z8}`~imap`O3Yvqcww9Y;J$`A!1PER9QuykxJU_h~n*TRc>5qkHJvvcSe021t8NPzn zk$dH<0El6n%XDVYNWWM1V6O;IsK}W_qM<+31mwiCLiRX1ScPIM0{hR-#<;4wA*@I@ z(T0}g)HoWswrd~^qv$9`3^gjY#^ywxvu#u)cf8ZM%=&JVHh$Um>T*@%B47B$+a?&V zzPexP#;7>jC8Uai64@NNzcBcZ8j#Uj57cbo2kCP3KNkpSeKz%vIzp16gH2kdoo+#k z75_&rWUaCrY7SPz=#MR;FQ)ERF&b~FiyrJMoI0GZd5O}4>$+pnt%l~x0#r0QX|YfM zXI?INnZfQTH_Ek^ey*L}Ef5Z?|A_FCAAll-YMQw1g%!UR1M}2%QYL1a1xmcirI}rR zaMV9TIVYKjBJqZy@4qRt4uSzr{8-3Us_wGd)L4`I>K7dV6S%@$q0Wq7?w- z>p?td;EjSfh0R*&Y{GN^aaY_)I^gnW@%heGW1Fruz%d^R^Y-qG;3$Rg!)-Qcw30I9 zS&J+X&i1ZK-&dP?dDXE%Q?UA3?W!_2XL`IF8bR~sDdez&BEO-v(gXpM_RN?VXwd4H znTVY#Z&bvDBC7I$Q`T&gr_65T!0p`It;#{QYycgq%wkI?_zK;ULT^1sou) zo61|AY|}yzq2zpXIA)`ZacWmn^S7D529wjulzT$MC?cbN*LRSkXkny3)6a_vj9BOk zw-R+!Q@x6ikdrgd7R%}uXLznkr~SLFIr7lS&vPX`H}NVqkE2ajqL*22we#Q|PbB`D z<@vc!XVhjqylnG)kB1+63n6p2BoS>*NTd~hTCLI-5Wolo;>rjFS{Ue=c=GSvmgApd zOmkzhv$_fmzcjCoCQ|n1@G$=F>)~O0hPs-ji}!mekwR~{#do^@RD@xl!WKB@3Y?s) z;uv9H&8#vkEk2gL{MA-q_>A`D3t=PeqS841vSnC~M}GA)fq0F(2WT+W=Si*f?f2=2 zYO2laR0}{ER^L_V8--h{{`H(I=)L>J2OjwhTkBVycKdwWCaFqamu6V{2BnwcZ^z!L z=NzgC7*LDDxUrw$#E=n-)|O-7Yia=&1)4;`o~1WPHvY^_3VUi{EtwUUKpk7EP~==>~M%6_&uS(XlvAx$^A1qT^02DGG<+%(LS^_mYW z#7PefZ6sP~Q_3*{!bE$MXpqCT{RGY;YC`D*(L>S7W0Wars`;Y+<&UK0=tL>;)}d z$Lmi?1SfJ{r6c0M5lx*+7DFA~bhC_fIvBGF>Lh0A6CC$q*WzD50)%e2A1xAh7|E(* znmTd1aZ_YR$sQptMjgk;l&(q>!%t&fElSEO3uZ=plBviGKy43F3jwT;L2I_ScT^v@ z<=q$S;}{I5U%46aZYn-c+Jc{s8U-|&h(5XB72$Is)416QCf{#1osGK8 zCciZUQOPC?uNwpMLxt*q{>Ks;encClCYLewE75IjMKgH z;+RgoVG_&%y$L$o(s`1QkT7v@bB3KUjg39P-E;Q5EE-;bdiTM&tCV4M9j9 zJ91epj3r}%bAr?CNIHH|>$zsqXwgbtKg&=$j3uc&DJXrGb z>TRu`I+!bfIWgsO0p2)rZP3V|IQ2y2qV^X@cDN%W!op6T`4n+z?Q!1x+OFMd`&7F8 ziTbeJAr+x@l#<>Aj^O2MRuu`vgTaI*3ykAcFa;EWQ5r%t4|>8J5iO<$K-5Iddkyw1 zAd_icgUCxU8)cJt{!)>oJN3Ci?*v%i0nJ6-1 zLb^7*AN7kiafD>o`bB*E8i3aM*1Kz$&CKuD+SMwO1+7Xcf{Fi_mCP|Od?`|;!oOoN z91#yehV)Dy%5@*V70-G0fQ6N-eLb8NixT|soZQ~Nv=Q|xV8Sd}g!xY8q*=9fwOAPd zVz>D7XV?!^mP52}XZZVX$y4u*nDv15O7|SqSLC2@%;14M}bK^Wk2R4cDgaS39H{pw`h9J zIlDi#Yhmm2X%2&#QG}fZY0A+p0VM!|9=Q6aJ?FI)ac)*81d6zlx&Y&^( zF64`>p8T@cjC5*xd>CXhw2Lg1{>R*41^H{Tl7CZ%&E zcKFD&vAwE}J)XT6H42D6p47#Vnn|-bC_ww~{K*F2#lUpe6GIY6h6ki=BL7D5Y3M;J z5H4Q(ps%xGa$Xj+ee9i=eRVZcIBp~M@09ezoHbPT{ zlM}e{nH?b%pZF=v@u>nIn@ua>`F(AWTdgIDhd+Vx4g4`gfM0+L@zm$^XwT{LayQDp zIF-8!+FX>1He}OE8ZHy#Aj^n{x6Ij4YeMFyctodp(Stk17FJ0t3+4E)OgT2<&j9{K zip(}GOCO-3FH@$Bb13i0TH%v21C?oOpRG2J_y&TvI`0+Bf=Ca4@I^=JT;0e9V15;! zSMFc?c+HgcfLBulHOv7V%>9=}m-K`K-=<+Q)Mi`T;Qf6fhTsmTeX zzX??6x408=r;s({OUxu~rL%MvrpQwNQ!pb>;y;VHowsm4D90TZ(65a@+cOe!PWWK1 z4YtUm8EPlU#QJ(rxXs~eP_)e{hB072!iYjIUn(&Rscbfqk{Rv3@##nap># zCA_NBVa^3sEEO}htTLCNMJnqFZD2jqc)J~_V&wu-{A|ym`qNkj%T-kWWFX&XtgF^2 zGpfe!a=q((BzpU5*S~ogeYx}_cn`>8XjXgG@_|iZ|4}sIzES#Nu}+e9vy*-NilzrT z;R;BUk9ycq$!KTVlpM1zm&n6Dm=4!%A;{?9pl$%Qmf9SRAxhP5F6(`oCeE!kod!%F zh|g8#ym_WFGpsIhUkI+}w6Ww`Y+WzApanl_qh z%gV{(wukuJsRP5U2MX#E)HVjYLJm9lqAzs2&i#b=HOK8BrM_Xjs88D~9mfOO;mOr+ zYCeM>&c%i9$=u*O?~`#X5}K;UTU2?xs`U3sT<}t;Z`)82x{iFO1!0&_@wbt-Z8S7? zLKm@1#Eo6Eq)AMiP058%JiIS_=lgpo3dbz#1St;%Q^n{y5X~P;BRE}5>es0%FdTf@ zAwR~C+{_2?uySIy7}}7Vg$2mQK(N5;1iJ&%%iBYp$-kD{$Fv7#92b`fCu?pVrwu{E zsax_s*a5?$IivMQL_8By{(VqI8i|Uo82brui1U@CEg4d}E!#Ho91v?#;fvOUn;{2G z+XrIfwx;TW#7G%hnKlai`UEGM7zQYI{Bs<=f}+I8pUHhC2~GNRbT0IYl6mu>K2%ol zba`P+%PcM&sUt0?&ufwzeZCtk<2#HWXQI%OI7Ny?=K`iik9cjC6Gjk*awdqq4 zb1fE}xW0}BKKmA)An-4||FzQK5jnzze8?XYeRg@_7vdexHI|HNbmU*eJxqu3LSyi! z>fG-02_kL`iut|>SZmr`d*bi*;7$9^56q%LWIZKrY5vhHU!SR9_E?*34=)=$JfRP( zuEVmu+uIKSn`<3LL|i965Mx?M&xP{x{y#cE0=0tgu4NHTQoM~3(f%)Ej2}n3*W@c& z`PIvFHAsWLba3+Wv4ht{LQ@8(*Ix4`Z>ViMb@CVK$C_MZ-pIZE6>@oAc>9xb?R68q z6@w8^>=h7C&Tvso|0ORYeY#wprrFqD zdV|7=&1~f5X7-~iMxQ4d`u=#q0_+9F5@OsHq0()S2PewqZqvtSE?q=WJMwnxwuvxG z7v4`vX;?7NqF0~}oIS+O_3@1DTK{`IQFN+y`FmP@e0LkPcBLpq9l8neazhn@$Mv9) z!V>9F^ybg-5Wea>%P+mvNiZ^1i!E~xMDZ#Z>RTP1k);?&sNG2dB_fUt5?gH(!fx^w zb$g$RwKF2X$T-O$wON`M$J$qDo9wV)w1#u$VqHPX0~75NtL#MS7Tt5`VhVaSg46We-pf^8xrRF1qje(SzD;q+cY-kkT#!i)Nhm&^A2sl26AYbXcustSlbZAvt)h- z<&_NAg%EF9P!#?Wk!Vo8ZNS03UVG}oD3=x(jC}$bx)KJanE&H}xwak#5%10)4#=Ny zm>~r&KkJ(4EB)@3vj1?@b-vrxKLopy+~H`LAMPEXH=z~7Z+M7ttwidBEnT$$_j}Yq zU)2CTqnbD#guoaNh+*!UV4Scl0U?(OazToQhOP3@pj~~O)J}%Po;7?n?{Oq24YuA^ zRLuhq(Wvq~d%M4!F&3~{h3xzCG+}x-Su8+wQoh(y?6zFw>Ej)B)HY|Qp97x9Bsa(b{bq~u zS3Ftv`j1N>pd#3rmJz&V$R@7eGI*P%mJn`J!PF*z<-TtKbrzv9xAdhL>bQn|mjb@a z+WkU50AcW9x3V(#3^S$K!JOsCzW!-ma)Dx% zGaQz5&RP15xvh} zMEq=DCpyA_+~KE37e9G}H7lZzUK0DA$xQ&~E1RF`8mbcQcM~v~u~;)aVL~@4lcH8O zo{p+SFwWv~zkR%R7#Iywy))&z$1E+)>I7%{SI;AQ+RNSDkzh^R2w^7!$LcihB|!Ch zhTai;Ap+VQyrY;z&t!TqE>yHOka%j=(9~JK`sYtjMw^vH6X|YvTTl+?Ubj$jjkDKHGQ}$wp|LrmB|;a4fE{aorp1<%!QO7nqS} z?*=o7G8HYncm|j4op=sln<(%(?R>3*qlmK(!G}LeR#%07^7FaIkx}w(IG_nz8)qWo zT=eEznv0{vb({c5P)x>kl=*P((0dLDVtxu!H3xTIYEI@fIn2M8BcT0NX{B&h41+h_ znBHPo`E+nA4Tc{n!ji#&66KfDYZI^1j62%7gW7Zh_~Eg#I9f20-&FJ|TkXk~x0`8J zUaE?@2?*J$Zw6;#&@Cy8v#zE5J{g#g4^kd?>wKQ{bP81G@#jz@g6-qyzq?n6-sQPN zl*a6#C%c{j{I&vQg1cA0fBhZkpbFId^PZAl zx^y%hfcV{)=MhXh9!izA)|Xrs-tdd`UOdO9sjeUkk=r+$2xqpeG}0GKw+j@r)O%qI zkfobO*fy#=zH{n3?bPTDJ>5E3-!Dz`@;vx4c(FmE;L*e)z?%PZKT^oTuj7gP%8>4u zQ>^QwBACnN%dV3J;o_(__E&+|552f-dS^X*<_bIO29pUIXJhSC2~Aq!P1&c7*?*~a zjy_fwXCfj4Rznx9Buc)SiA5eguraDd;B~Q9K!%UQb+_`e-$OtlNexuTrNL@_$sopj zDZdOGslQ@2+3$<8QKJ=xHb1UGb9TRZUwKjB-t$*&&czGDD&A8AVSa*R3!jEX@p)`1 zzRFpF5Cy*N`O9Am(wFF~yP@cCed-p zt)>X34kdBp-?DMCVRftL)5Z9t&L%kLgg;(Lhq9kuhpknR=Wt`fk0q$Qmb) zYkYX1NA8l3(kM`x5Eh&|l`NDlER5lXw%Uq?Vj5F8EYs|t$9zEW2d@V21JdicE~$jG zz%1V#9Q|5q%_0rj(J1_CdL8jNat#!o{Pd>C{(~xMKHR#wm1AndIhEMl(FIz0mnj=i zXsurpgqQtyyixhj?jHz~`u65#eCj(d6=%fPhQ01kv2=cPpG$7ssfDj6m7}=0zK4?% z_~_TTAFRg}0|TMCBsnqwF!OrDP(z0}W_YzVFK;u_QRfcWNQizhR7^ZY^H-iDKxvl1 z6qgPYu~oQpIJzKoCAV~pueDyL$aF@|vAg|7W7u3#Wng@Vph4IHq(z-vgY@T18QMiA z;y=}{c-ui@9+b|VZ~F;?AgZ|j!3v?cFup_)JP-%@aPa7QHf2YFG#nC8b!fPQd(*78 z9KpuwDAdxAn_FKZiJnMsH+|V}yZ!8Qb=$@&UsQkbz02y>Eyn1`)2ReYtKO^lb+Pz{ zhZ7S5&_rb7Ny!M^&x1Q~@sxB^d~brSDWbo#>9jqxV}qAEUv|_hP^s_*lUmaN5$+TR z%~tSyNe{NjwDjxRCz1_dxsBmMOfmPJkvN{`;VBDKD2RbX+BDTGeGAH>gYB%GoKqEa zNsZ*y;*93B2dt_J-8}c*hzeE)ezb72Fx2r+>E7$`l^<+Qn=57tso7ns7>T$}QQH5` zj)U0OE;TS;o8Uw)Yn-ChVw1!Z+roR>J=;k-o1UxXw(g=|M%y|fe;h`>j$iHi!HN^l9Me`VwlRH&omJAfK+thh z2Wg)5J)JX>Q8TyNW3B$aMGORogO9ipzLjmb%o`H2j;4OQu3+7qrBB(T&p*Dr`x?Ed z>GALlt5Uf^4KG=2_A#Np{xjrNX#T%89iWL5uB6`!%xnUWZ$RGQo@%!e@oKK^$`>n~m`$VoXZt$3 zm%gkH8OaxSTuvADZc$1{sd#+k?b&qn$?oNO^cm3rg=lx##DKy=b`Zm%QM<<{zujbe zEr!>XWj&ad>thI-sq~cgK2h(-DT4D>t|*OB%xF;2z>Eluh^33|p_`IRtn9j)@sM$g zJx2c;!fIO4*e%e}DHx9)1b#w%bm@5YXI~Hc``tH#CxfeDZC!i3%xSeFf%d9x|IIhw zY)(burAI&faiUgc(=uKicwrbxjIV|*w5XH4%V1<#J z%iZ#kp_C_!cQq0mwQ|{eD^@36=3OAE9XV#k)&D4F%lCWqE3cLMdoW2~k z0)v>)c^Q``x!_+^XaiaW=xRm&z(>J594oT%#sNIxQ#Nci3`MieUYA>G_D9~2or>g` zCw4{FzoCE`sYAUY zVa`R=`{+<=mQ;sd&!rM9k$zGXDcp9WKNIO9kD6%~p==3C=?+~mOxCQ~Iq+Z}aGTP& zsEus?H&M0SSln?wA=MJVs+n%a!4q>>zfNJCv$Wlu(`%VXkc)wloo@#pYgaPq3BWmy-i~nx_Hv>(l2<^r3mrxLpXbSnf8ml zXKW*QJV6VSD8mXYvrcEdvWatUr2uitCTMwGb#L&4%(&(lYlAAD zn0y=-?&V{Bvhq&JQAOX*$j&t(i2WdF7ThYv-Y zo00+F+sz|GH424QB_PowFf~@Q7=ik6mUNwkB$ukxH`PxXCJMZ!v%L}Bc-U0S)!x5j zb9Fs(dvtI&SJ+p5~^W1X87iUC|uC)E>+b3d?V2;@ZM5?gU-w4M8X50r?Nu zbZxb$027ZaxXty;z62~WgV&6ST}K7uQaK|F=MfPRXU}{_%ji=^OSR7dUEyWC+T41} zr&vNdLkbvJ3mU@UpO2s_X_?$h19SiS+yBcrF{HNK4@7~GY!iAl^&N+nA=uv4ndZTQ zCRm2k%9L99wJpaw!mD~+h^DK;t?y>WeP0mT5Hq97X4|fUFQ92s+fSb@;w_eTJIP|z znL`a?Mz6TYz)jhBpAI!I&k=fLHk$U%+o)ewIL~XEPQ5m`(0rM08q*O}f0Kww*svSTiAr zQ{%-+=Z6S>=7cs@1}c@}f24c&*z(x0-4Km;4FUsBeF$p0!^n5u?YauHVa*e-+tXmV zrr^mpJ|r*c$rUs5N`=J*xpBXPHanxGz$g5}Y43dJ<5lzdI|A`*SoJw|`+56A!iWz00>|Fwkw0aRxYo1e~tlyUUlI zxWXozC`=;-OV0%y|K9jsWB6Pv<&?KwNzUzSh4gPI5uB)Y0#osfhg}>`UW4jj54V;R zkMDP-xSakBzx;ep7{EqVWmB5K&!?^`iD;trf12kgeHu|yuJXnR<$X^ivvOW%5%{ae zsd>YjLca!>!BF@18>dJFp!~TM^GYV*XESkGCGINk@O0T-Xz*zyp7Tnud))+&|EVv~ z$|#QT3ZI$oLAvmu_l{bx0g#HMv8-zE<6!lq(4w?kzb-B=x{j`z8Ic~}V0$AboSCPf z8A9^;@F@M<$yDhYJTGQNX9AG!F=%3lh(-CWf+v3Z;k7Z)WD}nh&u{-3NTlyu?;WmZmcREKj}Ppq}x*af}$i!Jxyr9+p8 z1R+Uh;T`v{ayBkqtk}mQz#Kx#YDbb8S4_^aeLNJ@TwlX6hUdzz0 znQ0d&eTx2}`q&6z`TlDU^QJn7Yt8&-n||NCK6r_lZcQaVy28DLsCmjF~h`W56Br8GjSqxHIzcO!c5 zBo2W4W8wU&)o!soM^2_ls8!?BejCb$K7i4*(1?{u^?WJWBTq1<)xxppZ)`eedTtRF zwSpD&10@xc3gYh1DIx7RCRLqAfr=RMj^IKTQ6`q&yi%h{Ux8l3BDw;>fV~C7bb2`E zc|u195SE9G>iF#xb+d6kbfV8m^j|7Ylg8NdH9{m$#7q&cm!5~9Xn0p-gi4%l;wSft z^eeLtVQ>J+%C^$a*V8ed>Tre;4;NqngdRcQrii7k@qciN-|4Y+kkm9=p=XAE(Nz3l zAv5;T=kL0>`s}{N_dt-bnVDiO9k-jfd6F{cX70#F-pitzGpxpF#^EofhEeXAY63EndL$qMKQiv$_Gsl1&wZ9(n%_P7J-l#vbT&9s#??o*Buv*Vr;G3#= ziQ8Qkth(YKa)nI`@nl;tDk}03+7=hy;JcBcPo%2e4Mh}PVaFlaq0V_3y{8+^!Opxl zX9b}DBHF@veiYT0xJ7)f9^u&f?U|F%(lS&?syf~~x$*iyEqSM5!=<$ssSP$ZX7t(< zJ9K;M+HdkbRy@Em4)0O`#iUNwg#Efke=UPY&~iu(N1)T4k);42lq3u#9s^`1-%^WL zyS*#5oVsHC3vLIyX-ZR21HjCW0=VD={7xO}_CGIM9`qNci-PF$l{=k}#FliQ}_o=|vFzd6_9*0b}9BKqC>PDDyh}`k* ze0~_5);-B_mDxE4=4@9}Z{mndI2-7=*hwaS@mDUq+NnnGzK*^a>^i39K42}OP1_>4 z<%!*eI`wP5a$8ey>?Q-)q&EJlgWRa^(2hVpBl#&Q_x7jkBSU`P(OlC!=ov|dcZRlR zx%JY+A*(IeFrw%r*Rz55_C?;O$s>!vjiQu(Sg#|0j~A(m^7X2s@y7`gjqmKdG2i6? zPmuB2l!JSmA+hLSmqeN@kU|gcy^Y%OPLSeilrlIzDyNzByIy7(|5Z&sYwP=64ewJn zZ?GZ#h~8jI+h`<74cL>odY;@pzLopwXT=+yo_|)nMOgV=SxI zzjkv8$5)&TtMG-q~@T(#5NQ9J9_i=6Nf_ zt}GmLrqAZmG`dIwIB~lheiba*87bZy@gTI~_>z0ZlkTAT!^l}uv050GTj5W$m#srg zgLVop+Y^A!IQKh-zRKAy#k@cMHoV0%vUmIA@O|)mF)`+6FXp~Kw~(gc=Ghqw+6Qk^ zWW@ix+7}t8l`{=ucHbzZ*Y*$F8B00nf@lhQthgo{*&b~zKs{_Fv_KWseJiaSf{g)4 zSlr(}qHUSL_vLs125FOoi0vu{&YmnEFO0lC+t`e(HWEYf%Cjz+>8(tK&4+vTAjY;m z7zH_^>XdMGAXG~S?57DUi|OK&EHGn$*g}+2zc)?BT7{HB;qk~#^#QVN zicWQVIAYe;O5W2}wDLbugN4N09#d5lSkq!#908iNK6i2@TPY%no0uB8F21X%4{3B9 z&RrKvCJG>-;^|u}6DWndV_Q`_k%z6}G&wdw6x#ZJv6Bzk0Fp)r&oT9VKM>*IEK- zGEhQ+$<2cq^%HEW^7O!dwKg1@iBD%GCcAa1niqE8YASlFlcq+_Y0{?(o;cMi!)Si{ z-WLTcW`Ss`M^q7sgMZqy3{Z-Q*1vP%xq16IZms9w(dsJ>Vn}n`p|iv28KTCh(Ves7 z)b#X$1Wxu%FyGK7@12vKdB!NO5DU9X!c(J0Hnn@7%b%QwCBIuzz4zJemH)lN2Jn!w z6yVcbzVt$HOC&OM80yVQJo$Wvdg42s~#3XJD%#3<5?I`3R3sU?ay zFd6h@6dT#TZY2g1{$p1Wo=FRQb2y>EMOFFBcWd-9GpC0xUlX+zHjgq9Euwirsy@w# zRBA51+}-TNiUx_HdWR;(wWWL2;Az+rX{zBrd2ga>1Mm5acr<5Xe7T-wFY7Su+O?3N ziMR)*rhXu>=P7xR@S7o#m$%P@r~;>(n2Sfd4$SoM3x5(*5$2@i&npOp57j(y{EBQ* z@Jv}Cy(oju@&=E?V~9c2-OXIAbINIO*vZ&VV0xou&dB)Olg;R`LbS9X_h?~;1U;O| ziWaFE?9h^1%UxCSURto9BT#xxm#>38JH;^VJtm|j5>h&)k3;UXZvGnm@t~gYzC|*n zo&sJjW`0mlbmw&4SRj*t%c`ZHWi1m&r_Q$h;FATO&&i>k4%w$U!49F8e?N|?ed3v| z4|4oI`Tk=3;`}x5BRImD36y8F->d}XPm1{#8nu7Y#P)p{5g0x@{&&gS{(%#hk&TKoQA#?P*iZ;F%@H4 z%v1m5td=N4|0R50hpG7uf57)PM!-`nQXlp}K#iV>i6$22p3w`}AdX~2u5?5nRs#8D zIODDNQv1`hEeWE%yCNB5+?mp&P=c&z>63T}q$G|;&*1&F5isA)aks#ia%yVc0hVdy z39Paa57%*Gz&5zK=-UFH4u=<~M&To_&Z8dhZ*|MEQs=Ytz8t&HFgTU(ANe-8q?NHK zyZv2!rgZHx>x4a`*ipavh8pd9kn+>Z#KSR_t(s#Ue`!A zpiKagIv1fCtg~0|(xK@XX3POhfINJ2)eAH{u|2%|SSNswOx1raRY!1X|IvsDd<9F2 zXaz2+N*Mca`L8tJEx{M@a^CW|KK+WBjGOCIE5xxhXMjwz4qJ#&BNJcJwku31U=(BUr?A+osBwQ8J9B`>t~MGTPVCm*B2u@X8nL?;S1<-#FyiX#pv_T z&u~qQcVP=y9#;SVy5hqnIenF~5~_b3@>CwAJT2dRfke<`?8;I&nm)pLOuhN}&Q~#Y zu3UF_D(tSLSesYJopf2J;Ndl_Wt_-sqT}zNMntO@Ht#Tp#>M1!b+N|9OXHbjb_dJv z1gq(NjiYn(u=sRn%a8h7qbv(WdTco7bIt`0r%aCepM=%sJB5u^!mX4#!u*h2gV4j) z&%3T0tM!dLTMJadhf61l@7P9OxxjsezN!|(oNb?|k>o@buJFgY$-bc1|EWgK8$Qhl z3Fi(U-=8U(VvPg)Ho9`A8HaEFI6R-)T_;xCCd-_J@gUDOlp$r|T`dmDLo0#G#p4>$ z+{F*yG`;z`-XI7P1eC~vp>h)y$zcWCx3b#Y8|aBvfIQ~K_qv}QLb#9|<5-FdP-qPZPoxOa50y^=iGeEn7- zIY*Xcs^%3h<8!j4_~p?l@bppRpr^}%M-8X^tj+kG4I?q1?n`JDR%qnys6nd>h=T~I2Fjm6 zN@jHJ57qi?D+VI3v{{q0g2&h*zStD zo44~vYteX4_3P-WeUWGAc-b}q6u>{syHE4WD1Y}LrrT_L!D+U5h(7*t$A!`36Y^L= z@#WdyN8zVk$9?n58&_u8M%;aR+WI{Z#`ae-cd;m6-KtR0RbFj4f~5BQXO&;@`_td+ zMCBS-!kY~W&K$j5C$IAlhT)7cU}^!n8CC zm#K?8&yW7xeJ#a|2G|H`qg;P%SvN8YyCV4aHed7Q&vk?!ZcG|$oL~K)-HNK;td|F4 z-@(x$O<>#;`y_-DYHeC5HWIC7$r`sOwqzWAws<)2RUf(YNw_OKbhkcjoZJn{SS0g( z2-`mx&>Km@KmlXDp!mVAvy0c4f4UA;(nCfGqZz;!k`0%E6WdJu{Jh6?bO>1W z-jfLUEh_J^u|&+fe74(VW(=t8!yDI9l(rm?HAEEDyS)(B-t}xrK2#y#I)gYt;F%c*Z_!ETE(ul5&U}@22nONC z2aj5ci&Nh%8h0K2-n^JtZE3mKIopX6g;D?@T4axV)6E-*p!_7Gr$d}Bkq;*R1m>A| z&QHzh+^emP*i|r!4CbH^$-qq<#7vOP|2*voQg{pt4pNs$5oWP0CC`Ko>c9)s;m*8%K`STvk&Vy*SIOe=YOL#CW%r8v71tp zK2cT(J;>Of6{io1DnXsmp~A6ETLlio5iwqj?)9VQ1}6#{Ay;K!1X)SAIjFywcI6}? zG*Kqr;B`L6UeA)p{w1x-Hz@_HB;M+)Rm8S7b`TWtvl`vI#;~h;d9te@rBO2Crq`&x z1@ASn7_Vb3e&s?Os0e9$jhLlg+9)Rz!@24XU`=bPrR64NWJJM<#0T0G@fjdW!gmv==;2RHC$Gh>^*)v1u;71c}b z7qw>KjI>|^a*605nrrtBht914QCDtH-GLXWL-&@pZ+fpjdaoi-`Z{tiX7zhrp&*bHL zq>InhZ?>%N*XAL^+MX}T$K2vlKu{w1^k#vyn(%aj5WZT5ZT0wwOgbN-8;h{Y0%V!*C-&H`hp<&yBD>>}RSEB@Z zB+m(a6%KQ}f4uvUFRHTh>n^*Gj++Rihzebe^69E@zA~8Gc@p5g+F-W7eVy)L*J8ur zc~I3za==7$=g%&`Hq~Duhi*kV)wa>06ZGLqHy&if8XK=$?Eh(3miW?QRXhW6*}lss zH_^joNQTrk0c6q>3pCb=J3V^+I)e#=td`lzN&}?i-2?+wwO8vN4)$y>XfS#w8y!tY_?a<5hH7_FYB&3*TY3IO4zW)=_r$*-$5I#5_?Zhoe7Fp8Xi zM(2vg9d7UE1Q&gV<7F!PRnB=}zTxUzO>$?x^TQuj<`=UPkx1&~jdM+ja8ao<9Hces_cf{;PlU%m3s5_C@qP zeSrJjX{nPI#?4>s)j?N#*c_>^aCIOV!9oj2JncPQ!sQPOt8$DiiDhm&^RC zPu|#$!omU^d0xuXFR$IzqmcuMgE%~y#0f3oT}nfc=T4Lxj$ST zG5bJ;qzcLN#Br!Gj48^Pks(OGTq~U5Sc9Hs*dIVk+1KGC?q0rvY*{%i5(2n<<%5qS zJ4Sze>{1MX@6Y;`(?S(XEi5@>TyJA{f4JeSc{9(z%Q6>=?qT6&seAYBwqCF3U~Wi6 z#2Rg<-7Q$>N7P5gfSxH_G*@#RhZUw2Pa1@rOu~$bf0FPm$^`D zHvWenTwlNbxYp|Kx7+O>eMi^pO_@FMG#}7ATt@0QKa>h`(;kgJLb#lGZKf} z7!5+s7lpvl+|m#UQCb>cs({SoUaVQ)hDgy_Ju8Cc2=}3tsYnJJU}lu5j1jGkT7*jd z2j9&fesAi%-`(B){O6B9dE37AjgK&Zg*@C7))7(=LRC|gkwuhQg_54GYg?NN)=H6R z)*l`WfAZ3bs`}jgVe`> zX_}L~)QUwZ%>u&W{9+F15IF+t342fCc73`6s7Jnh!S5ckwfzW%9paAdWx6bn>(g9S zw7GY16c#`<3INwhFk$pjm5>pzac4kaA^S37e{|cA-do2J1gKMmw)K`VZCwe&cPgn9 zkyeNpJq1ZGb(vv(+`}XTPv@$&ntR{+S3fv=LWGaj7F9y=^b5=R#W|5!bHDY>D5_f6 zwlo~>$wZdPLVeq(nbLJwjNJfnTEx5p1(>;+R3-p-0f2qjJV_W_*KwYORhM(jV@AZ* z+V8$vOF2ocOHm?9F!u-`A}$l~aAk4MDusjyxwrB3bXy9SiPFK{9v;Wi1616dxTFIS zolX;13Af&?H|kr*1estKi8QkW2=J0Pz*!tR1ZZI-Knf5pb#9wQf-shX6hg`XS+cri zx1R1kGP4(3lw)&*y*q|N6l{ z_=B&2Q!;8@kh2XyqNE)W5=e?>W(4e`v9Q)<-&+rw&ji8KM8I(W;xdR57{UnC$AF+x ziXz0_sc%CFO3}=rry2$bgbb7=!yySE-AAIDyArd?ZXMkLn86bK@!Au|X~I%)y*e?B zW>l6DG|#hW8KbY8oo4>#3;NL)_?!yB^d8~e!-0y{lDRUwZ;qO7>1IbSdq#*TVhX6q z4oL5jSP`N1bd9y&B1^fyucgT7S0&#J-3?QEkL6scTN`niY8|aAJ+3QY>3al@X#oJV zv5rpRW8V$%OZb=@u^}BBI0A4$oX;gZY(xO)Oyum0Bv>%P2LV7>L}mn$2{EJ8s@;ZT zdLI4pnQ(+wB0Z;5ITN*J-}~vqUwm=*!3QNGZnxXt|GoO=o!@S45-?|}TuM!fF_MTp z)4~`k5Qj%#MCR$TAV9X}&h^SgEsS+Kvde^F`yi@QB_$z2A5K`bXhfH(991w( z3?%aKW*3sUHUyqd=tclZ1xFuXd6cYT9|k5^2+5=09-RwlDa_=7!Yp9Ia`X$DW2VSV zbDcyv(nkkj)fdV*J-z$lx900x|JLD*_{mSc_#gl3^Z)1%zAp0wJooLYN|*`EPfu5( zWS6<}L{Z9QFtQt_$`rzdDo*D`>-_rl>(9RUJS@3PW*s1d*$JZeOe$LHOgb;y!wyDR zP>wRYWr=8gHz5L+Oazh0%nTMvMnsgk#%Rkd;lpj5&mUNXh@}r7BMdUbyj9i5)~6EN zZb!Ylh-eX}=naZ0QI6Zb@B2$;JeS@&p%E6qLCjJnKx@}cv?-|p9ep#I`Sp8e&RKZP zU>I{fA=tyMUFQ0>**!7UX}QM(;8aVU>O71lrH8jq7kqs_l{PjFZtJiZh>+l-G#r>E z2%_6<#|(ocVvLA7b2t(}5Ddt)z<@;3&QyRI(gGm2>tGFF;VNO749YlZS)qkP9{@>M zaH>^`WLoboET8XqKA%-}+qUmso?@>^@a{TF;nPw`YjPdEGarQ>aocuptBj?n%q6tc zK=(0(G)gc_B1kDxW@7U2bP(0yjslVBs$hde!m&=owTxZB5-Z`cUOK6~yeNVr7eG+u z9Ae(w{mrLO%Zph`#Yj?`MS4r~0SQ_ujk1?y(~=LLKeir#J})vH#+EwU*b0WlQ1u+7! z4+QLp6=}+OXiz;@@AmH3H$Bn5{^HHs2irT>3h9rJ9mj2Xaba{s#E{$Djt^cq_8nNX zZBLM_mq&t>(|MlM1_EY(x|V4|R9(8vW7eHM1_IMwI2%0>jj~XYz|EP%sgEL{PdRBx6#UE*PyP z0wH9OiXb9K-KaV$ULlF&Xb2l7;_*5S=lyWYa$X`YJ|6pF(0dg}(LM8Of7x1O_F z^AtMa6l@ME^Q7z38V`>$*D?0KHycqeOS`SxI+U|43g%q-wt_va%!Fy&ibq&`B39wD zP!U!Lhlj@}-v)|KwSM&N6Of08_wdMVA2}$@kxQLPnGIcF^7Q$VClE)h4A<4m1c?ff zLI9e`6wzP4tn=yQt>4x)JAU`_@W#G+B*lL6lQ)0;(_j70H(xT0%dF%2V39cNeSC1PRsskjZ31kqvJYSc1ONZEbS^u#W!YIC zw>js!?H&7^IC<(gGnIu!mCJ>VjRg%rwuF$NfrM!JF)bdzJp>Y>VIx_%0a#TS%2kz> zc_L!R-fPXvxX=i{!E6j^4h{YP_k};4r;=xdgOCwmISBwoFfyb>F*GxR0T#re+CvI| zey&eXuUXRNaizc|^KJCws7VP|UQaxl!$401*`D^i~B}&h(H2s_hC))U|z!ech~E- z$@TN`@Tr)qHmfgDKU+eSZ~68TfM_ z%Q}DfL=GVq5k8u7!YOfZBk$9aFbq(a4c8>C4N?U{s0tycEvGaYR0Ki*j}$=Cs-k0m zb`d9B{=bYTtcy2=+b(wP#xi^vwg$u+0 z;%NZ`3J?>TR)Q7(=yZPJ??(epZUZbbJHQ*HMCKIXrqwCAM+d+X&w1*c&eR`@M4&Y3k z*f}ujz+6V{B$-!2K@3J3VAAe&j6Mb=S=I+440qRNqgya=VT$ae2o9>(>qt3GnKifv zNM^z?r@#7EzWQn*qJ7^x>U6qb;Zl~TYm-rznLU^&rbG~sGm&E(e!JI_Xqh~fa{0lW zgfRvhNJgd-0PbOT?Uw?2V}6v_qcb__$+PDtk^MOg3Syk?OEP(6s*S;)JV zB7zzQl2A(g(UJZLEXl~5Zl-*2+j;`oPf=&iQQ$`QYX^15RxZQ1?-onr377&;xQPpE_DVH+m zc||0i=B&VpF~BbmQv@0?GEXZ?5`$obhli;jwYM>?%ieTbp;auU9LGUK065GDJ=iG( z0YfMpQADO*IC8c7hGNo7E%Fk#O`u<&S|2qY6_Q9}%`y-SIb*?dTn zoWhiqfuj&DxBWQwSHd}Bw62K|gS1;Do)*4en;Dpmbs}O6?GbUBGcyGEfBdh%KA*qL zIlnni|J@(IBM4-dtRF7h-@e{6$k9!N3P?;6;W4yK^y#;RJl4}R&r2Uin5*L(*3)CO z+vr}+A%YP>O%nq~@;5bAHx2>;M*}qx@E9pG4IrSGeXK=fJEsk$j~K_OM~xZ>>pY`& z1t$nt=7JIe0+eK8NnuVBMi|y@-U=va59UllfN0o?w^1BDlze84oo^anx~uy-ck>N=xtZWe6*ekgQ-X&jwCriA>y=^ z+G?Ui4oL(P)S;yWA`gpP5&$t{=3ohcAQBWN@OhhD?X(iWtlHe74K>4Ic+>IS=bMCu z=`!a$&4jy9LILaD%*g3)zkB_i?WBj^Q5onIPZeh(`k%i%{_T$om@(04z4w^Q0|0AN z?adioIUGmlB=Z;!f#`15_df3T`!r4N-s`BxF@}{Xwc3{DZNH+9(ffGd8PF38cpKm) zCpI5r`a=hapsI*#LaPS@Vg z;ei-Y&00)zx_n3vLol4!rwoaOgpjhDMIS5(;Yj45Xk2TL;BJBL!I&fiTS5WK;f{f7 znj|p_gh`feM#-6Y3`KquLhzgs2;}?U+z}FjGUMf8Y3gQR2j1?E{wvH(#g z7}l4W(9Bd*;pYQ zXiPX>iO4mwL_g{qA~Pb)!wt-1)RB;;d28+N964DV3L3l$wxFQ!C^;ixE=-i~&DPUo z+DGq85td>el;>$#dL6wsKVDn!)4Cw}{ds@+?zU$noCpxWO5s&<8!*(V%}@Kjt`t|GSGKsfkn-CK~?}O?@D^4WNcYOCC4)x_Awr$Gp(uYX`_auS3+x0%E82WGq4CoCGg@5pv|N1*s zozLgbpTaF0?tgj&J!Y}c|JXc z4b!F;ipeR65=Vrp??+>dJj-|AK9dj^LCjx&bsmE`At4~BH*)~w1tlW^r@VN;|Neh# zN{NdOwcFiI`!MDd>cD8AQ<+bX#Rea@*wxg=TN>i`zFC3sy1)4rcpN)Kb6X@yd+kYl zUh}v?+gl{1w8~-uotZAD4Zr>SIv7$2QchWiuIGg^?R(v?Efw#UZnrwElgHkA%+uu7 z`Yt)m7(wQHKnjH@u`0p-Mj?RFr=S};6y_22s@@65+> z{K5PE^~$h6W0bdH(2k}!4EpN`bQp$20^HdFd%c;CA(S-Ymru84&7VHa=l3TJHf-(H zbwr230&l}mcmja%01CIzrWS_c0ulfKFw|=ugelMb{kP9Dueh$;7%{KK-P+!N{e7DX zH&s>Vq&YEje|omOc+SV=yeyY@X&s^a%hR#51*VlNFfJ1liUJTrM#R_*wYEcY`Rd~t z!fX!*Gl1W`g1;IQ{K-Fr|Mm;{VtV(d|6+Y3Al+|At-W@$h=+&u>#x_8QHbi5zf>V~GHnq^Zo!)PCSRpR!8sO%hO;Z}*zxKy>GRx%45JX7!yW6)<@ow72 zfB4;pK8!KAOduX(A5F`IvMe@yKMvqr*2&Zv^;TnDeY9Jm<>j?6TY>1VJ+V|r19-+s3f6Oc>g`J^$kR1{}Y>9LPawWRj-ZRXNv(mDdmN zWp3`;j{DP14RFSTnMx)haR6W_g#Y>{>)&1b|M$R1s zKu{k;MF&I>>Yr`hV^0FIfMk>T$S#U%b?rw(&~VhCT7GmKd$lbe@ff-)%Ek| zJMQ3g+_jrOJeHzN8+H;=HGUb#@eL2u$&h+Dt2be9xFqVPB4sKWT-&Yn-r7<3`^b!vr!t2b4(++5OtA00MMSvFls9@fO@J_E z*)BA?>X=Ys2@L>Z2}iFr6KNn#izE#QNkz31SW{F5ARfc#^7@Z|EC1vj)E%uiEekK{ zxQ({=s-Pj>Ni_opA-M6ab|ZJgV|NdH_b@N(l=9@RZf54g?{y!*5@_=RGN0BDZ0c?X zKF%W(#>4qy@OF4(2WwRs&;fL)>BuRWnsm@K;W8HpVip_j9wre$sxW#FRUvT4aT{EO z0gsm(l?>pBn)BQmV375I5&(qmB9JC2C7XL$G9yPDeHdXpd~wbK+Xe)04hYk`;& zyyiMa-PY1fQ8fej{itRmb!1o9Q6T_RPDO0g>MicqzJH5MNXg&atzpJWPfyPPAQ8yP z96cO?K!)7WaUcMSL^v^;22aHd$6h0_cbymU9<|=ev`H?@D&T(G`@`cFuBMLNkcf~N z(MFv&PIY6Cz56f@6r z5iS89Bn+`^X`QEasR*va2!R=xxE*zvg;1GNnGyvOWdOinb1*{)F&#FBFht-`1p*@~ literal 0 HcmV?d00001 From 7c43dc2a0454b58b950089a673d2ab26ee3b15c4 Mon Sep 17 00:00:00 2001 From: dizzyi Date: Wed, 28 Jul 2021 17:04:13 +0800 Subject: [PATCH 6/8] added 1 test photo compiled jupyter notebook file to py file --- binPicking/script/detectron/Test.py | 1 - 1 file changed, 1 deletion(-) diff --git a/binPicking/script/detectron/Test.py b/binPicking/script/detectron/Test.py index 4ec85a9..831656b 100644 --- a/binPicking/script/detectron/Test.py +++ b/binPicking/script/detectron/Test.py @@ -12,7 +12,6 @@ from detectron2.config import get_cfg from detectron2.utils.visualizer import Visualizer from detectron2.data import MetadataCatalog, DatasetCatalog -from detectron2.data.datasets import register_coco_instances # In[ ]: From 8bbbc89ed3002099fae7055a30d2f08feb7f9abd Mon Sep 17 00:00:00 2001 From: dizzyi Date: Wed, 28 Jul 2021 17:14:21 +0800 Subject: [PATCH 7/8] update hyper param description --- binPicking/script/detectron/Train.py | 3 +++ binPicking/script/detectron/Train_on_modify_COCO.ipynb | 9 ++++++++- 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/binPicking/script/detectron/Train.py b/binPicking/script/detectron/Train.py index 7def0ed..2be16c4 100644 --- a/binPicking/script/detectron/Train.py +++ b/binPicking/script/detectron/Train.py @@ -115,6 +115,9 @@ cfg.SOLVER.STEPS = (120_000,) cfg.SOLVER.WEIGHT_DECAY = 0.000_1 + +# ROI_HEADS.BATCH_SIZE_PER_IMAGE * SOLVER.IMS_PER_BATCH +# E.g., a common configuration is: 512 * 16 = 8192 # Detectron default 16 cfg.SOLVER.IMS_PER_BATCH = 32 # Detectron default 512 diff --git a/binPicking/script/detectron/Train_on_modify_COCO.ipynb b/binPicking/script/detectron/Train_on_modify_COCO.ipynb index 785ffec..4a883fc 100644 --- a/binPicking/script/detectron/Train_on_modify_COCO.ipynb +++ b/binPicking/script/detectron/Train_on_modify_COCO.ipynb @@ -117,13 +117,17 @@ "Detectron default \n", "Base Learning rate 0.001\n", "GAMMA 0.1 \n", - "STEP (30000,)\n", + "STEP (30_000,)\n", + "WEIGHT DECAY 0.000_1\n", + "MOMENTUM 0.9\n", " GAMMA : Learning rate decay factor\n", " STEPS: num of iter for learning rate decay by gamma\n", " \n", "MASK RCNN PAPER : https://arxiv.org/pdf/1703.06870.pdf\n", " Base LR 0.02\n", " decay by 10 @ 120k/160k\n", + " weight decay 0.000_1\n", + " momentum 0.9\n", " \n", " Cityscapes finetuning \n", " Base LR 0.001\n", @@ -139,7 +143,10 @@ "cfg.SOLVER.GAMMA = 0.1 \n", "cfg.SOLVER.STEPS = (120_000,)\n", "cfg.SOLVER.WEIGHT_DECAY = 0.000_1\n", + "cfg.SOLVER.MOMENTUM = 0.9\n", "\n", + "# ROI_HEADS.BATCH_SIZE_PER_IMAGE * SOLVER.IMS_PER_BATCH\n", + "# E.g., a common configuration is: 512 * 16 = 8192\n", "# Detectron default 16\n", "cfg.SOLVER.IMS_PER_BATCH = 32\n", "# Detectron default 512\n", From b4e80cb81d19c3d78a4393bb45ae1142bb0c1a36 Mon Sep 17 00:00:00 2001 From: dizzyi Date: Wed, 4 Aug 2021 07:38:33 +0800 Subject: [PATCH 8/8] added preprocessing file --- .../script/detectron/Preprocess_Data.ipynb | 202 ----------- binPicking/script/detectron/README.md | 31 -- binPicking/script/detectron/Test.py | 65 ---- binPicking/script/detectron/Train.py | 166 --------- .../detectron/Train_on_modify_COCO.ipynb | 316 ------------------ binPicking/script/detectron/name_0.png | Bin 89413 -> 0 bytes binPicking/visual/FetchPreprocessing.py | 101 ++++++ binPicking/visual/__init__.py | 0 8 files changed, 101 insertions(+), 780 deletions(-) delete mode 100644 binPicking/script/detectron/Preprocess_Data.ipynb delete mode 100644 binPicking/script/detectron/README.md delete mode 100644 binPicking/script/detectron/Test.py delete mode 100644 binPicking/script/detectron/Train.py delete mode 100644 binPicking/script/detectron/Train_on_modify_COCO.ipynb delete mode 100644 binPicking/script/detectron/name_0.png create mode 100644 binPicking/visual/FetchPreprocessing.py create mode 100644 binPicking/visual/__init__.py diff --git a/binPicking/script/detectron/Preprocess_Data.ipynb b/binPicking/script/detectron/Preprocess_Data.ipynb deleted file mode 100644 index 0754da4..0000000 --- a/binPicking/script/detectron/Preprocess_Data.ipynb +++ /dev/null @@ -1,202 +0,0 @@ -{ - "cells": [ - { - "cell_type": "code", - "execution_count": 1, - "id": "eb245494", - "metadata": {}, - "outputs": [], - "source": [ - "import json, os" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "id": "2f94b5da", - "metadata": {}, - "outputs": [], - "source": [ - "# The modified category, stripe all useless classes\n", - "# TODO: change it to real modified category\n", - "modified_categories = [\n", - " {\"supercategory\": \"vehicle\" ,\"id\": 5,\"name\": \"airplane\"},\n", - " {\"supercategory\": \"accessory\" ,\"id\": 32,\"name\": \"tie\"},\n", - " {\"supercategory\": \"sports\" ,\"id\": 37,\"name\": \"sports ball\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 44,\"name\": \"bottle\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 46,\"name\": \"wine glass\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 47,\"name\": \"cup\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 48,\"name\": \"fork\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 49,\"name\": \"knife\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 50,\"name\": \"spoon\"},\n", - " {\"supercategory\": \"kitchen\" ,\"id\": 51,\"name\": \"bowl\"},\n", - " {\"supercategory\": \"food\" ,\"id\": 52,\"name\": \"banana\"},\n", - " {\"supercategory\": \"food\" ,\"id\": 53,\"name\": \"apple\"},\n", - " {\"supercategory\": \"food\" ,\"id\": 55,\"name\": \"orange\"},\n", - " {\"supercategory\": \"food\" ,\"id\": 56,\"name\": \"broccoli\"},\n", - " {\"supercategory\": \"food\" ,\"id\": 57,\"name\": \"carrot\"},\n", - " {\"supercategory\": \"electronic\",\"id\": 74,\"name\": \"mouse\"},\n", - " {\"supercategory\": \"electronic\",\"id\": 75,\"name\": \"remote\"},\n", - " {\"supercategory\": \"indoor\" ,\"id\": 84,\"name\": \"book\"},\n", - " {\"supercategory\": \"indoor\" ,\"id\": 87,\"name\": \"scissors\"},\n", - " {\"supercategory\": \"indoor\" ,\"id\": 90,\"name\": \"toothbrush\"}\n", - "]\n", - "with open(f\"./modified_categories.json\", 'w') as f:\n", - " json.dump(modified_categories,f)" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "id": "9f9eb29c", - "metadata": {}, - "outputs": [], - "source": [ - "## data's directory\n", - "DATA_ROOT = './coco'\n", - "PREPROCESSED_DATA_ROOT = './coco'" - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "id": "2ee597db", - "metadata": {}, - "outputs": [], - "source": [ - "with open(os.path.join(DATA_ROOT,\"instances_val2017.json\")) as f:\n", - " data = json.load(f)" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "91abb0ae", - "metadata": {}, - "outputs": [], - "source": [ - "#data['annotations']" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "id": "7ada02ba", - "metadata": { - "scrolled": true - }, - "outputs": [], - "source": [ - "#data['categories']" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "a43e5d5b", - "metadata": {}, - "outputs": [], - "source": [ - "data['categories'] = modified_categories\n", - "#data['categories'] " - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "id": "1ea86c39", - "metadata": {}, - "outputs": [], - "source": [ - "cats = [cat['id'] for cat in modified_categories]" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "id": "f3f16c20", - "metadata": {}, - "outputs": [], - "source": [ - "data['annotations'] = list(filter( lambda anno: anno['category_id'] in cats ,data['annotations']))\n", - "#data['annotations']" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "id": "b945cd6a", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "5 142 True\n", - "32 253 True\n", - "37 262 True\n", - "44 1024 True\n", - "46 342 True\n", - "47 898 True\n", - "48 214 True\n", - "49 325 True\n", - "50 252 True\n", - "51 625 True\n", - "52 378 True\n", - "53 238 True\n", - "55 286 True\n", - "56 315 True\n", - "57 370 True\n", - "74 105 True\n", - "75 282 True\n", - "84 1160 True\n", - "87 35 True\n", - "90 56 True\n" - ] - } - ], - "source": [ - "## verification and count \n", - "count = {}\n", - "for anno in data['annotations']:\n", - " cat_id = anno['category_id']\n", - " if cat_id in count: count[cat_id] += 1\n", - " else: count[cat_id] = 0\n", - " \n", - "for k, v in sorted(count.items()): print(k, v, cat_id in cats)" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "id": "4e4c00f1", - "metadata": {}, - "outputs": [], - "source": [ - "with open(os.path.join(PREPROCESSED_DATA_ROOT,\"modified_train2017.json\"), 'w') as f:\n", - " json.dump(data,f)" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.8.10" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/binPicking/script/detectron/README.md b/binPicking/script/detectron/README.md deleted file mode 100644 index 9fa7d06..0000000 --- a/binPicking/script/detectron/README.md +++ /dev/null @@ -1,31 +0,0 @@ -# Train Model for visual -## 1. Download the COCO dataset -Download dataset: https://cocodataset.org/#download - -download - -- 2017 Train images[118K/18GB] -- 2017 Train/Val annotations [241MB] - -save the file in structure - - coco (DATA_ROOT) - L instances_train2017.json - L instance_val2017.json - L image/ - -## 2. Preprocess the Data -open the 'Preprocess_Data.ipynb' notebook - -update the ```DATA_ROOT``` and ```PROPRECRESS_DATA_ROOT``` - -run the notebook and preprocess the COCO data to strip all useless classes' annotation. - -## 3. Train the pretrain Model on the modified COCO data -install ```cuda, torch, detectron2``` - -Detectron2: https://detectron2.readthedocs.io/en/latest/tutorials/install.html#install-pre-built-detectron2-linux-only - -update the ```DATA_ROOT``` - -run the notebook to train the model. \ No newline at end of file diff --git a/binPicking/script/detectron/Test.py b/binPicking/script/detectron/Test.py deleted file mode 100644 index 831656b..0000000 --- a/binPicking/script/detectron/Test.py +++ /dev/null @@ -1,65 +0,0 @@ -# In[] -# import some common libraries -import numpy as np -import os, json, cv2, random, pickle -from PIL import Image -from matplotlib.pyplot import imshow -import json - -#import some common detectron2 utilities -from detectron2 import model_zoo -from detectron2.engine import DefaultPredictor -from detectron2.config import get_cfg -from detectron2.utils.visualizer import Visualizer -from detectron2.data import MetadataCatalog, DatasetCatalog - -# In[ ]: - - -## load the model and the weight -""" -MODEL_ROOT - L model_cfg.pickle - L {cfg.OUTPUT_DIR} - L model_final.pth -""" -MODEL_ROOT = './' -cfg = {} -with open('model_cfg.pickle' , 'rb') as f: - cfg = pickle.load(f) - -print(cfg.OUTPUT_DIR) - -cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth") # path to the model we just trained -cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold -predictor = DefaultPredictor(cfg) - - -# In[ ]: - - -pic = np.asarray(Image.open('name_0.png')) -imshow(pic) -#pic = pic.transpose((2,0,1)) -print(pic.shape) - - -# In[ ]: - - -outputs = predictor(pic) -# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification -print(outputs["instances"].pred_classes) -print(outputs["instances"].pred_boxes) -print(outputs['instances']) - - -# In[1]: - - -# We can use `Visualizer` to draw the predictions on the image. -v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1) -out = v.draw_instance_predictions(outputs["instances"].to("cpu")) -imshow(out.get_image()[:, :, ::-1]) -im = Image.fromarray(out) -out.save('output_name_0.jpg') diff --git a/binPicking/script/detectron/Train.py b/binPicking/script/detectron/Train.py deleted file mode 100644 index 2be16c4..0000000 --- a/binPicking/script/detectron/Train.py +++ /dev/null @@ -1,166 +0,0 @@ -#!/usr/bin/env python -# coding: utf-8 - -# In[ ]: - - -import torch, torchvision -print(torch.__version__, torch.cuda.is_available()) - - -# In[ ]: - - -from detectron2.utils.logger import setup_logger -setup_logger() - -# import some common libraries -import numpy as np -import os, json, cv2, random -from PIL import Image -from matplotlib.pyplot import imshow -import json - -#import some common detectron2 utilities -from detectron2 import model_zoo -from detectron2.engine import DefaultPredictor -from detectron2.config import get_cfg -from detectron2.utils.visualizer import Visualizer -from detectron2.data import MetadataCatalog, DatasetCatalog -from detectron2.data.datasets import register_coco_instances - -print("finish importing") - - -# # Prepare Data - -# In[ ]: - - -''' -The file structure of the dataset -coco (DATA_ROOT) - L modified_train2017.json - L modified_val2017.json - L image/ -''' -DATA_ROOT = './coco' - - -# In[ ]: - - -register_coco_instances( - "modify_coco_train", - {}, - os.path.join( DATA_ROOT, "modified_train2017.json"), - os.path.join( DATA_ROOT, "image") -) -#register_coco_instances("modify_coco_val" , {}, f"{DATA_ROOT}/jmodified_val2017.json" , f"{DATA_ROOT}/image") - - -# In[ ]: - - -with open('./modified_category.json', 'r') as f: - NUM_CLASSES = len(json.load(f)) -print(f"NUM_CLASSES = {NUM_CLASSES}") - - -# # Prepare for Training - -# In[ ]: - - -from detectron2.engine import DefaultTrainer - -cfg = get_cfg() -cfg.merge_from_file( model_zoo.get_config_file( "COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")) -cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml") -""" -Model Zoo Link: -https://github.com/facebookresearch/detectron2/blob/master/ -MODEL_ZOO.md#coco-instance-segmentation-baselines-with-mask-r-cnn -""" -cfg.DATASETS.TRAIN = ("modify_coco_train",) -cfg.DATASETS.TEST = () -# Detectron default 4 -cfg.DATALOADER.NUM_WORKERS = 4 -# Detectron default 40000 -cfg.SOLVER.MAX_ITER = 160_000 -''' -Detectron default -Base Learning rate 0.001 -GAMMA 0.1 -STEP (30000,) - GAMMA : Learning rate decay factor - STEPS: num of iter for learning rate decay by gamma - -MASK RCNN PAPER : https://arxiv.org/pdf/1703.06870.pdf - Base LR 0.02 - decay by 10 @ 120k/160k - - Cityscapes finetuning - Base LR 0.001 - decay by 10 @ 18k/24k - - update baseline - Base LR 0.001 - decay by 10 @ 120k,160k/180k - - Benefit form deeper model -''' -cfg.SOLVER.BASE_LR = 0.001 -cfg.SOLVER.GAMMA = 0.1 -cfg.SOLVER.STEPS = (120_000,) -cfg.SOLVER.WEIGHT_DECAY = 0.000_1 - - -# ROI_HEADS.BATCH_SIZE_PER_IMAGE * SOLVER.IMS_PER_BATCH -# E.g., a common configuration is: 512 * 16 = 8192 -# Detectron default 16 -cfg.SOLVER.IMS_PER_BATCH = 32 -# Detectron default 512 -cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 2048 - -# Number of classes -cfg.MODEL.ROI_HEADS.NUM_CLASSES = NUM_CLASSES - -# Confident Level -cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold - -cfg.OUTPUT_DIR = './model' -os.makedirs(cfg.OUTPUT_DIR, exist_ok=True) -#cfg.dump() - - -# In[ ]: - - -trainer = DefaultTrainer(cfg) -trainer.resume_or_load(resume=False) -trainer.train() - - -# In[ ]: -from IPython import get_ipython - -# Look at training curves in tensorboard: -get_ipython().run_line_magic('load_ext', 'tensorboard') -get_ipython().run_line_magic('tensorboard', '--logdir output') - - -# In[ ]: - - -from detectron2.modeling import build_model -from detectron2.checkpoint import DetectionCheckpointer -final_model = build_model(cfg) - -checkpointer = DetectionCheckpointer(final_model, save_dir="model") -checkpointer.save("save_final_model") - -# secondary save cfg as pickle -import pickle -with open('model_cfg.pickle' , 'wb') as f: - pickle.dump(cfg,f) diff --git a/binPicking/script/detectron/Train_on_modify_COCO.ipynb b/binPicking/script/detectron/Train_on_modify_COCO.ipynb deleted file mode 100644 index 4a883fc..0000000 --- a/binPicking/script/detectron/Train_on_modify_COCO.ipynb +++ /dev/null @@ -1,316 +0,0 @@ -{ - "cells": [ - { - "cell_type": "code", - "execution_count": null, - "source": [ - "import torch, torchvision\n", - "print(torch.__version__, torch.cuda.is_available())" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "from detectron2.utils.logger import setup_logger\n", - "setup_logger()\n", - "\n", - "# import some common libraries\n", - "import numpy as np\n", - "import os, json, cv2, random\n", - "from PIL import Image\n", - "from matplotlib.pyplot import imshow\n", - "import json\n", - "\n", - "#import some common detectron2 utilities\n", - "from detectron2 import model_zoo\n", - "from detectron2.engine import DefaultPredictor\n", - "from detectron2.config import get_cfg\n", - "from detectron2.utils.visualizer import Visualizer\n", - "from detectron2.data import MetadataCatalog, DatasetCatalog\n", - "from detectron2.data.datasets import register_coco_instances\n", - "\n", - "print(\"finish importing\")" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "markdown", - "source": [ - "# Prepare Data" - ], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "'''\n", - "The file structure of the dataset\n", - "coco (DATA_ROOT)\n", - " L modified_train2017.json\n", - " L modified_val2017.json\n", - " L image/\n", - "'''\n", - "DATA_ROOT = './coco'" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "register_coco_instances(\n", - " \"modify_coco_train\", \n", - " {}, \n", - " os.path.join( DATA_ROOT, \"modified_train2017.json\"), \n", - " os.path.join( DATA_ROOT, \"image\")\n", - ")\n", - "#register_coco_instances(\"modify_coco_val\" , {}, f\"{DATA_ROOT}/jmodified_val2017.json\" , f\"{DATA_ROOT}/image\")" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "with open('./modified_category.json', 'r') as f:\n", - " NUM_CLASSES = len(json.load(f))\n", - "print(f\"NUM_CLASSES = {NUM_CLASSES}\")" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "markdown", - "source": [ - "# Prepare for Training" - ], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "from detectron2.engine import DefaultTrainer\n", - "\n", - "cfg = get_cfg()\n", - "cfg.merge_from_file( model_zoo.get_config_file( \"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\"))\n", - "cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\")\n", - "\"\"\"\n", - "Model Zoo Link: \n", - "https://github.com/facebookresearch/detectron2/blob/master/\n", - "MODEL_ZOO.md#coco-instance-segmentation-baselines-with-mask-r-cnn\n", - "\"\"\"\n", - "cfg.DATASETS.TRAIN = (\"modify_coco_train\",)\n", - "cfg.DATASETS.TEST = ()\n", - "# Detectron default 4\n", - "cfg.DATALOADER.NUM_WORKERS = 4\n", - "# Detectron default 40000\n", - "cfg.SOLVER.MAX_ITER = 160_000\n", - "'''\n", - "Detectron default \n", - "Base Learning rate 0.001\n", - "GAMMA 0.1 \n", - "STEP (30_000,)\n", - "WEIGHT DECAY 0.000_1\n", - "MOMENTUM 0.9\n", - " GAMMA : Learning rate decay factor\n", - " STEPS: num of iter for learning rate decay by gamma\n", - " \n", - "MASK RCNN PAPER : https://arxiv.org/pdf/1703.06870.pdf\n", - " Base LR 0.02\n", - " decay by 10 @ 120k/160k\n", - " weight decay 0.000_1\n", - " momentum 0.9\n", - " \n", - " Cityscapes finetuning \n", - " Base LR 0.001\n", - " decay by 10 @ 18k/24k\n", - " \n", - " update baseline\n", - " Base LR 0.001\n", - " decay by 10 @ 120k,160k/180k\n", - " \n", - " Benefit form deeper model\n", - "''' \n", - "cfg.SOLVER.BASE_LR = 0.001 \n", - "cfg.SOLVER.GAMMA = 0.1 \n", - "cfg.SOLVER.STEPS = (120_000,)\n", - "cfg.SOLVER.WEIGHT_DECAY = 0.000_1\n", - "cfg.SOLVER.MOMENTUM = 0.9\n", - "\n", - "# ROI_HEADS.BATCH_SIZE_PER_IMAGE * SOLVER.IMS_PER_BATCH\n", - "# E.g., a common configuration is: 512 * 16 = 8192\n", - "# Detectron default 16\n", - "cfg.SOLVER.IMS_PER_BATCH = 32\n", - "# Detectron default 512\n", - "cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 2048\n", - "\n", - "# Number of classes \n", - "cfg.MODEL.ROI_HEADS.NUM_CLASSES = NUM_CLASSES \n", - "\n", - "# Confident Level\n", - "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", - "\n", - "cfg.OUTPUT_DIR = './model'\n", - "os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)\n", - "#cfg.dump()" - ], - "outputs": [], - "metadata": { - "scrolled": true - } - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "trainer = DefaultTrainer(cfg) \n", - "trainer.resume_or_load(resume=False)\n", - "trainer.train()" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "# Look at training curves in tensorboard:\n", - "%load_ext tensorboard\n", - "%tensorboard --logdir output" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "from detectron2.modeling import build_model\n", - "from detectron2.checkpoint import DetectionCheckpointer\n", - "final_model = build_model(cfg)\n", - "\n", - "checkpointer = DetectionCheckpointer(final_model, save_dir=\"model\")\n", - "checkpointer.save(\"save_final_model\") \n", - "\n", - "# secondary save cfg as pickle\n", - "import pickle\n", - "with open('model_cfg.pickle' , 'wb') as f:\n", - " pickle.dump(cfg,f)\n" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "markdown", - "source": [ - "# Check the Model on Robosuite Example " - ], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "## load the model and the weight\n", - "\"\"\"\n", - "MODEL_ROOT\n", - " L model_cfg.pickle\n", - " L {cfg.OUTPUT_DIR}\n", - " L model_final.pth\n", - "\"\"\"\n", - "MODEL_ROOT = './'\n", - "cfg = {}\n", - "with open('model_cfg.pickle' , 'rb') as f:\n", - " cfg = pickle.load(f)\n", - "\n", - "print(cfg.OUTPUT_DIR)\n", - "\n", - "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, \"model_final.pth\") # path to the model we just trained\n", - "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold\n", - "predictor = DefaultPredictor(cfg)" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "pic = np.asarray(Image.open('name_0.png'))\n", - "imshow(pic)\n", - "#pic = pic.transpose((2,0,1))\n", - "print(pic.shape)" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": null, - "source": [ - "outputs = predictor(pic)\n", - "# look at the outputs. See https://detectron2.readthedocs.io/tutorials/models.html#model-output-format for specification\n", - "print(outputs[\"instances\"].pred_classes)\n", - "print(outputs[\"instances\"].pred_boxes)\n", - "print(outputs['instances'])" - ], - "outputs": [], - "metadata": {} - }, - { - "cell_type": "code", - "execution_count": 1, - "source": [ - "# We can use `Visualizer` to draw the predictions on the image.\n", - "v = Visualizer(pic[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1)\n", - "out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n", - "imshow(out.get_image()[:, :, ::-1])\n", - "im = Image.fromarray(out)\n", - "out.save('output_name_0.jpg')" - ], - "outputs": [ - { - "output_type": "error", - "ename": "NameError", - "evalue": "name 'Visualizer' is not defined", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# We can use `Visualizer` to draw the predictions on the image.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mVisualizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpic\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mMetadataCatalog\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcfg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDATASETS\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTRAIN\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscale\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdraw_instance_predictions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"instances\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"cpu\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_image\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mim\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfromarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mNameError\u001b[0m: name 'Visualizer' is not defined" - ] - } - ], - "metadata": {} - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.8.10" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} \ No newline at end of file diff --git a/binPicking/script/detectron/name_0.png b/binPicking/script/detectron/name_0.png deleted file mode 100644 index be2aad732b69ec27e5f90292b353944b9765c89f..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 89413 zcmV)XK&`)tP)goSI})B*Cg;6_5ZD2_OgL9CQ5Rd-oo9*TUV;y%CuS zph)UfR^X36-n;MKbA0&x^Wo6{{D1a80YC;200IF9gAp-I2>@Uq5CI?p%-^m=#Kfu! zGhznCKbk54Vj_^K8JS-M0Fi%>U(L+SOrddY{|yKP=E4B`?Eojxln+RY;1sn^ZrT~bT09B=aeagonj<1KAuSWmgybUge zkYDQ8jmLqQ7A}61u>ofJ9sqdzr|<2?c6<8l_cx#3*wB?V4hGsQ*snAc6avSlK`s6h z|4*$MnxZL2MN>>_lVTK%ib=5(8wH1z4O z2t-WU8i&AgaFTM&>)`@5uMDe-f}1?+|hV zk!uYTH1D3-zjKy%hN(Zi`2(AoD~HPffk0*kX1^cqqT}zeRLFb`W#Z+-r~GLy904#g zvJ_IJ>MDmwn&By3F#`EY5tC``DG?cr;D7qclc~+0^hGRr9kW=;t|LVA{-#RiJ;XB< zDz4QqBPPIaGIaLiPwtJA)LOqOpHwC$5Qu@oU;+~g`)b0>h(K!nyTAlBa97y)e*YgP zFwwyF)#D%E*?;igJ-WXC;x_@hapT&_$r&@(T5GKUN-2~tFdLQ^3}tp5Fc=p$lQe>C zu}mq5A$=NV5DckQI)h>!=W-h~)8%pG6zmmmD zg<|zBvbgEAgxc+YXt4i3v9XfDd<%(*N?LcLh0_ReHZv%UT{is_v(;oK;!u{9uuw}F zN?}xknJMO_H3Fqu=+6%j`_iNa5|~Kwlnz*Yf|zox3)o=l>PW=79yWWJ5CO|o=B{QS z5DKNG;8M(%PD(AL5Jp5il+=TG90f=Ls97oH!;e0B@bGvThBBV2g|TM>5jcdH3iD>O zCL%Qx(I!eF5QC}#aE;}s)+O{g{-Z*8bnouhw|7tP;`HVlH*Q>KuIu%BwOW-@hG8hB z09dcr<2XKke0=lfjg%@)J3>s1nl97&-YJ1ql}eevKk!0WK7*0lpRgPT(-h^tTGYkE zCClx!Td4FvzR&?JKrmCH(iJsuM|=r!DSngHOex_9Q$Y%Xx!~k7`CDe3699#XKvc{W z>K;u}N76iWw}Gyw1T*jq3EWhKQdu_%8pfEBdOq7)7?_Kxxd<{Bm|93}SiXf(4-7?N zJY0NjVWBB34TB6YR=4hP2VLwFr=&w8W;0{oqkJ4URNS#l+8{IE>iD|Ee2%Qb_L~_I z73O!|dH?b8*?PTZ=1Y27xI5PP8YXPk>%&Wjdz-yF48%N*)5UgsezDz+yKxc|QBfu` zvw>j-bOV|iK?G5}aHSj_>&eMmdi>+vcy0aM8vur3a6hnGty*itFc8t>$HyJK8jOhn zBQTLmv#Kb;h&i{zm>C8$B`#sPX15$VFyod)fZatCxVvW{ws^LH-9Yybr7a<1gXOYy z+r>oIaV7y30MvEBO?2D|6T4pq^9-aE#*T+!FfwRbXak_CW~Sm&&!rO2!A0C8!Au-n z(X0}_0qXS5XT9`q;{|_^nsi}cLr37M7HOKZdWbeoa zP&JsUxIJSo#3?@HcH5U-0WxEn!zfyKO=TB#GwN{HOyP>Asm}_k?h*T)JU%{i8_dj? zwOTci)n=~@r+iinTLVNx&&e>bu|W)#V?SJp87MEU4&T}} zGXwpV&7=ROt^w3qnVFeg^qJYhHd)?hh#rwYQ{XxaW6Lz7K}W&nu(}I703tROn4u(; zboc}c$3@{*h&a{5R8_;&QmHkhkthIJE;A7aNN|Xe0MJwoaNVd>DltPlI6z1784;(5 z5OJX%e{}hZ$d$+pR4SLc@E7hD)H>c*1r=siQ&RzmN|_;|nyH}FIgm&QVWMCnE^fq` zYkWWF$^u|!W>ZbvvrvNh(I~{c6v}X-KT4{p0U+keW%d%Tjv5cnL~8LY232VQp@er& zNGzZMa6AU6L~Ns~G>7uUCH=G;YP|dYhqQPdOvf0=v9_AINgTy!655T!dc9Ix539Ys zy52M{6;uptxDbUlAu>uUyJm>N4B9~lo9s1W=ODVhE4oCrU z4Y$-@HKn$Amb5({)mh+puEort5|%oZJ%TFEeF2CIbs3w3XQbyhV&=ky280q5U{FDi zoS+8c!e%g0M?4fzD*P04^$o3V1Hr{iNWR~k~1FWS~15E6zZ3L)7#L;4R4=_`i z{8=&tAvY;t3L69X9b2+ni_pq_j77Xo9=67ga>NV_B9FqEhK_es6~kGxvv&YikQl^F3}QPK&7d{DWmUio zY9YoSoZIy^4h9f`87~b_9@BNV`phgM{xg6_kDd@0)!Y|V-+4#;AR=A_925*7Q_0AO zVM?l&acjh~W~$A%HEz?*xv4KHC4J;TS!~%g1_YwQ%nOdhxD1psR^vNB zAU4(vy#dt>7yTyS`e)2Z7O|*2hRny4w0>pFuhymJV2;w$D5V^x$O!9WCs>#~}th?#I5g0@mq zQzGzFY9G>zqnX9+`*QJR}>^{5&iw!5?H#?8WIGO@uPKR(`UHvWXu*%OPkhbSsh zQ-@73QQJ+R9hYSdis&bcIk@rNYZirY^Q1gYt~lEvT+8SpICWN63iM$VEHCAQvhI3@zSf z@yKB>TnZxyQ&AO%vlND(12+;3CaqDTwQe!pYZ@(=Iz9q*L~Ow_EznjnJJzFK>#O3C zTwXh9K{IA1=DVNWMb}p|x_Dw?&PfbJnADkPk7LBJGr)*NF&)ZkwK9;0uXUJeYhx2K zgNtLDK^MYRA=de(T`⪙o__30lNRWSkkz2Gcoh%r$0#Gb}oegQFSQnJcOxpL~az_ z_7y-)HP+lT<-j=u5V5o=TqOX`aI~*aDP~GN{xe07tpPCi&CIsx2QcP;iVOH^dU7GKm$6$=qr9E&yb{b2F`0~Fr5PC=;R%yiS?A@bBkJDl}BLrzs!=xyTW(FmhUxzt)~69c(qEH(om;H^jcH(oD~E-asRPF$vfVwxrq@!zORn5sdE zSfv@We~5^jJKz$5slh;ODi#_O(X3l(c7zEAX@{ywu>h(dn0ji`F>Xm0LS||}#vYai zlTFmr$N?=Ue?;kH%|rzuu1v5>8Pi0rDyh|=NBrYPFzPUZ0HVkRxW{1U1T8{RERc!y zFcxM8=0ap3I21vGned(-o@Nx0EU%dhe#D3Y!OYZ}AuUt5>=5RTB17-O2oMu$S2d_I zin}|Qsz^jhHd~r7X_0;G`tJs-fIRgP3Bc~Ug353_6rY(vY|v&#>~C=On>8`$H2aEwFs@!EL!^UxPDSdo#RWG5r4^RMtaN1B2_9q3=lP8_Kd^gA<6&O%=aLC zpa6xqENII3%FZpORao3UGoiQ|n3;QXm*!GxAenqS5%Ypg%rBU~3e52!xgy2@F+_u( z3~h~EOMH%CD#T9gI)w^kJb_NIE%wdhndNiLo-TA$iJa6zDJ4w+mI~n46tku@DF93g zLaoX1($xuBe)mp~TLOq$qSr#>+?S(OxBpf*5z~K`F&ez6G~o zXJ=>qmJo}~3tY-6e4UjxMA`|=#h|dPec&#ueeu|1E#HlJ>_nI;#YUMu1u;Nsjj0k$ zVPui_uf<#EFqzi}8mwr){;X2@NnG|9uY?IC^ zX=KVMN=l&vYLj|cpgIggWmZ!a<-#&Ghvy0>C_+3Q@9ZioQR!~awRW{y?RL8b(zbp- zzGBoL#C6%@$a&$qt4!mce1OM<0s3=+(l;@)lX$b#C#}?N8F_LEFeY|rNoJQNGZ2kIC01%+YDJ7oen~70!jTT-2xwye# zx|vHe)3qfYM6ExFdCdv;TXb7WJ-kVN%=-2c0l0hbeqcEZP-2Bu3$Jw9#YMHecaDCfsHX<@jQ;&VSKE!)1 zHF{WoT3@J)&>YNqHWeTvW=ECo3IoJ56${oXq&IGZh$3cAJ?Ta(9wdY+ z-w$-l9hF4^f#SY668fRC6rQ@190_iMzb9JCZ!hKv@1t37;!>di=8SUOW~`K|>zgfM z=lt3J8&d{ZkARcVekH5oE`QH9&X}G97Uy8RS-35_)zm?m@l> zDk0)73VEONql7La)^E1w9hMkYtMxdZbsbPA{1paF8aC4yrYPif2o|Xl4)#RUTGKid zCskDyL`@w;h7Q2+cmMv`_9(;G%_wH@C+t%7NP3reyR!W)SL*U)q3!rm@ zGor}4oe)8w6QeU#AJBBh5&3a|dcGX#;rrGGkff*@?jKB%R85)hAIE|b3O`u$);X(% zT{ztDC_K>am*|nE%XGk`*}%Dur~y#o4g^1MROY`Dsz5OXQ{H)uN$U;1m z_wOr^4m0c+v1R2s0goR)S+CcsDk99B`797HNj#4kz%YgZ3>0@dN@}V`#Y|MIYQ};V zIG{TLgBdo3e(zHOyk!Nb^TjUj+wB&>(b3`JF(|fjE{^Qmgoa}~3hU22{CJxz-rem@ZG(~z|SX$u|X5RrBopiSczwQ%~A*N8b?p4FtT_agx6fju>nWDM@B$(Hrx+_2r35=OJT^Il{Q;j0Wj~@sV^>`rH zBJAzn-d<}>MBEE)Hk*r!iyoRM%YocA|7$|T*bFxoeLFgJLE(MDPi8!nJ z_tFmM*KZJb&Dw zPKX7_^{3x!)XfGDJBL;aFv7&aAo$;5?*1%9)^m-$CTAAi8OP*?Sw00w+46ZUpTgG^ zAgGr(Bf&}f^}L!_F)W{gy2G7YH4>r`ni%fF6 z;nt9ewCKn(mM}=&)z`d_B-m7`)LFJc3G_pNye1(05;LVabSWe+B#4jFQ`^t^{9Ty* z8<`pG3P4GbJBNY}3)1R#cwlB~GJl1apk>A$X55=Fx@S=9PhoQy5?oBzh49aDWp-B< zOTmCDY~v z!Az+}G#qeLDx5!;d`IS-T&tQc^O$2y7T}90Heag-wuIP)vMw}CAPvJU(;K< zCa40$CCrdN7lDxMJ9V^gp5^uBDDk%ZS6|}9gJeC>cH7tANwX=*%RBt!Qp;>GgK?T= z&#XrjRHI!rFW9tI<|sIe{BAsAZsW{=cbkA2vCY_!m^y4rQyG6{X6p@af$?T^3Cs;) zfWRyMp{B}QD5@>G4PY)Dh%>OZsS0!HneZ_DF8K?Po%hynjTd18b!-)YDt?u=gTRzM zmc$;o$Nms2LOIhoXv(0v(et}W-v$sDu0(ipe6o~&A}V{A+G7)?QEimiDz|B2`z^xr z@{|IADLbby_nW-+NdNjb>Syo&^Ok(5CPf*K3;gVnozr=*|8WKIW&HLxU-LTlsAn*Z zr$}R*#qgmXDpI(YyZ81N3u2zUvQ2x)mnc5UFH@q@VQ>~|#nvyxDg6A*2&IWKWDqu6 z)DRM7=IZ2e@0F%5#(qQ?Bu}`W%W<|JWz(8j92%nA?qR1o+XhZ7>hcwWrdN#l;ROP5Va7;?%)Wx768LhCoYUshB`9#Gu>yt_i6H zo`cKK!fGOl*gJ2IF;f#Gq9B-6h}j}V8?`#r6GItY&UOY^{60&WW{u9WpZw+&Eq@Oq zx*>B(1D8uO$6inrrp@JxjJn$(iVAzvW?paOYMizn#6-*)%&7@Xy(4kf&)HneY8&z1 zLirO2SO;mCuh6|?FU5kX1~wyN>J88#XZIgGT-yFw0alkpPj?ER^1!)Hp{D3C*;`Yo z1Y(QwW|uDle(!45$fygV!UFm_!?$LA{4#SizKuh|AnH;>kMp&Ph!wCvS-6ngX*RW}_<8|G^wr zgo3!3dBdjn&XXsey56|tG0)-JET;kLTsacLgI7{wSD zTc~q!H5iMnOTQ_pC{4#HY4!A0sQ#UW6swS2zef$0F$6P zwlRXj)u+RE|G_BA$FgrDc)L4q?7e3)o|lGgP_`_-iy9+t@9-hM8~< zy@SUR_gVMh0Xm2v>Nz+<4nhExI+zKe5Y=oSOKd=dB!+XrEH=Uy%-7S)`hqB9cP0ro zkS2-}B!{4=YV^!7OpQzN4Xy!f4GK+#>#$<(-2#^m(|o-X4jU24IPy>tH8$>LGzyB+ zxI0b2RMd3Qs@#nxC7b47t#fWD9_m;Wf>nVS*kcWc?*t_e2I7YZQpkLG1QEUa#I`>;{{{O^&FEhYqM-v_0UnggFtpb4 z7x0$IZ+`PNBKQ=eL=9Q4rW!e)UiD_CQFt7in-$XmZeervXaHTnq`_+|EG%jG3ct1) z!diGC$lMcxu12xk3&K=;>=(c0T983tWP&sN^SAmdIz>z^x(ZyosiQsXS`UJWBtwx- zfWbUC@*^TpaVaH2HfKac6jj}EoBo|?!c#fIW`<;Vwyz}y1?Hd3M2L|cNwEg;IPnM} zUfsPq0x=aLGU5CurzpIc6|{J8-_&lJyy7(<0pa!39W5pEmQiNb5>)SH@RP1ith;^)F2L$KRh=QvuTOFBBjKU?^|_ z*~t!zS1IN4MegQ{t3JlbsDqWEoD9BEdCBaqrqY6DILr>)XU^;l)za$9%a-NUIorQm{1wm zM|AitN^`X|vmDxGNu6Y#!A&i(S)?J*;~&jV>cpL$OQKY-yZ7!hm-`PNA|D|@pv7eY zd+_k_as<^gYT&7WQ7~zT^=3Bb+9Mjt(I>tpE^l`}(tq-)ND966(Q)QwcD6)8YB*L= z(jEhNEzs2~mlx*N(WSYUHZ^94E|I@e6}KI3H^bT_BKL+TE|sW{Arlqv=p!?WCh3e$ ziE>I!R>dJ~r~46@iUT2oB>keDYo76^r59er#5GlrqNxX+UJcI$H*sd!@Ma;+erp^p zLSD%m5ne`|1kymj?(DiXNK8*go>{XlrOox2!ZZ4+sG+AckqFlF9R|uAhZ1ErJlBp= zQD^qek+bvDh~XLPfnBz@nk@@X5q!vE=`Do-T&ni|%&=NnkOzX{CvX4!-S<9NUf~O0 zrIh{s{bd%cOaE|ic@(TmRD&{sb0(F`ZsED=C=;U~!qp zfxRq4T89T7@DY(Wlapl{h;qQ7BCwTt#W})35awb)fB;D**aHs@sN_>5U{OS5-JGXK zbsZpfR~u>fc;GNPUVUx|Cm+x6?+QymL|hhdCG3Q|DJl#l0+q-svDeGQoiGf95J$>i zQw%C#{?U)${`ouachinpe4c&zAgX7uPyZs@JF<47hEc6Di6zTv#zdlle>#Ijz$?Eq z{K>yOWuihf&fq6X>zZQHkp4+$E~qI=X}d%~W;RXJYp;IY7bGx8D8FV|6bv!W#H&=2 z^NlSqM3OjUngU)#MCXOZToRm;|AV{V1c*c+reb{y%zK~0Zkx0&9qCCrQ5$z-=A9&1 zcm;2Xw4@*jrebE^f9SgF={Jq?8wYZky`};~$~|+24X})UO?%N_bWSjnmAEp+Y#Qsi z%?QZd*|AT}3m!XsPdpVsu;><@U5c4OO}yEnWc8aPa_ggQRU^9ZZA^%z@;YiS@{4RC zCFITRL`;vMzPeslk!4QZPXI7`_~d;r1^zc&Jy)WDcB*mmik)c>d?_BeQw7gt%yDi`^2Lh9C8=BVrGn{MreWi-TN>stH8GWi~8{ zIOfWwWcPxfWk;Y4DhY`Q#6J7VH>^i1Tq+gj;#o5?lYkVeUEG%iEh6sCs{AD6{sO!}gii1-+`aM{MZNyEn6I=)`1lxvf!$q?T|3NScc7k2QuAwQ~)34WGekWBo8d46g(u22m99d486=Fi>ZH(i-*+ z=(Pgj?t$F2u%!(^jWqIt9--13eCH2vCM?ZEKJq80oDx8QkG@G|rCym5Tkkd25xK>J zcBSwI(7frz?OP1YaEYRRw+{Fy2I<6KeW6PUe}goMl({qbjGq|yG8f_}E;7pj4qVuq zwy0D6asR-Q3QMaOWhZHU002My(NDVVf9BO%%f&A1_1f{zG5|2M^Yim7SFTV7@sb$5 zr*El0A4RK(fB9egO8LM3>*;WqVZFcKTPwY!!ug*TF@SkimYcX&r{hzv7TU-8&ijRGCN3S}(~Of!BUDh#6zC!Bdq!)+S6 z;b@uwpHt!wgaCc9x*_{6k)};8M{pYQs)SCMxFd{3G}VbPaQ5jMqR+^zgCbfAQ!a_O z*zv+z^cx8kh<24U?y4<6n(qvM5#)xNs)~vb(8n#tBlAK@0A#0ct<`%Dy9ZdZp8&>j zWab`*64C15@>U=PD?(E%L}F2!=RS)O#QIT#_iuzpiCHOD{sS4?U> z9#G&!+l0x|`+n=2ucdtnZ%qa>F_lt>pzN4({k<}`c(G7Yjg~#7V36e;fZ(hXONOya znQ{?Q04kU{Mr2ewzf4tI^Knm1Z*~ zm=EX_z(BykWFh_SY3UyYJMA#Z9tZ)uTeofi_|4yVBi1FoTFR}HA548LhIz@1 z27x1Dtfj8H%xcciwk8N5AQa-z5fRbCAQ|G2ht617{>y@u>D!h7K6E8XjtJz)GU9f* zau-CPNF2m@2{F>whzhf&2r@jIrlxEhd*4!8qTCGW5v3v#1s(?hoa{c^Ab(_ z(xy^YUQl3asuJ+gFQgJdk#~O0-@}wspaEib#-`J_FywSMz*IoV2qmWNyamt5BKAHP zeb(62njj)T02fn%sfv{t3x#|49#A%wyXXfDQ+0&n?9lb=*Do$EC|~F2=gadckf1hd z*An^{iatpv9A3Ok|M@rS-~L(q%Arp-2)S?Z=Pde>{8BWvrZG^jsjaTu`sO#kee=$9 zUfXJBQBmSRU`C`qvS2a5n8?d8!RdFIdwEO1w{4@m?KP^ z7eAv(^a^>AQYu2$)bBPijzR_O#c!qK>Ie!E;u5o9&4+u1@s34?DOAj)hx_O#p0f{q z4)gkWAB(6O{AY9A`sSeGp{Kr-&;OGqIUQA8?u@1hwT=lIGODM+=l{ z2^Cdw9y)I-o*@gI0wDS5moO?8^G{+0G!`m3Z07mqsR#XD?g)!Qrdq2@zl(lrZ5+q# zcI!1Z`}_MRCnrn($7>KoaykUeuXr$dK1yhP{g85r5}r%~yO;U^p{4>8GK(VBs0;7`In6~e1jDSFikZsT z2wp_#2t^b|AdyDWh>FiK1(AvnGiN~Wj)<8tc~fhShsj1+ zZ4*FTD+j(u&Xj~@MV;lrFzz$}^|A|J&DKWuyVJP!lVen^l%mp7UXg_YX6exl78O3W zU_BhVdhqA*wt5kK^oGtD&2L`CFqN|U^s{@%C#Su(*;}hb#O2!sy_9l(e!gC>yV+Q; z*WI)4?d>f+fM)lf0iviJ7V|GYU*12_kIt+IADv(5{{eHVtN)S^rzoA)qxI&>FFyF> z=kNUDr59iL<{PiZFfssSDgt6_tskA{D^CV(&BvnUU7D%5=~ihb&YMU2)@}cyA`2Zk zgza!nsFw~~kQ{(1)H?@M1%^_Fbj=3p?MG;B{1Dqj?!}+X~f$9*e?75*f)?rQA!|sL|#AZIFJomsSBC^E+ zTppBrVB*FmilZVW2r@OFRLki_dnukFnlo_!!NVWD^>degAc#euu-RqZ9n}8*zD5x| z0dOqjYrem~zw`imm#&;E#8A_;Pl~_(7lyz4Psdw(0a7}2qx=skXh==)mj}?5uoX-S z2K+3HUwrUMt-~8{yqdVx5S2D2vMXf($XnL{m}x1M7b{P`@^zJ;>WP8~2r1|3o2d^5 z?^JzMmFsLOkwI^y=haL!XTK#M5sm&4fR$2-OP`#E$d-zzmNEd*cLy?4DW&#C*aXTH ziU_k>>&%UE4p8Cf*G!&~V$yd~1?cFW=E1y{K?zX|2!9BWOfoidgXMgj`N1KGyg-6!F4RReV5IkPTBpAd4xBq6?JI+R7EMT;G?p^dVFh@9D#ZXr>M9iLz<^4Nx&Glc;-s({QpI%j zId=5>x9l&XQsG>InwdJmYCsfDF;~usF`l?q5wEpUQ5E7MvCt43nF^OWtkwoIkxZDozGFXbo8orNF{m&G+uM%1Lq}HUj$z5&nA*=Bbd6`{lySKYr_H9^C@) z`Dzyz7qQJ|wbu1|?E&tUD_2@;bKSqlg9i_8-n{AG`{tW(b`<0vf5eyn#_iKd?UOV7 zlV8YQ2BIBOEG!VnP`px_sE|od%1{=SDN&4+v)yht41?cQt@Y!NKf8AI^7GH#0Zc`# zPv`a+RIHSNxiA-lsWu{_QZrz)2n~A}W@Z*-M%PUvaH&WccD-D%G>AekZO$xpw+ZrM zvzX-uv%#M=mSzSFclxe zp(@hmq>ayDD2v`-%1jj@#kYKZ*%9nySXoDy$*Z80s^(_`B@e+oFFO%bL#1=m!5{paIEBmGdL=-V=W8@WiCcWh*LA7m?9z$A)4Fv{fGmm z=uIueSwrtK>q3_VR{tUSN&sfkqC(KS=Uu`wWX>Y$Q1K+jya2&m$NXSmuZd0AO&xTF znudd6@1X_}m%?>mt^jeVy&EG0lp@y_0gx%8vI^N1;q<`dO6HH2>YWj}8nP=q2_J7S z_Ig1jcEYr?y&)3InO3d?X0~3h{p8){%RRtCi>ODy0IpxZz7&1``QFKPU=X1aDxpxE zZjfUj=CgzgqcSREV5}JHf=$7u(yCC6Y$YNHoa`W^e!WU7z#-CUBuz zOQ}RmTq0`EDYL%8$c&bS@QW{d@X8>S$ux%PyMIJf$OY}=QbDExcwEzopc=V$=FU4K z%C&mm8#1jF-I7!)6((jbg$g_D@uC=4Ujl_I4XBPVrTu*%NNXysmtit-se_{+E_EIU z5Fx)iff$`Yu1793+9&fxT<$WSkL06%vRoE`nM(5^0v=D%Y^crUkuulq9lB=VQoPdw zfV4@aX&hBX#9Zs(OhK$g9^a%A;b}T!qC8#-?()4}*fVn?KA4-o=+58z$`Mux!y)O6t{bu&Z_TsfQ5u5q!lX#b!=6_J#2rqII+G1$?OpEI5?JXK!F)DsIx#w21CR}3E{ zX5DW`KF^(IPT+_#66;)}hO==TqJCx(84;D5l^q-tU(C;K4tq!b5f}5too4{=F3#)W zVwz3*%IG-?u1GC9sEHPLu9i_j{L?cD-Dh5oyg&W))0;PMb_n$9tFPX^efz~1Uwre; zH=i;801gfgT5JDc`x5V81uztM0x=P&xH1r!{b#B~15;Ue1`75HqvFHq;1Lc180_2M zdgIp38@3pi1VBVS`Q+1(F4Yh?it}$=aO*Gxt(UcfsyZhRgOPAn-fN|@?cK{6!r_Fx z&Dbt%r6R|!Bik?nz+`c52!x77?=OcEMeF~39XJ&Z07`^f2YRY8C&hn(!pY-TjxnLh zBGDqxmtl*E;KPVAvT?{X*VlAL4|Br}<&2{g zNi7&X{Ip*hc{3>k<@H&G{c#0Z5%Nt2wN<2me^|&IUl(=)(-|*l>wApNyBK3u@ zKm5Z#ymsxHhKpc^?!MK?Qyw!X+`{u^ykktoWqQc zyxy}C?zi>Z?b>~bg1qORyJITf{?==?)};dH^1=bdEEX}*D5NE2y>n^?;%@R0@CZm! z#LWCmG<0PE1K=p2fUms23`WVEhM*&mYT8^yoYOHg7U{+UT>1?Lh>icP{_e`9sP=>f zQU${w0ol9{cb(ZMOl+EM2p-Nltr=x=Ff*v46!ys7^3eJWdX0jcfT=W1&FPE>^%X;K zjs8bKPeOc!WouCqTKYiGaSB;s1~E&U;(jbPk%^7ehk>Xn zX*Q@6A#&%^Wys?JPj!7>?pLc-KY8xWH{U!uI(qQn!7vQhuU}uU*IfseYH)OP^uY%o zIBJ^qZ~U7b|8W21=Qij2_uqQuGVNEKwkni~wbyr$#PPA7b{+kCwK%Dv;qWmpLb zQJbd8xit`d{K;p2EVhejoCyF+l50s- z5tR}%ES&HoKoq7Ug2Hhq;&7tYWWu~P9)?Uit7_rm2iij+VSqqc7Z;&Y1j?Doq~I+C zv|82GY8Bt8a}kk6$F{>ZCXVSQeZde0$E?Yeylns{&kS%4)8v&Qo~FrJ(I-!yeE!Y; z+H5wKXZO}xFD@?LC8oV^|C>i=>-GBgzWwU=|L}W9D;!?BRBPQTOWphVyYJon%F9pA zp9qOujKg~4nK)JTRs?eUVH9{atkDhG&7E5iiD$@N66Dy*WK zM0gg{fw@*H95t2EALJwHW1e;JjE`cDXq0htk!CI zD0eAkv)TBl@y{zjXZ=el(=0q zaOu`#;L*;o^Xs2)F5h_a+-@4cfn-Tm;r_a06syF5ia`_V)MXZN`H ztUfR1G*j`~=A;Y!Y!@c)o`#O&BrPUxn3yqHsaClR%np*(^4`r8lz}D)X2vmgLsJRN z%*xVy_-LT0dZr0g%ZlmJ{@xx4N@~*tBQC7kWQ;R#oXak&!INS=gs47Mvb)T=h!c*~ zUHpteew?c({?!qVnu+tJ66}{DVR7PZ7z}J)V>l}S5GjWU`E$6Nsu>k4G*pI~dhJI% z0{2rq6BIc`>7(!aJOnpCNP@^yz93K&ZB6T7MHPi98^}wHk)OD}pTYB{A8BnOCb-DZ$52(=<)f#~**Z z*=%a9`}_Nw&Bo7+=^o(8lPCUW%K5c@RiF8Dan#aWQ0Q3y{{8!^`tWQzYu-m18}I&M znx=2Rcq7w)Im0UeyIEo&WV?qCNC2qT0zY#E8YMh!8kO02C|=Z{1hP~BN;4Sn8s__n zNbIEShl^N@VfE||!7A_0w-r=|N~IbX(*`3dg^HT^a0W6W74-uFy`P_eN(87`&v#`E zT1#cmL#gG$IYUM2DOXbkMp1|KT|LtT`(H>96O3d1W3}>;EFvR-l4gSunQ1bQESH4I9jeXVl}`WimF+3X=3q(9>ipj;j2t!MrPdoY{WaH{ca72ELn&_ zVXp!1GZqzUcKy=+mA$jmsR8KiFaD?NLx4In0EdT%{pTJ!cjw@z5}lr&`lYtp?a|SZ zh`3^y+5Z0iZ~fM9eeZkUOFrZ&4rd9vy444;zrX+5Yp=ce=9{W|czC${Z1)mgw-=S? z#!(7vARlQU6hxm9kxzfLpgN3+jM)S^a~Wi6)1*+9Ode1ez(oaXGSD7MxxhUA3=t3~ zRx&f6w53oPchgYIDh75FfxQilnBlovSer~ucwlCVS+UGDNeozw_hL15qTIJ$40Fv3 z$jC&E7`0L%iX(UCXmu(5vmpxMvrs$F1d#PdH}w*F*MN$;Dzy&GMGb&*7#lOmie>oy zJ1<4v>!RZ48}ycAUm}8AdZ)>f*Ga(THMhhBB9mko^8dVzG=>NeYaY!BFUXFtgB0{}#HadC0u#tjj9=bd*P zuk*710qEc2o3PnznAwkeIXOA`_~VaPt5tUh{eH}>7p|)6Z@+RolA{chXvU3#sEEnj zc|z2DyllE7{QmBcdPrnoADDoU5zo9TufMw4T7pj#_y>O{#$ngM$Od<>%+;o{RNAH=B)1Ei<2=pVQN?d-v|0pP#$D{{HX({`bE3J%8tB zvw8C5$&DK~_V)H%XMXXEUzpkT>({4gdh+Cn6RE4!%Ihq)+pT|N*S(JZfAggqi5@eJ zFXqRP5s|^G8K@RAfJHiG>}P)&piuEzFTjkNS~adrS_D49H5zMt^CRuUIUm{iZ_V@z zXfY|o(oTh$ZLl)#rrpIjtOjY;V9x&VwqH^P`E%{I;s4^{0UO3i7JGKq%qH63=yNNKu5i4$SPRz**PC zf2L3)CJhH%3~;+0F$em3s)cd_0Q*NII6BWpSVx~kTnfk-7R{2F0d7#ROk`AooQ_&W zVrDNrV~}C+G%QoC)sb`$I(_nwhXgLfr>Cc?_>$K&O}pLh{QP`> zfB&ETvw!9?eQsU1IO#Z9sim$;kn!P7tb;)ny(v|*1z%6jpwgl@-f>{Q>sRU!VrUz62Md}W_2+;x9+dVxv)A( z%^|3j3)jMx>dG^YCan>*1kWKo9xXkc2?~oJBUsHq-uxZjzzo_{4b4bZWm2wHpe9OY z%vKm4sGMKOZnR-lh80aFT!?^ray}9fGD+GO9|$e=jIvMbGG;Ihm8aH6;2=b}KL zTA7O#EFJ*If$aPhhKXzCT9ueo)Jq$gGVYS_M}gldf1KMkJpeHEzELmz&>Wk}R40CL zyz5QCmK2TafM4wyK{qpA3_-ynoc-lG;3Bu%?U?!f_uoG|JM$8{v$HeTfs2a^Guv*r z$H&Ja;!JCe*~V~*?>8wTFWx+S?S(6~^2v7Bt!A=ACgMC^q0GcbT~HDa);%zpifB5& zn6<~ug)asdjfe-$GJnI)B7m8=6f=o*k7+4|3Y&;m)*+Hk^Vgb^nN6zWxH}tlJ=7*9 zq9E%9Akw5YYYjHGI!vW-tqhCUXQD{L9qa-34A}~y@hCm=O zCU~#(YtJ9T?EcC4;AC8i^)r%vadF{_;VcmM(%N683_ddG88vVz?-%iLLyoMTaU-2# zV&*qqIDkP-SJ9FY6 z!c=7(;Z07_#N_@rsrux$q{(hntF8>C73A8rM0Pv)!udP7Ynkh&PhpE2%{*h?W+Ln+MB zOp1AZI~W5}bh19S~KEO+{k1C3KdM`87oybSgkx%UR>3a<&mhPY)y29E=*4= zw4)r(NE|>!C*;7- ztPjj%oqPcGzV>w z7hM4A2ci-A89qMDP$V`thJ2-)TCfosxy|148tpz=`Q^J`Zd@(^yhzYrUkLR7onM_$ z`jzjbtZe&2O&}s^%>wrl8${i@K*cD}_KVITHD}WiV?8{s5DqwM-ZxB$Bl-}yBJCs@ zV3kWPRn$aKn5lq`VvJ5oKMW#f);M}-4Gb=9D8&q9UV_7sej<>Qqf|?QLd;ln9A&7r zRstC21$u4~)B?SFPZL$^hJ+UA%eg-f2aKmSwn>IUZg4$7O7<#zomabc`pT0O!7gaW zqRHsR$ zSt(izRZmLvAr~p<&=~$smRgcQv_-WocjbyIdGd-XS8a`S-$FqD14=)VBhtk*p12I_>nwpwwDOL;B!tNq0cf`o`#G9E+pfRN^Y(J1&;DMlHd1ni1 zDa<^zsSs0doX=X4fKooiPz|IsM|;7-QE*4F7|0NwTbq`lz?}fo*mP*(D9*y$c!6D1 z`+4I5V8Mt{Lbi_PfrvrGRGJi~x+-oQU1(DO4CHEM+0W1g*hI87=`%&<(3J>4EnEt> zNzAhTr*fHR?fXarqrUcW&mMGld_hRwiJ7ynK2RiO{>}%I^~PpV1{o{ZJZ?c(s+1b{ z#R-yJ<>4b^XuiW-CD(_-#BMJgsxBr)|H@Yfo}ahY&s0YEh@nzSPq#f&H}UzG$H(jS z8o=`}ym;Jn*{RG5+ak{>J6}g=K|r_zuaLYEOLYAiGu}j z3T-9N7S>XRTE?k`P|h@s)94tOKlqpn=YbgU2pQ7Pg z=`GQ-#p;8&5T?d+91(r9F@aDVD`xqmPrO56F4fcM?g?&OTHQN->bS)}?Td)E+wIq0 zdbO5PYyH}buYT>NS4HIi{9pa=kc!c>$i38}3w>w$dr>M8z5CAF|EvGae+k2fAHMVM z&wuh~zx~@kdh_4@;?p<3a`Vs)L>&l0YzHbp`d;nvRtc04h9>b#z<=&TX?vhWo~NwC zFa+o*#Ac=tWM05kE?QIE^towC^IC)??WiFRq%@+K&R&P2Vnsz%#?cC6Z#{Gn>*Yza zGXp@VREG@X@^-rd5H;22OfylG4oC8JNhL)XK+M@`U_kf@|26p-j3{|ai|GR-MMO;$ zKBlfU(WI^x=GfnFFsRNXDo08MK1LqN+*@$E0d8FZJO@w;`$>KTWXu+@%KVt@95i65 z#4wHeDWdR03-%L42J(u)NPb(ESucW;Kds#R57RXLum89I$8o#eKR7hA^Yb$R|LDK_ zyDs&=8eB+}5h;jDDW$g7M2%xeGvyp3gntjtg0k|_ zjgoOKtkqPB5iy7bF+@z#LhyX^O?5Xmj~O#OoneE;?&mqdf$BS|h=O1iX-%XSDy4X# zfdOpX?`Opn!9fGuk&Z^Q2>^7c7b% zkmNBz4S`o-(GS0kf#4E`~*|N6iB-~ZzMpMCfD|I%x(f5(6M^Pm0r^VB|)-T&oq$kPJPf91v9 zZoAuU|Lq_Ciy!~+kId});qagS-d7_Ih?7zbEkbRVN?&*}^SHG8eP1PCc(N==w;(67 zOEE^@c*az%d4nJ22(`#0FoBeSxlDvBybHutidZx%@F^lOs2K#!43NT92$f6-t;tl% zYA7KvgibR#yTUm}kkrg+^_b&A%v_0C4MaSYQnJ*~lO=uA&C|3F>H2;Xn|FaZ1L?dC z6SLFpQ8(x-$mzeKJdfiU1z4?CZ~f?x`fps99FBGzkU|{SiRU4}8yuMcZ;0|nzOZ`1 zX1htIP`1`$Vw8(3m$F(7wN?YEQ!2$92_n;MMtN+KA^Sg)xGBXm@sX%kPuth5-jC6i z1Rn{*fV9a(7&@>OvSc}p$ruLJ>etT+HylsTL`>clF%+u3>cR))_YpmD^n8|=p(A!{ zO{BG$lzJgePG<00KBZzz z#D%*iBO(Q6=Xa9W$+|883?C&OVVy@nCDtihVJX}OA;%_Xko@n8faX^z#VhJ+t&bkv zKRCGb^Pj%?){p+^^GvAD?DG8TB@aEqX6C)Uy^lV4_jiB)FTL`OZ~ou@_J8efxP575 z;|M^E)W~?CQWu^to`<|!J-X6Ciw4pXOj3-Dhy-HqfC=+rfE6ya)YhhG%k#u@WwbaB z!dWGc9euU)&|wON~5gV|8IR1QS%Bi0<-sUHmB_R}+aW;srfprz2va3N?YD)U zGbB!stNHjXQ_o`*O3LWpP3jk~Wv=(i)oZ)$?%usm%q99D4OSrLf)WUco-^?Cy8U2g8`}_HU_C#A%yw> zQxg-lp?LkMn*gMNBcP7E5jjBuUZ@y>BZ*llR0yWcBsFObVZ@ytpmLZeG4k{@N+5DSK zNzEpaNwjJ+$(%RE)n-CEb<8bAy1h9h?QY21z#)nj->GgFOIU%JNcTnQr$6nh+wUL# z;13J){d=GIWoxaTUvpQ$GkyFlk?LF!GcN<;cYgNfXPW}GtvkVjUJOT)WTBE4K+jmj9R`*Y|ymKN{Ph>}!?;j@y#gV5q1qz5<8WR)u zjGA*b!Pt~SL4EqEkLpKKGm6g`m4FKl`BI5YwL1Nr1aM~0HOtQuUgStp3XJZCEm^T{ zbfcUh$~?tGS|5=Z)34lF5``$oUUe~dt;;~)f0q+8W;uxJs!-)_mRLkGDXW4a^{ACN)-3FSho%ZVFQs z5|DAJ#Dy}Y8@1AEQi{hvKr<+eNiD>w0)v{2KB}=NRlDB;90G;#M;|Q!pT0$&2z1dm zu}CrGFvTtw9#3$>b+hmls!-&BMpae#(H_nnJ2CHyzw#I+iKwV@3a|z9$^+`NnmYw# z-SZ|fjQ)7!9dJsSY@vav7OpW$JJc;@-G9?TRc>GO2U|9>d2NvsNJzzz9({QjjF_8< zJ6HDJ{$w8b{7YWX2yTCQv*PEA-U8L-6q9F|pY8;_2Q6+qtDyU+MiH5PLZ{mR_yM|o z`dbf($xnM;>Og#~o2Pl!Tfq5FA25ZCc`Att(A-wI3HLsmuxX(zY~+^SR885!f~Foq zO(D4&UMt6`a12tQG79t5w6Y9^dypF23#)#ZOb$EpCKnQ7i}T-1M5PI-z{g7i#1@&M z9v}9gVnLi}jw4Zl^bTqQv9kx(#F|7%u;k5Mw-yqiuGzUqvGV#9Vz-Wxdi9AviyDb? zILp`(ieeuaX=HJH#G-kW4Fg9hj2{|lgnlTDMNO+irT8+9Sz2a&7`uvn(LlIA)yqZp znU|{iz=);Ve4an`AD*FxE-o(mu>-&RjT=Aw@YLTtwbpTr8vA%=13_n5_wl&CMU?G@ z8$!0sKnzD3?wnF z2qMm?k}0TCCf~wH8Ic;Rnke}(<%JkvKf%QA4*uGs|Rg@HKS=JabVVkBRyiNYOrQ{g;aO7jyQ{ zc!U0)r7eHvBR^#fOc&z|fQjUuq*ywD_mQU}7BRjnun#VW!ra z7N_)>%ph4^ZbU|vnF`4$#uEm=Tfm&iCaSB0Jz!C-V0LzXF(yR#+Tgpr^Ds?r~)qG$*C6&`dEiaHlM5X@K zo={-)B%domP#N)ii()Nxhf~NnsY~Ayw=)0`@e>dunS9MHtF7HGn3?$jguczhGa-C> ze$mb^3W^^-Sql@FVfDrJ`oI3mPZ0B;T>*NZF%fx}<7%~XRMP`Tq~Y%`%05gY@<`zN z(cZ9k8TiyMesJw8h~Y>EA4q0qYW+xtP8@g`=T-x7PnhX3%Dk5h?3}dliiv0#yyx1X&MawQXp?D_COPGR7&aIPkkXl1LM>pJh_$AZ&ou;UDXnsA z5FS>uq`@bk)KV6K9&8$=jcs6F)si*V8P#QGyO((wY2m1tw^)mq2Syn}avWD+26!8? z+jIv=>s8sT>uRVm#0owtiPE(M$XEXc3#P6BCMqqRlKUgGs8q^Gn_{phyw8sdmAKWo z6ksuEAmeTVIq@G0Vlx6rR2x4=63%bW%&TDwQz<3Y)L_4?WdGNB^+LBU{XP73iC5KU z3OxI3`AcsFCL&KPwAPpanMXB#^vV4@*Dpn3f~lGMxe{FbOwGKY%YC*$`y!DMm*|GK z?gHSzl5(B>92-lD$N*I4>K!8RvQy_1Vq3#ZGMETMAy}IPUD81F(2sy+fG6$ATG&G` z>9e6^@Xa*T5+lRJLq2nbq9a@k#m9#^can415?U>W6Pqf8sSp*;@w7ew4hH4jB!wsn z8DRigld5g1H5zAjqmJX`WeDaYvRrxtCTI*3A^;>J+Pr7bDQ$(EobENHsr9N-RDH>1dWMG;O=NUdLPdN#9X&FZt;HP`o z`P6o7e$GvhNuejsv_~V7qY-B#%?I)Ow_EOz#h?UZkTc%eX%VZWLOiuLv^vxhlW*AA zJM+DmY}`%T?ZhyKkF`*<7JO0s=g-Z_O!+xbF}#FKJ(AIYOrR(`$nr=cKg0(p0I!F- zUe!|B|Eq;dp<1a_DwVl_tDpPLRHziHt5OE81DC?pCsf5b3{kx62Rm^z2ZZ_)w_>FDcPl>xXwAl|s*5zvRF9?9r)LfDy+Y-FtHW{dey*ktUKy(^ArW zrefNBVST=p&3@awWHD%@2o60W?Q&#pSFJD=U>ao_CuvQlm|s1U>Uqf+7VBT(xh6A9 zDkds=(A~HP5|+N53foYep%iwJ8wvfDOVltEQDF_j$i&Pf8wnf{6=xngQ3?a$?(y!29~|F39s&IE zJC8j1`k(&Rb2pC$BKUbu3}Qwp==p@N-aG`3{~hGsUw#nA@;6I@s1NsCiu;$5p)UH} z2@ui84^IB@?GO9qK7DvnnC?Eg_=kV=?o*dlWuo7B@mi!lBDQvl$$G}aeQx29_4M_9 zV>(})aK$Y)p@3@Tsrg77gp=^983F-fCL#qO_SuqTMoh&xSa_R?K{z6KOlp;lnb=^_ za1eO76!PG&Em=Re z3kI0C+GgD#%zCGyk(#PSK(E%~lnKjrc^CcM8bVYmc!_7}xd)&4z)On8%T9@sxiN^D zOid@58ic^!a7NaLI)di>-cf9LL5 zYwaf=KlzVdyLt0yFY*Ky1~^yG9BxhlUzHFX`CR#nh^n+t- zpKD}pSi{5u)|yJwT0K5j)tD35W84Upy>pf#jo~dt$g{~jvKLSjFAC+<^MLq1g=n#d z2$zw6p~66TYU^SSFT}NgSRgi5-zO1M@!b?7Z<(CqjQHy8pGhZ>b74FQL!Um`?(OX@ zt*`6az0>x$|LF05@w-<+nctoR2Dn{V3e%tJwTyuMwFEz3HW1}kzLrjai0&P4^O5j_ z_wW7q7Z2x{EAt8d!)~0^)_zP@m?JlMAt1!CJbl^chV(o&Ex;rJ6b+E^1T<7GeiW%t zWmZk*BNb*X=CfMWKu`h|Fte#LgT>k+D4oQ1lTsD0z^iq$ z4`6JnXqy!et^tJJi{libM-_ffoRx$~msh$vqt1%Cd+e|xxL z0-Bi-)~jJP3|>YbVX`lKM(BzD!i-Q7LY}Ar!N+q&bZR+i-Va|0GFt%}Hmg+}o`(`; z`f;{1EEqjAb*j4~4)-VwBc_9cgZ@kZ!CCh0b}J%v|0^;!p0>4QO?VivjQ19@`|E4M zmq|YWpKbUnFV!W_%zDYgGftrA)INE5^0hlxM5GT+eCDgqwQLUdnIqqiSatRGnczF4 zwHnE@+1!8u5Y!u%TukR$!=|HdGcgq~$T=8y^QgsCd~yVnp|E6AU?^7cs3Ngvh7N9j z3gEL72{sgJf~M9cT@8f@b=co;O~&zjwJPgX9ST{ujRuf0vw=#r;)N~_>Cl(Jh_X2X zAlT;c^7_){P8Rv_m!JOh9iM;#1T$I-SBF_SW)O~ql5(+Rf;>S3&pQUOYBBoSn}8ki z`(Q60SRBWXXdTwW-kww06xC4bTUcVqv-$}fj#J}YS0gpirXc(1-tja|^mKEHi-?bN zTkRhR@9Iu!v0Aw>C7yzsJ~Q<5*LMid5S0BRdwYAYzyA7C=)oR&e+6e^bTO#oHg`?i+=%%_D z0I}XGK*-H7=SbcD!;Il`RWk&0zSBq*D!u!j3ZqA0p1z|(rK}3bUV8qj{NXPE>~_2T z{rx2s^E z^5e_D4#RNg&Ykak=R1!cJ+j3)8cQwu?U!%eJ-+yZw?EnKc0FUj%x7n3tJUhc>ql|) zwYOXJv;YyTfX(YEy)l)+R6u_IWCqP<;Sj<}x}9KR;?=625}N3$5W59&%F&HFQLVM} zqr_$qh}*5GV>2?9px{fX!>V{qrVnvr0~LZYfHgr>rxr6F>-OyY{OqE@3Y(Ao;M_Ro z5;cAvYC5s%196ea+-vz_-nU(S?qxF+3V@V_0(kqUKXoTH#b93;ti>sr5%a3oqWg&i zX*1M_oown8fn1bpC>jA-+|z9_@jaD?RE+`#2^KQv9?a3!-}rMc zeSH7)U;XIgaU8w+z$N&(E1S040%+sT)Tl;XMe$s+88LmH$@xM7{lYi+cP)SEmCQus zl|R4aZzrvNlPYk$o^YHoGM}Pdzz20u`wWz=Qt*?Fc_SNTZ9KpS( zspnEsJ0l3TOx}?}W?e)OXsz468y zue|cg&wlo^FZ=8l>c*WrcigMI_S$RyS%2H_zIF$|b2sh~8I%3?H*Wsm-G_lE&4}oC zUcFOFC20X_;p25I)xqgVch3g5j=>KZ{CF!LsBMBYq^3SPmMo63KsCFVRH|AbiYDzS zm?Mq}6iz{A!yPrzEW)?g@Yd2Hdz9znd%;9d0gCwuDKavfL<#-G8UcJ*Mmad@p1ZmQ!rZ7;Uxql}^e{^Reyw%d&`<=~zcNz~3I z0dK$m*}ES<=sNJVuYK+G^mH7@|Iu&1j7}FXFrRq5shF|)@cuVH+SLK4$4^X}7+M^xEJSQZlg~jWdg0Ek z%hxZDIeqE+?dNt{oeikQ1ZrYL^|O0V%!&bzFL>B&cH2osL}WXTyKx++iHx|Gs76lP zu^xD{DvYSOY$D^-c4M1FSE~x*X_DjPlgCevuU4cwsdj7M!+rz`d2M->6^wCEE zUVriS;iXH#xlwRnNUq>7{`TcO%CTS5Qn8oHW(j{RtC_RNB%Qazf~32t|6AT{{~x=Q zRYd65^>68ay1Mou9&;@hdJULs)wDfp;{}7Mkf<>+75ITnB27hBg?7&D7og;M7OH^a zCm$Vu_}Pg_yMA~8FTU_ZDOVpt?S#x~z24ufx4WI@Zr!@`yxWSxTnkT4RPCL2ej5GG zhV$JhqNY-6F|a4nO;Kv8r6PtZ6wkh!19bLU4`>cgErn|xM2$!lwqCE+s}+~Zj6#6n z^5OpeW(C!$wcT##wb0%R?RhrOU_=E76Gci9Ir_tUVgM?oP!oZl3XyVU25+KptIbYo za}km4I0gOcy>ea?Ot^mea68J+-uvLu<0r#rpX;iNIF~r6GbKt5&=^B&Fni@|w_bhm zHkkjH|K;BW;NpAb_SZ^{3Y4Dz#=;3$_ahN~GJwOw!w)}x~U2<#Fl3g8hH^gGtoRKJ^fTLZdrZhgiS=nS=bO)4qM`)5SdE$3I#8& zm=mUcae}I-$mY_uCwD(U5PvE{Y-DNzjN=3VrmI5ZIM&U&4mBrx1En%bskOE?-GB0A zJIT@B282m;D8&qFHnm2jOuF4QxqJT+$VBw=np_b}IeG^&LvbR|kKvY0vs$*-VtWM8PFCEq0ZaX#; zsjbhOYsVnsZGRXIz793#<& z;rR5d2Q&!O12ZwTPwziDJ>SljPviWl;0}q1>CW||Pan*M6@T!f_pcr8#b2`iFDLLU zj3^9-pFZF_cka0UIAcMCcRzZNb~GrQg-1b}Ya;RqIu7DpKJ_V|U_hR6oA(^a`ll>3 z=4S$sDnM~~M5-?geb`N2dFDpasdy1U+$6w&O|56}7O#7*i9qCL&F) z-@1M4&h1PffTL|PcInd(-=B8dNwgN4nljgfe(-UT>iKt70yUK;BGVLS5~>@50#Gh5tqX3;%4{Dlg@CCUV( zEu2eF0dx)1J~^EdQT15{Sj0{5efZh)*AKg&S*+cAfr!Y|i0Sm~EErRw7r*}XusAWE$cS+7{(Z!VxNxyP zDqTfZdy=;r zWRr+qT%37f&*9`{!rp zAAkH2wihw#8OgX>FCl`V*IxjDXO!oyduHeSBIp`GT&rch@aH!5^Se&BuU-1p@BiXI{+(|CpVQ~hU*k{xvS%;wWe?Rk&iL4uYQQu9 z(6a$gS&yeb@uPP?{mVDr@RGB#f4H~a-;Ebq9U*d)$;<-{W1tX-iOry31H|mjcOOe@ zyX{3AC&13O;iJdaJTD8teEtkWRY5ei$>L~Broy}JmP1zY_S|J+t)7LzIJr+C{WPiUuf~?v#*_UFK1fd=H&;!;IFiJ^PGTu_LXBG zp8YiddgYA|uHQMk@%#ln^DQ&W={cC02wLEuFfz}$faH6geU6VldjR0xm7_~vd->~~8*pCMD_iI}Rbi&pFqmx}M-mdNfepq? z8EjEuYN}iq+J<2`J3FbXqOFNEKVD}N1*o+))CxePc}h;i&$k1@gk+xqk%|zLpS+1^ zD2=sVuU7z6RYVF=i?g`rouEW(oqQpGr-X{|1wRja zM>u&j|NaZWf&q9k%_U6d4?cMIE%-96S^gU>Zu+T!Lr1~-g@gT zcNw5?7hdA(DP)ZU9m3c@6$W@8~tc!a7dqT@78Z4yx?Zi@^Q`VO?4qKTO3(v3SOczeJ?9Mg%UUN4BL7r3|GG zr4CtG@@u`?MQ}k6PcELmp{20I&zm9u?%cUEogXi;&%%O!4wrwyYhIeq3F@;hfPgF4 z@2&Sv%8KiHGP*cFJ3l#p^pkt^i--FkKE5=6)qj8hV19PDZ$EtZ<5Fw!psUoce)WZ` zRSs<0wCy3cLu!XKUZU|bjs8RR6G_$d?T^o=X)1*)^YLy=PrK%^wN>WmRJD}aDAG5> zYW?WZBM_viCzuNH{rmTvF!5?4H7lOdH)|?uu&55T647d?O*D^hnzNb2)I>$BRsu*= zm#5({7m5^l0N~;8mtK8Q1hu%q1S-E~_zMux_2(X5zJAXXT#G2wtgh|x?#d?*uYgue z{LbBjuU$L)`pq-_g4^xP(()E}x- znwg(*<`w(3)C$8pcSkB*j0>|*=*lk$fp``LIlH1t%%6Pv>E)xN&W3qwX_l>0VYFl2G7O~&j7#wXfmp5WIPL8ju@gbE zB!a?4_KnR;nVG6F&z+zCaufu_v@=&BigqCXtKV)0FGLFuK#RQuGu2vwR=gqxSglsW zuqrvC&@2bDJ|hOd>UHb(t#)zp%Qa}t!UqQcri&-Ysj9zjusZ`4&@VC9Pyfqj$Df^^ zPjy4LZewcv@Sc3~=-R)1>o)PqdnR`~efR#M?S19;!6PHvvriTTX#Z+^+K8mtdI}B_ zU?1Oogg8(j?(v83d~)aJwYW3G{?X<0<0ocX-1-nRnHXR;HLFuIRXp0-rNe{WZZ~eX zAi5YM0O(Tl3>V~~uuhGLKltE-%fIoPF!9xmfh$CN2m8D21u>h;V$5JPV(gmi&0#?G zFhF}4VBQ60$+ZKZ5^*p$l{yTEwNB$WZZ7~_jBO!vM$Fn;6LFX2Lz>hC==FncQVD@FJGU-^3RI*)6wwfB4FdKOzn6oE8j~-p@j!w7N z-@Es`8Xs(GsrBS^T&-3IhpSKSpKr$_W;#3FJ1U>OeBCU4?=+TDl8 z_a2?zx_;!hck`<+zxU%Ga2*U*)NP_i0;Zysf%hJbCp%(h6#?OV_r>F!o<>2!rGj9l zPo6wxZ)+jIyl50mbnDjbU%dC8u}@MVF)b7)C{itE>L`V|1Q6-MXE9UP%h()zbMf+E z_9E(GShZ>LbMLzJ@5Ywd25n8&t5sn(6DBl4hcJ}RL`^>Y@B=d|(L(J_VUY{lZUbRE zmUF`&OCMn&hk-D75KJ8-EoH($g`=5}iIgrbF2?B~N@BA3_Di<=i~nu==%Z3ZSGFI0e>=ZndGTkzEGRBFzfURqmxA$UzE5yLv{IeSSk$)Liw{1T&Zf3motv)D z&PGvu`;D8=-?{RQS8n{n|NiYCz4hrd$%Aq@UF^Mn>(Pxh-?J^yq3OH`9Wg%h`smXK zaU<5>KtK4|2Y>a(Z}_kGE?wR`I6QguC`JlVv_y#~bZ z(9>a}Uc$<|);6nwfbDjxa;74M`E0>f!7JvA6G1UGzcuMxJ`=K)A0YbhgAcFYyrrsz zS=|S~N-DsU$4~nD2Lzfc`=fC_wbnhr(qX6JBpS!<_WWE$O@+B=Yu<4(tk)@D&yII~ zO2oc|;*_Vr&&)kMC6Q14W%29J!RTkcm}aLm6B}O?3Owrqn%HW8J#ME`%Q&`GZMDdB za>B;Y?b&Yg^_QO8Z7=TK*FXG|56&(`6bxM5=o>fXjiG*}9S=`F`swqpynXRGR+(sd z=f3PGXBS9JYu2X;;_idvVOV*n2BYU+dHuZ~e6I-%6n!(~Cp3YFVg2@}=h9kXTCX<= zex6c)f6XXJm55BWa500UAg?gtNbeE|hnEib_Ya(hF}QLVGn*R03Ks|8o()(Ian;Ny z6M0!gKnqnZOkz+GQ7MJjn@uV8soOO8WkrRUN@1HTWtJOHL=crU!*Im4r27sI4kXlu zLe!ty`B^OT_HvlkGb=~r9VA1otJT0<_V@Sq_V%b3Ied90j28d@x%d1s=<_T1m0xy+ z&m*IF)@M2S#lOcViB3n}ZjH+dFi>eyh9cG{8OsUUXQ!vfkDr`fKH8j~ZpX2$i@tGv z`ughPt2kRd`Gk+Iyn3+t&%XKEYtO&%+U?s|+WyaZfRnSW%e`S90pv>X+5N{zw0m&n z+QGG3!+JHW>S|qwbuFt}R>QEj;-B0bCo!&dvs$lKt4AlN0DaloE;C}P-WvroefaR< zq8lYWDWM0ne(>Y+(70U_r!<*qwTeb;4Xe5!-3aQZ?RhIYV2Y+-D(ls1H4M9PTHftA zO&&ss_-t2(!PU%WE+L9%RrLJ)yi4*D?=JAAxdPiVK7c>h0#tY+>={69AV3WQ0`V{m zt6FQ}^?E(js;Vlgq9*by!rm|0j;CD2eD`1YZu^4YKCO3O^f6of`D@+X8GrNSgiEcp z)?p}QD6Bp+R?U10;NwRRKK=N^AN=60Cr=(Q@cMQ6`qlE%je~mehz|DUxBr5@^af4S z8=t-Z%@2Qa`{R#3XZ+Ds;iJzU#=~2@p>Oe@y!BqR?1-7j!Ii62D-D%~>SxMRVJckT z`tY<4tG$7Tq1=6Z+8@cD@_31YF1D+wnsrIuxqz;Q;oiLm2bT`v9c~myL|(sfn34vb3g8rgN1B!}-MyWvIijF1#9w z5?$y3(;>DZdL={r}`M0Qg< zKi^vPK;itNb3ZndPXq;YyVZR=K060Mph76@eIIB|YORH-4v{1Ot*`DMtC_YWSLoHR z-4?kXfFdEU`8O>X691POcDQF-x~*km*3EWDq6_)St3GS8;$P|My$>P)33v$oJ8hwT zz^9MokKbNh*?j%_YWI(H_3}4l(>mX|p!F~~9ln^1X#n;%d*^2- zM@L6~lT0y6|M2j@^9+88#Z|o+!F*gpAsW+8M&!-cwh!449tM3C{dr_ z;tBEi92=CVmg;$EWV}&gurse$t5SFfu9g1egY&=g-4{&7e6P&NM5bNHDREzLA6oHr zxdEky*i(pW_E~+`Y*vW*vIaQ6{DqgbSp(+Xx5eqt$f{?5BP|O(wds>bCwniva`(X} zUw_eRZTknjUV-EM7t~4@7wznP5;Zleq}S{A#%tgB*-zj4>O0@Z>%VnpSmDwY`sf(< zKOx!uqs>=EMW#4ub*X_ag5Q4W_T7h1kfz`nSMmL~-~X$B?viV$nH?M+ou5871!`p` z5ivuB*-&UTlqYA~`;R(P=1c%BgR0GH2*Nd|Kq0y)d4F$jsHG0YPtl8+sQ?$geCg8p z*;!>YQxPlV2YttxVm{1Mkf&7IeE-a)l)_|WC{`3AVzm@6kyw>l3U9Y(^9{oQ)B>M} zQJB4*!vB#=%;kt`4YNFZ0oM1F=Y#Q(TkFSjQw$q}d2y?k#P-2EfD8H%bNUrda1|n8 zurf`;X0;SQVN*4GjNGCVbE%~c`58dhhIi@jrfP z0Y0CCX$0K7c8Makblbh~M`oWrIG)r0OxJIJ<^1FrCIG3=)@M|Lu7*NfPS3Wp6``kV zOAH`eud1oWu7!CqfDZTfFJIbMX>GS%t=GLJ+$HGFbI-l^-g{M_>^*m%T8wfQMRFLnVaWb}?ZF6Svk-OQ|bQ zF*5t4q7H5dL|h7x2js-Hjo`!Z@@=uGE~!}*wshC|S3(}9W0uR3sKqbKdDu!GUg|Ld1;T*&(4hq_=~Z>-uG1=FP#;!QtWd^w{x>h_+F9#k9W~YAJUgJ?^U* zkJ*LoN3oJziRcy;YUgfQWs4bO)J_MbEm!s6eJbTv)7E=9=?P zW})iO#rX;9)zBRlU4|;TJw;OwDPLz^N{`7eZ;G&pCX8^ zu@?VXsD`CA&L867H8m4Iq6{(1J|D%8V7Hu$Izyy>DIiu#XO-!1P|1;4eSOH6H-_1& z+5X{P(N}Kr9=46HeRS{U-G@gfr$ed5&*`m0DTSwI1X|Pd-9P^te((qO(rdW+9PK8% z`zdy3#`0je`Mr-O;%QaKSc*Ahu{;GAK8oVo-+2CM@-X_!5~kx#-k zPE@g!3SufmI6FI?<-ZF$XTLSTOl(ACAkvOPXx0*P(K5s-Gyp+BT%2hY8K-8gDPi1= zrKX|*Fhlgt`zbXBuZfU8aH^YV%=&0P8LI!>%VwwtS$+&l{GH~r>e7JqX`P6)5IWLN z^knFD2fa#pX$WTT;8H^L(Y@`Td^mmQo7b)$Y+l`7yz>A%e*W?4xVO4kudFGv;?lty zz&Ht${_eML9*w88+hH8>_yO94ufF)`fcK+tpCvFqeemS72TyKYJ94!j)_Y}D$cSnfok}|GM*8^PL+a;kA(m|!3h%8}eUw`F zGOD(>+3c;>enOt^w$pCA-fUoVaif|PQcp#7decpw7b z5Eu>*56{oe%!h>`Vo0IHNYKmDs0zsJp_TEo^d|0Q;`zyYDlQ&nrceW!L06i(%SJ7`jy3UuP zWF1C%IyIvL4Hf&@XYH*|+Xs(7`SX9_)f?9~_m0ob&sSo1#C)emhQ$HaoiZ{G+=co5bX`sw$s7c+%zm3lZhPW(^B7xLgqs_VzZ! z#Sf4dC@4xQ`be+}e*U@Ve){uw9J}VI4TFJ+npzWWt;O}pO*Ao4Eq-DMn0cH=Rb!$y zP0UtEo3tj+lx)!807Xs7T5BS$))j{ZwD4{OA{3AhxK*{Y)6@OK!w!wGFur{e9htcp z&f#M&mBk=v)|e9JOaSMZd}gXBFcmR4H>$14xa;QDa`YEc;!V4`z|7@D5qcj*yRiBT zu>b&~`yYRJv}%W&p^#oYcu=mbwqx606(05$A(O$wx(I&y<7IrE5d_lsA%ETNnV;vd zUe$?L_s``2^N)Z2`pcL1D%OO0`MHOmPOGx5rKs^x`23;`m0y4H;QF0wII-Q{t6Lj> z{@$aHAD=%N`O>A$(Uq#Zd;15EzxCqRt=}a2t%O$tK4C6F~8J;U2=hZZ${mrWBf(Ziv0LVZ$zlhmpwcg*`bArNg z{K?}->-8FxXXAm)Z13Q}VNt(Be~1DaTN|gz$FoGYKw+nP0zzjJsuWU{*4nz3)lj$1 z&d#4|iXG~r$(SH2>_^sRDL*C7$U~XJlmWA3(^JVUJ3Ku6?6bS3hMM=CdNBhv0G`rz z^&=BA74`BBuW?W_0++&Vl8dn&0%ocr2%p0~f!MtTv03E)ER1b7n9qP7F}-^Z$}$}H zzkKxmJHz(=!MezJerl`F9-W6t4B5YONj z;XcjC{>57t*RJjll^;LZzWs~S!wp}$v{!a#SMGe}@!bcdwCk5v>vh?z%U+?w4Ie)` zz5Bo(pY0x9>>i)jOGkUp-B=%NCX&0S58vHR&9JVu<_TOLuTb>bBS)_RFg$nj>h0@S z?mj$z8vXz=GTJ`9|M=F`BO@@dcW|_S>B_~~v6Jofv&U!0=es13o_hJ$9&A?rbqLhz zS2hssAMEe#AFyLsBIYuVBl@vO@CHc0^x$Cs^yC!O{|^KrhU`R|GGSc^r5cx^RNwlT zhX<%yDGaO9#LU*KVOX!u$MbEI*%$-sT5GMuT-bReu%FW5)1o~I0AsLEm^V`A~S4V9`1QBSV~Mu*mm>Xyn{`vu>H-91e5!(`{jXsv0S1$OD_W)?pZS({8<4=O-%Q zXITN=yGGR8?X7ENoHpfF-(qO}yibfAYpe>wNAJBe-v4lKqjm47pOOmG&?1O=r8IFf*@LRe=wllt1|Sec@s6VDI|1gBNd_?(XUK!(zw#oA%ksDR}kW z*S_jwXgkb{^9DSxAHN7q{myG&{fGbhhiRm2zA?Z|fB4R4cYgmWf=;EQ8@Er7pO{yw z$0XDlhZCL(g%fS&TR69fL5KVMwKBxWj0f`ZyVE$X*XwTU9W~y$efyoWGf^kD8L+*9 z-L4lUPwVoq1{D#BqJ`irjfpnHurr$4REUP5NRv}ffpy6wP+_mIueJK=Z6%6;hzQ0Y zKg*AZh$0IVO$^#Ega{hLV?;DunCAe=`a~xSV%-SD%ylRnhc}dB?A*PDX8===)!lwE z4>caa|JmW80!9o!^E|OOfv8XXuY}h)yr)>ZM!Fr?|yXW+M}CS z&iAfMDf;ACx#(|v{aZ1#Tg^Cz=64VpKUt!F(Cr&n5}qYDzOZ2T9v$C*a^_t!01hr) zSsxr7A3q_rdc1A09${b+3Ws1FY}Q@Z3D8uIgM+<;{e1=*ld1T@zlAZLpX}{zsArdY ze@xnnaN}GHQ!xyN)xbk7rSPuFZX9c&p_J+%nk!UH8#763bttQ0Xd>GdhKce{5B4^t zFw9m%QL{=U>==rAw5MB`I@1D;yaREffO!$KkLtn=*YGE7!eDt%Q z-Z|RbdwlW2&0$12zL-e$)`9JD66D7w=QcPR{5#CvH_mnjfab4f)dau%jI-r-haFhUh5kQ z(D#4(!C(EoZ(1gHuiv_Je)`x`ViW^QHaGq^63>yc$X~yBNjCkFvvE_G#^Y1R+%>Qypq8 zTuP};o-Fhn050wCcL3H}^Dsw6O0B`6FV(pF&fX;-iC|iafUjhr)*^bo=Yu3wURjM1TXFmK*v#1x(N8MdSeG;A?kY`(I` z?e^=x`}J>p27`Gw2JKf8P5#@>&f z?9R`zDveAkkJ}v&6=L}R(e`i6vSdekAm*2~A|f;QeV_ZO8&Ig~0t%=C&{uMrWYc(Q z$v6~8W*Rd-=uPvW7nw|wncnm_^eP)0Gs#Rw8&TAdGfi3L+T;uVx%O|IoH;;B5Bh42(J_-b5SkW7!|ASM9Mf(FFl%Ee zrPTZv<|5wy>-}F+ATma-gFd{SJe5+4%-W`D8wFYo*Cvq8yoX(l*=r5V^-i5AFz*Gc zYaacu{`3%Gxms3Uc3Smg87+YA@9y-gD+6!thmAR?lRc`5iIHb*!y%+ph>?IAu6>T^ z&2EGgs0kgwfQUpKLY&+*B6?|%hm@|=XsTl@0w`!m<7{3Y^HezUK)83~?q{E#o|+xM z`KxCDzxt2=PyOk`CqMbom%j1JM}PQVBG_IdS9y89=1)#mC(Eo-o0GTV@KFP%QX;g8 zfNjba!1NhytTZvRb$-&dWFQ96G?vxUKt)*+fBNK3U{IhR{NTZtb^@9iatt=4u9UbQ zx=$W!T`ypk?nV66E=CxQ~3V?Qj zcH1y(BNHHDWOh;Sen>@hJ!B$mx|RvWY`N~Uh?x#U62rG%yFX+d(!eIfw3yFB3$^j$e7p{tVQ_#8dT!rQlQ{l?$xAWRNPt5N9 z!~Y}qXFvRB|JP#n-ji~+F0bCdHA>EmCJF$+Cr^zFT{p;kyCvAB^Quk%YPt@QQl_(u zZaM6n4M+3c`QH2h1=nX6B@YMt?J$%l8deetHn(o_)iOMKvQ%q`G;2u91|gGO9`5tq z!-aWum;rs~qksM0gLm(}bokOs@o(MRS*?y9Ji)V4y_|3<*U|SYNq&b9J=;*~2F>gscWI000q(wwcm3 z-o;dJa3;U;bKlH)l51yXH=Lv@9SEGy5R;OKmS@t~xw^d=&y6>~#Qeb~#*J4hU>OUC4p(?$NyNE`vWi zm-{cjqD_<^{%Ce|bmJfYr{DQ6{_i{g^}l-gYhQ`4zWm{L-?R6=|I+Wii^O2??whYX zJUQQlPIa6Lnkf)HdUo;jVtxB~&#xym-D>YyFtZS%nIaNEMg$_>UCi4MhLl{j8%>>6 zbWjl@6JhEF4FgdVD2s{;BAY_a89g5niU-5QID0Ku^~i4 z8pZj2HW+6qV1|G(FuFQx{W%0;Q$r>!g~0PLC`Gen@RoltGvje693g%0y$?^%&JK1L zaWVHNGn>s;tDd>3hP0Np$z)NTC4g}Qk6_#*(K@EnYO1yxxr1A;;Aq?c!3=NTzIo@) zP5k;hVDR7kH^291)}CBk`Rc!a=LR5Gs5kxRhH)xabVET1bt0i(5RT?RD+dLZ6J}_e zMcixpo!O7u#){zORebn(=i$X(;9^z~fDw^!XMxP~g5~nEzgT7iyLDr6d@wsbUtccM z!LH@?`lZ|Z-jiWy2L}h#r=!39JKcA_|M&+#=>Exn`7-kpF!_UjyB>z!5PlHxXs@09 z)|V0C*75G`KhY&*O*Tb+}^~-gSfW+wKP=r_~ z|Ckk#$TaU-MgX*aDDzp<_k*YsHy0O|5SqSES!8cM1HvY>3LpTFPtIeE2qw8!hOXCr zkus#*r_6+mm_$^;R7=s6v#9RHn2R8xrL3wPLJ<)aqM8RJFIHn3QW?jNk(r5zC?Hx9 zWo9!a4rv&~fRs}zC#UD23S^>YqON%7OQj7IV-o@ck{E~r0z7^CboSMk0RfRrtj6Ti zuyGrQT)H=?)ZfTV*4mhH(klP}9NfBJs{(M78{W$87!qD~cb-1@L0~qs-+1?}v69u_ zJhlqhxEaSFy;4p<6REdZTMn84&Q?zsVHtupEyuv7paq6?e0J8wKnCC)e`?xkmP?Jh zN7l}c8D2f6@BDcA$){%=crkBwW)?xd{wsH3WDCFbCqLk4+JE|>`|e-v?d{$A;3u}b zGxUS@>kF`BAWHcjhP{uT9r5K=D5e39PWh`n)L&}s( z*Fnk!5!QVQfk}Xw{O$(vIW{UPxqtyQOaPX0aS5qZgpE@vt3LVQ&RG`$U_S2-4-Zn& z2PfwW7@2Zj#~2o~E)_{Ri)bz)2$4y|cdm)4s1>iBFaQxmDy1OvkV~K{TIw6X0D@9h zDMg78#1O4G*hlad3Ns_~c{6H=S#rw7?^RJS6$BJj5B>{#LYPAg2M7CE;3tnxXY+ZD zxxt7Tfpbn;#73gQ)Hajx7xXJB@4>y#XPq1oB_;}Bpz4^0=-plQ%j=;4$3s*iiSp&w*t z-mLpP6nS-#F4z5%DD2!kJ-d4M-J>_(y8G|`^wS3qFHcX)q8*sBsd)Dv06h5gSsvh5 zE|)v2e)j0e0#MJ+pMZS?x(2uf@G{`*0{!m($=F)4sV3R%QKj*Q0KlLB=%c@Xa~}~H z832e~NRUWf^%k9Y)%DQB1R8w6mKg%bIDsq%GF9+Fh=?45svrPUaI5J;Xic5!Jlz6>u>oRuxfi$fsc^^t4E%04Vt-70hpSKnT);6#Z*qOP7ZFg2gkFf>!c_%a10Ml zc2CZCm?>p7gZVs)m}Kc>eRnRfT18#+Z0C>v;z?^I z6_vB|^&!A83{Svh-kt>{fY*WkW;6Td?%u=w#lt%{e*D(!mw)5yINF0d$NRUBD;7n+d2HIyboGBkSc(LUBElg!PlCi3mshurk4%LSWZV|cl!+QEgi#l&h;#@c z#54@Vftf{$ic(-iY?`p@2dKhW2-h^wy*jDr5Ca85@D_nBY*M!zQ;2cI7KQ-<;uPkBLB%{{j4*DI8XHU-VDk7b zXzDKagZDpp_Ust|{MN61?YF-9_I5U7i^}x14seY(sCkBwFm*RRm4^TrwzDP#pMGr| zI14_0);xSNR|svOltj&jB*ReH#H8Fmu$A;bdUSgG<-0%mxcT5gdh>Yw_TBi}(eWRA z@9gMkpV_Sped~{Z$f~S>UAvlf5B7G~PfzDfxM%`UP+~mX`}WS_q1iVspWI#?PLJT7 zH(q}D?%X`|rxo0AR#7K2LMp0+xLmFA2B`^o)ecs5OGKRX1S3;1 zL&PG=gv1P^c#fzkVd^vi^tdy|?rM2Od$hl^5GnhM z?r3iwn)&(FPf`&=6c3$MOA#6?L=ed*`QUiHTOL3|!%(V-r#l6$m&kc<=r9pPf8gVE^l1`;wM4R{lU`^F=ll;Z8SAmona; zQD4VDPz{$o-tVye`t9F*wEFbX`eDxS=u!CO)7jN}wzFVBDy7`MwHStyi#AlUVl2aJ zHy39QhxO&<%P$}Q=<&(^EZoA?M;|^wDTn*Jk~5e?bcr$At1qR;kB30blc%y=op#Os zxn`AR46D5UWZA89KKk`v&uKHj{KjrKK6@OUl ztZLeI-F(&}!jSXn`2_-$Qq0V!T9i_!0ZdN7ht*Cypn4vWT8ZEqjDE?V+X3U^@ecay zQYcgY6|MpDNx29%mbo6zR1s`Gn{f=Hx#Tp(ktSjyCdEY1fa=X+q=2<7564BSFqS9P z+b5nTQ&4^Q;NgG%pZ^#C?4SKt-~RLOeDLF+Oy~EC3IIDfd1e4#{py#A897V?X}tEx z2`(JjGBY))&0jSm0bUzuRY`yl87Yj97MY0e?!5But=~8d_ul_-adNQ}g3H2{!^7Fm zqV;Rd0=qjcdSWJfN8+z(uiG0bglo_3;nTARk5Bx~V^dX) zXf$v{1Xbg}7nfIiI|~uTx`s?d>ri+zRSVV2SB6=j8dl|v8j|2cNS)4oyxNd?}EovGI0ojMUtO_nM7rriDX_Xu48wB z#{kGUEpE?#sQ377gKPg_Bix^szgyY5B{uWA!3_uS^U`iu$sJ26wOt3G{gA7^g0D`= zTS0|6<768H?hnv#tUsL&JR%?e3o|Yd;o0fw>FJpP{NW$|o0ngH>G=5gYhU}Cd)$Be zPyZPJ{MOgLXceW6-VD5q4@ady8+`^&$zZ_I?>%>QXgE5l$%cIC&MW`>|E2xGchA4| z{%XjUvmESq2Ego4Q-`D?J(aSz(|&Pz_4MT8=*H2fkDB*R(x=a|0={;8;hyHt{^$pR z2(GTe{k!qAM@?2XF@O>|0ibA`O34KS9p65}sm&BCH~h{UFa5Xw{)fJ?VM78`s7dC>+d+ndTz7lv>-22z6AT(YOJOFb2w^6AJPyJrMzzh=dpk zv1#I}PY7sg$Ri4>Ph$*Q%@BZzQYupz5Mv{v21a1aVd(p_)ykiH%B5`$(dWSwFTJ-> zTV*VCm|XdGDGQn@BD>C9x!zPWl?)MXp85psM@7l$k=IqBK0)KUl+tuS&mmJ2N|B58 z+Mfg=R?!H6+AF{eRU25xMmfhkggntM2!LaATJ2u}qY;(Y=l$`=pM3oBC*S_|p926h z`($jYHBkmOjk`lcMBXqb^_XA?JW8I7>4xdnT)RB*B&=VC=Zya5SNHzr+q-}87nd>e zr&nP}#ot?g5)-~ng5CxH>2mowBpsGSX`lm4raKX^gn)(IWwFg_t zk9tE~&U$3m{Yc{~oMc4|Y7jh`$ZNI|k(@m-v&ez#fMHZskz7hCrAUz}8HQs$be&^j z3Z`aiDxeA?AYL^krlMN3Wbl}0T$0rmj@R_);iK`B0HG0hqlLMgmzlTE3#fpqNiiv= zs#=_mRLNGvL{-F@8Pj4?v}8YZjYZJ+j#VNH>~DN^@mp_pU%mrhz0W#!vFD;1J6K(D zH`|}{7n;kbPcLVo%$foKEsAL|EvE9FKYPDzVpy+dmshha3xLzEdECWw#vw*U6sFY1 zul(v8E;s#?DG`n~a%juqvbCu4x{+2a$D5>k;;Bu1{GFdzUBQW2%V7gv|Z zM+YiGpYmJo}ms0I#b#^la~ zMGHcmaoIIZ8(ES5>cdZ_i^xSBE!H@gssbVsX>l?1as4|HgeamuqQXYl^SlD9uN_0L z&A>WZ=302`pMQIut7?7IvzTTK6@nd3_Jk%9DpKDH$}!g3{7MvWTN?OwiVd|Vb#G^we2{VMvi~l@f#1IGED{c z1?NQ2#x-pq(AXWiG4|WLq<*tfvD=dYuD@4BaxPAw#1NVgVGEMhMK}pV0Bk7#I?tNC zpcU%=_Ch2tboC!n5A7C5yR{d#KFLcrcPrbw;r?r4vm-fwWu7lD*RneO+uwZ2 zwepep@#72H*@3-X+}rJjw0p5Ev(66p$}9Ks{d=psw^mIkB1gxIIx*XCk%y{Om;T)w zdp=Hcik3zaq27~6C+81O&I1CbQgG{*nQ90g3M(;RUS7>_9-9ggh#{DX>d5n$b3v(O zWX`3uO%p;abqp{&K~6|*6VGi$zG6N{h9aVvPGgjYoVzxAjUB_Y>`!;O9kf)B=P8W=r*Gqck~kBv~(I7%bb%IdAbi=?*28ao67%1D{R>C<#iOeq{rH?2 z4C4O5>i{48_|e~e=iWPSy<7u9fKMKt;%bdAy(9o-NNc2cdQKOYHl!c|h-Kcc4tBnH zIIr`8YEU$Uo0z|9@ zo5c5Ddimb{m+s!XfBNjCna$U$)#cfmj&uf6#jU;hF?-CSnSwB2m^fS#V% zjl<>FUcd3+1TL01WKtavAhH#ifBNL@|Mahoi=p;a*UvvTSl_=rw&vmnV{SbgdvtOh z{Pr%_!~V{k5Q|B`=+ft+VyZ-VetEUGvoKJ%4+n?GZ@vBY+1c5}a&_D6_{NR(dX0dw zoqg%;uiv=y_*;MSM*zl5C7X!GCSt+1iOY2lV@lYz$xy^1BHqZj2?MBUciI4Evu-}?`t^!&Ga}klwPb+N7b0y#~Mw2y~;^L z77+o63=|X*6v;*X5Jqqe255G)&$buoD{s8DTV3o2QBX_V(kDP)TxY9lH_#H$78%@So zTn+ob^7_{1t%=58{na-wzVlc1B}5{SoEWi)4$I4ni}O3TZ$Dfw7yE~Ozv_oH?;0WwOnFEm(neB+Fosl! z5JHe5U=}FGNV6t1k(~uv51G2fvuDrtW})5NdH>Ug^&^jH0|5^Y_5smMZyg_&q375X z5fy17cY)Qk6jTru6~f@M95Gr=oUs}d)j-rXm9%97Mr=Y*&@Ki*@bZd4m_<~<3kRLR z3QP(Ni1T^3ySqcE#XCDq+a@+KFaoQr1x`+%?ax9)%4rzUXodR3M4=&OpfE0e1rM+$ z^p6gCc;nA&HM6;?swQFl&SaNsux|E+mv32hQ#!$v$pcsVoCv)xdgOq{g}F6+1rX2} z2{9Ph7!y=rwE;t_wPte5gKU5P35nqNX%Q!40f_1E{%8N_yZ^U;0iQlvfA#IF#q2z` zu?&lkK8ee+%B3fG=XbttV-SK9>Em7eTkHMqt;4%F4j!IdJO_VZGzVcy7g;UvAy1>$ z;h}(wt~)x|3o$6t-IrcP4sZO*o9nAfL%dvF9_$~)ruoBv{jVd38#izK@gM#1-oc@l zPlG9P2uxy{Q_8t8Qqu$xMF1vhV?Yi`lbRqvWNu<;LO?WTUJiQt+1xjlIS0?%WhY;L?{Q3J6UMK2LxU10lzz)^Z{?F*YG| zq2WMDv#4BLUOsvJxCy8NPfniYGJ*Me9U+7!a4%(fwTyxDkcMGsV&mE*fuTsX5$KaL zT((iww?HK+%*ZMBO7aR}vly$!eRk6K zYboX8;%c=X00Q_d%CUOF5hslCE|^t`AOp}*zXH8zd|dGc0BE2~3ELD$-4}j2W>24- z0KjW^kLp(Pw>ahML`{u@5zq|Lm!zTMhP-ib@28|qG5cuVM*`Qj{s8bqM%`a^4y;y^ zvYySRVz22#)#c}JQB%D2cYgJz*%!<%pDoL)i!|JWl5gBO92o;J+Zs-Y2#NqzV7YFn zYPM`z{W#J_jyHn=+&bK|@!bd(0(*@o69o?YySpmNfz@<*b@?lAzJY*D7GlFl`+Iu? zaQoKH+1}BEhYwFre|+}rDHxtzt@ig8h8CEz$hsehm}3A@H3@-ZY=)FwJTMZwhJ%{q zBxcYLLmJkBad&5D$a=Lt|Lo~0R93S-?OR7jfM_OnZrzy8W+zV`pPrp{-3*yALuBj0 zmoJJ~&F3QJqF~z1x)7S8+KC?Q&Q+v{M%55FFvl1#o;|S%H;)`bT_;d$`mx$CU?eHR zOcAkw0T=}`!88mPmzSVr)^$kX;_7M`dWbs!peiN>z5d7Q&|0XCfab!9W&m2YY#ads za{wj;^x9e%Z2SDnHVDS8q)a{RBTpkER*}kZ*2w@!9Ds3D(ioc`ahw-hBan$o)*1|} zZo&*-`z}9L`7D6q5f>1LI)?!!nuw}__VUdUEfUTGuw$^WBIDn#5!imZ6qN) zo576eZKHVmc=yin!GkBK8`I&7NDbT?-8k4$QIFD@5Xk%VK-K2E?P4(_4rr!DPM$ov zdFO7on3uG6rUjW0Z0P%sKK*`9>EitCum0-27&&rJx z5CR4SV}uZinT9@_*~R6h6xrR`iHsE3Jp1NmZ(NOSU{PbD#bSpUXY={x@~UYX8C|ZI zGXj#Tm|Cq`Q6ZwQfAiN44vs{`BWxX;(wUHnfAm*BynOc5We(AG0YG^Pj(CVvCp{K~8U{P&-2 z?#sv!06@5XxW_~Ym_29zA-Til_<^#~4KP z$=Su>!LFzTjw03%gHP3-w{g)0CM^nSNU>>Dg<~)iz`%h4A&cES+%eNKtcM}twR&I! zz&F3}5{CeYfoZi`Ap#T4x;B@hW`QXe5kw4(-gRj;SzkmO#NFLJ5z&Rkz+KzqoJAyZ z2$6G<#e8n2f!SrB&6J1~fe?e6eN~4oW@CSI$+7F)W%!J4Gn;iCG0od9HqF!ZQbn&) zjtwHB=A=Wv#Uc_n9k^jaG$SA~S4J7z%{+XhD(OrkCZi2Kx%H5%JtPExzxB=6Yv!vi zRIJPZ(P&3UPf)89ag2ECa`$41P2JMiWC8}0O}23ek9iQIWN0>?qV5fJwN|Q^Q$ejG z7Y3?TuTT?z=3}15leoFpYj?@id={LsjTFUraH!cs06@HR^YGD=v+a%R?+7=K4opF6 zL9YRkDX1wT?k?tSj7$Wogc!gKQA=8{SC1a{s&;;Uo^x?L$Q()3JL^FR5t%~C8Hhx! zloCQ9(GZzJbJ=e+8!~~E+{Jld78Ni{x$G|HJB!8D)ddo&$SQ4(gCD7cBEo=3s48vK zFkuKG7cm7R^dVhDBw}E|L_VdIhymb-Km4Jpsg~Jn*2ZX}3bt5u1Xham>s1JpOAaBR zSs)hBTCKEY`J4tE%h~`y5Q$BR8BB>$RGS!@7`iSRn$|*b0087bfydfSGXN=rR*FW! zpkiDbi>I9a5mZ7aE1kF0!T`-Rlf2~to;-f?+wZ(8c^K)M5iY~EQ^15%5CKsqJ`rn4 z6x3B&L*aU&iD@#Zfdqg}L5odT%V09yBbhKJt?HJjM%gcNOgs1o8vTH6;-k%`ola5F z)0m$m6F&ujRjTakZ`}XAr)L|NGH&7kYDgHoDU3)BrWv{idwWe2I1mElQU}B^7Ab`> z5>v{#sCYdnF-h&9MKxoN!>|@Lj-ixnq9F#YxWba;lnNsT4l33*E=NLFSAAdtz@y{i z*<$|m=@SDMk&ERTu9-Htd3@;iUPVl`2?2=FTn1>2T8YtAMMR4mIgvt$ufOqT&c$)v z&<`M`h%_-=U1V3a`qOWH>v(?wfRTA9N<>wU8HvfE4uYAOfFT43E(_`)LOq#i0YOSx z%;)WFjzngps#E_ z>3{Y=dwaClW=K}~6ROekXwIu2nbh~E|07M+E9lAO8v2Z+iSGqZsF?vqRU0>$jDB*% zq>UVCHT~5>6K;KYtXX;|F1)zZCT7dbXtQt4U9po<)k>Jx@{RcucguHg9^N_Ld*Hfp zb+;fQ24X@4Q$S`j8?uOM46$v(?VCq!(}WnL2oa4zWI@Rgf{7qN!%RfJ!~rpeFu=fw z3Q9!JF4i}W_6^Ahhn!5UiJ^!{63Llk1VjXFm`xQ4hat7QJ7OA}hB>bKfr&4sK^Wub zHh`mp{j>9Pky1+0B7rG{1pumOV9F``!}JQK@lK+z-T8cR^VUtLS=uHvkxd~49Mk&> zfTmX4jJ&@A9R+|By9TCN&48JhsB60zV+cV-IFN}Ek(N@#LeqGH36P2?OdbG?$yoI7@B(G%11-^5TEfA);ClOU4w;_Vv;58MgV4T%620<)PKqk?HsB#JRG(edH_ ztck=VQUW5zpx*jK)&x|7z=RlxXzfz{NQ9K7_`Dqs0SQEP7;@8uCU8zkq%eo|dW}eC ziePHmFzzg7#B3^NmUGI~c0lWOUy8I&zuz3KDgrMHBq(V%=h&a@o z2BL~c(XMT{2ofn;KwI^BhZ{r{wMbOFr^z8N5)f(0W@2C_GR4H>gHr%RU>cuhrA@%< z907v`4)ZWT004NJ7Xa`y&nFsDO-;(^5o{_!ANj=!w~zzYQ(%3o+w~2`_ka3Ez;PMY zwAvV5SFMdrqzML4(*eMMKJ2vO5*UGfJ<$~1RxzUrn{wcJZ?Rv=F z&OF3$`s^8@mPh$;f6IcW?WLJ(0AWiT-FIr1gvjD%<+3V<<^fiW5ryO7S1G6FPB z;}4vOLTuY^K2qzLOX`>BRs&cXi(5uRI!1fbvaA5wMuZF3uuli61vt26l9Q^8sAID- zU1@qNg*=9owO(ihKE8hHX|~PK79D97lZ*q4pt1wCRjnfbiOtta7EK1+k2B7F+P;v@ zCNcm}6$G-e!gah??(oN0WV~0Keyh5@>V|*kOZOf;em31(1Pr2JNd%Nb^ofD0azbvcEWjavC?YkHF8ZMqA;yx2ekcH7YAFp( zY*0zn0(!Lx8OG)6Y|(_Sje@3P+y#}wF?ECwjz zaEJ*Cm`M|YSL!>k9dc1X0tLplVP74AxoP5%2ET&!daY_4+tH2zx$b(k$_7XioUd77 zt$5kS2Rf}3+)yQ8CV*3zQ87(reWpbV1_B=U8jeh?R+`JkMLg|+EqRQ}JJ^U)a2m5Z z9%chp#9%BXsw}8gfun0dgKu~kBi^xu+s4Tt(}nnt#u?t5Pia6f76{ZU`P_v}*zwl> zVx+U&J5dmoV#ds3C@P{7LO43y1Ar!mJfxH|G-SE7fkN9w#=Wj%rVlTL2-_HlxQG&y z6^t>ayzbhzPeUovwjD}A0A@mirfG8OIk2j#WQ?I{8fMN#7--hDO%pFtf}&T;rIjQ= zk58|zuYLo>mchj|5BB!wJM)ywa(QXy@+}YcIJ ziCP+pYC*G73J|VVtMl`76{+EAHC!5-4z5{QPspf>x9-<9JFViXtA$f1S1rz{i-w3e zAn}x%0)VNbY_y0Yv@jJH)}Ay&>&^B|mTx$m{vL@-U<$x-gheAx*$72Ufv(lm9+(z` zYo`FSjUs}LMjglOG!XiQY~uL3Gss}ffm^*@XsS}s!1UUM_~!EqxqYzp36WygmSGKuDGfgNZt`&eaJaijS!2_{FzoH` z0+IpdA^9jqQ|lUD4_-=%-h#+YMC867m_y(9W;&bAODQpM6DY(m8U%<$hF}21G^9iv zvk0;4KQVxjG1ejr7bh)&gR90O1Y*`9VL>n{gd&JY90LmgfEFRZlarGeBY?S@1JuCJ z%w$B<1T`%>hI&;}+m}wA48SptYYfJ2m1=`Wvk#_aT@DvdRwGWWE6az`fIi>3u{(HR%Ag!~<$gAu^8%wJIYqp09eC(LDPH8~vast?S)z zcT%er0P5|s5f;_?%zhSUL~NXM3xMDF@~gl1@4jyyW*7toBS3Es9EQw+5BB$B49x6i zCWfez1JR<3v!>Zw%(LpllM5_Lfhk5XbDh+{Bt>DIiHk@f@GS6Ztu$8tHBA#Dr!){k z4D2{05*Cq^(ja6=rO5f&scWyVw^LvM0MNz|n`X5hI5b`p+h4ARH0;b5fk`C;peU%A znv_xyk%`BwiD?mv)UVdDjb73dD5#mbi6uhB*?dlA1u*Hgh={pRin9&oZN5coy2v}q z0SGY$FR-=96qu1Hi{|y(Ok>*uSk=8#SC@7?fF_H=+@6%6}98PK7(Q-@GnU-vc^g@LD}RUI$J`^hIpCG16vZ zTmBSnkGl)2YjVqulO|vdk!s1CC1>IC|PCRb$4zY#=r~!svM9wfP(XbT^o_?*|V#Ic>|!QSwj<7>p{ed))KLH z)}|pLk!hJnO1bzbiVy;tHi1!9i-f?g2N3`W6oFDo$PE&=Fq zOKIC?Hk+5r6^4*h!Cs9U2TAmaY+ z9^w;!W+Fu-*VX1O8>hk>_3keliOX6IQRhbD6eb#wmk6SPr&K3`e5BdI3`~V6AlKar z==0JP2t746;uixLg`aA0UOA_!dVE^Ij?u9;vDLLs%t%N;<{kP5hGTQHcY}}R7*NyC zV`dZpjH+sm)TKJc);Q)NoSiTv9xE!mJJgKPP|V1SCa3o5$!TZ>{ndZoI^2Eu^xVG@ z0I2VC(?-BRWQY5EUDr0uF`(qqU@!nP3J5UPkc}&M)WlIRHvZT?1KE)KrQmMgnmY;&OS}H0_XjW@P5R?{{_^ zm*T860ugKq)Qu=yT=g6|#8`^ul%cF0Kh!ruyO6z%kU}T$OvHO>Qq|S;+Qh4K>{Ldt_7aI7Mq+x zjqfY7StAfcRHT3gp37I!?G-=%0b^o&^cr3hVTOQrj`zO%UVYpF037h7TsVXqM+cF) zX#<*N&=^80Y9>suPUW&6Bd+MMxQ*GRl$Uf z&1@)=ix5E9HUJt!m^FdD?2O5XX7hPkFBgl2YFYN1QIJy@aD1@4?h^-2{YnywXiDkE zjpI}jfD%&5t^k68NdTBwt)yX?&t@vp#yJ=a%gfV~lf#3ZSzwMarv!+B$%jC#E|;}} z(vTtpIzF*LLDf_W6GJUDtN@6aiBP04kpg@7;{Ra z*f23+ok+J8TiEEy)SO!(0B@+8%A}`jfepJeQbty7{N%fkY3f+ozLw9O#egaNl(S5i zWA@y&+`Dyf=Xm$w$prw0FTeQ(Lh1qnXxH%G&aRpE_B3oTv>a1hHVKo52 z4?j5((W3D9<#t(k-(ZT*sh*1nLFHYxO0|2wO9g1|b#t@d5>qQ${ zrX0hgXJ^i1JSce>im2@@7EjO5Vr;rLN|6{s-wz>##Vi1tsqW1<$k4?IT9)gzA_Znq z8?qW`6R2%tQAI7LVx@=*hgfoU4}*lS+`IGgy*rEsXennfOGD0P4A?{hfW<7{B~-+R z19Kqpx&Q#Edgy}zl(Ys!FdBPSbiyDuir#EE?JW-g8y=JZ(bTgCRD%jo*dqToyQoqr zH4j>C_@r4jaV$=~k8BG7dao5oRx_+oLG5xJm)wS`m}baNOB9`lBh>#N$M4Le!<{qE zD0fFf#+^;Z;ZCx5$fmQCl`_wWvqxl>y$J~+D+jyysNWew%X_~hX>D^pO$Rf-@ z6X0qsz?oHfK=ZSD(^%6mfi#jE*Eiq-H?DD^K&2HD_-!q4#sGx=X24Tx0p+~WC+u-{ zX6#TkeqzB`FW+@d^_#`QuA&=Q8hu&<$ro z@$raa@_4^Sr=flZz$$(w&+TK)`uJIl`72{{=+*nrK{c?+@^@CZ_-yf71Zj)k(EkEK zb#iSw3y$YQ(u6>Z%qbebM%ZpJWhEamA|w;_ds4=|{r9KmHP^V2B?$#u?Okm$Obo&Z zK}0#*BMT{b@2Pw57u@UdFI9dwqWs+D8NK`(sUonJ+A)6ZJHQFIH76Zj5>^MsiyUj! z#YN>dxN4R#0C7f~-Y>@%PFT8ZTP&n>HpRU?rmA07=_#EYc>A63Tzr`&6H?-{Ar8zQ zIRt;NHoCb!EVmC-uO%Dkg*Lb)t|j>MJOfDkG8I~%_O(~NIx<_V+LZsXEkCQKd5Zl) zj)(+mO|kD;kt>FTHOWp6xeRe@vff{<%zhIBgUVJdw?alj5IC1iU3!#v%0S4GQz zJ<)lJulzDLhNYs`dCvPc#jZCh-IwhxeM7^57FG9?a9DL5HLF(;yz#lkVc-G1Dd@p) zcd7LLq+fhNcmG}w!KJ)l|DfNR&#J+SfCYCnlZXQa`Bry|^QLV_&Jxcap-#(Q*LrzX z{w|$c$hpnp`j_4fkPuHY+5Z%ZepqAo)geSidZ41i^RIjkLz#)At((ke-G^dP604Af zer-Lm*=={Wf%Q8Lxqj`|T%TR13v9yX$1b~F7t5+M?hm6VY^|mqlPWnWp}fc(ddve& zlRCUX$U`5V(IVD{f^~ zL+|?X#~Nc{xQ2x|GX@s74^jt?N=lDJV&10Vw8b6izxFnj6J0bpMlvZ87&XzG=p4JA zVx^k*ro{kY6g--&4-+|B`J1xKtIt%tMTH1(P^=3VFDx5r2>2^cN{tF?`Q1RuzPRvz z6uiw=SRwO4OSj``DeI5*NoZ%HUQ%XCk(+}Fashhtet*Ao@^Xo?TZhfGdn&$q@(mYZ zkz!Oo2jTW_e1ycIR=(7qplJAz)gy8n?ds9Z|3+%MBGMyXRJ~0p`{%ES(+f!qmX8t5 zJaohO&ht*X(67(zN;L1b!umZVsjp0|ZNEM(vdR4$@0uTl8{od7prjJ<=KHiOoh$n= z@}Pa~Lz@Jw?fr)j#*`LX&w{f5gXGK&&$~8?qR4T{z?jWgm~{=9JU6Cw6o&$SMi;no z0nBP)#D9X(T^CagA&LSB19Jv7am!b=r%L|&WN({j^oNFqCJ=WyAtD`2gQIC)I>1(ws+=-^!qc?{6B?<9`_?f}QSX5LfjS zAyNivWA>)tG11yLQTAX#R>3|&cZ|vNWHK|#yG0j25}rkRZDqM;Y#FGF)1gjoZSs1a z{_<(dkFL3+3O&IXeSU+B1slbkTC%@j`qTIfIw~L5%8`)!ajsPy$bp69J%FE7=CMZq z{3DRLb+mj`)QFVA%5PqE;W~J~mHOL1FeZ6@+q4{g>QHEZf%t=t^)H>SzuMg%z++z4 z00@?uQ+?0$iW3(PrlbdXng)_mP zqOx&ZqaA%1#dmquW~M&}3cZOiAqnB7#9sB50qFC&6_<=YX0KE5`$Srj!K zV)o_?4!1Jv_!%_wX11L(a;`cx==2)~Rgxw;dvEmZm#KzpKOWaxHmkRoX;!4A;p4ER zuSYH#bbuMPR6|2LP;zYc$pFP@OB(9_u-|N>+7JnD7MEt9?A;+rH{x)w2x<==o9>x* zMW-vZ1seSMk*oR3yu9Rdsv)z4kiDy3-m4m?SIho2Ut}kGf5Mnh2H@UE!QqoNadv8f z%gC&4G38_To3PwD;(-^^Q)7hwlNX=dr^@J4UVp=>q2;ZY_4eKR&3s5(Z;Ejd#yMQi z%8$8RZ~2w|s!)sMBHD7E53v-}MV{Eb*?~A@6jimV9Y8cp<@W z!mrX&K7J986Y+nj}Y46=kN{%&BYGz(IcOsR-T48QMR>?=FKh>nX2B{8Jh`Lpo+S1mNC^XJ)v(BK_~ z&hr=>-T;-}3AfPl(R-Q`g(Nov8x* zpps$;MS%*uBsEv%$)4F)*je?X>=zFk6a1SrKXaZa|318yz4C|BFR%3JE?y7_Mki4dxQVM{-Rh)A^eO8A2pE3?>5L_VK(@Kx=>V zSAc?0nrQiuU%pnz7Xu5SWi5MAM0ugt%1ogJ7J^3rDlX8w;0BpaqhOHnThT~qgA!vF;|-u^G*Hs?A#jesZRSS-!28lu)_vsa z<&<_W%0K-jLka9}&2W?EcGq^Ki0{AK>`2%A^4Rw&B@sV6_NB?~0}M_o|5fO5eU*BA zv+Tp(Y10Y5ApRn^!7@&_%;!S&{l*EeVrVb>>m7XF+83qMtR=$1-3MpeWv#Hs`&TD< zdh5$BjY7XQNI!JD@`of5t)c(N4J{cYP{pD#a`gYX73^iYl|LbUGT)okhrlQ@D?LT7 zuh26+&uSsNF1Qv|RZ!V+?jHdFe$Iqbf|ua#BZs5x5&AbSN zn9{;EkAJ1|d(ZRi&bVfC@S%j`J4(^)tulIlWuraPg(^7YBjFk*VFDSdkx2Eb0(4Gy zWVk&j%!P&asVH-~9kE>3VU?Tu#?E`1zbDybeXtf+qrRL=82i1jTP4| zHr=;B55M;wVlGl^ZNf87@QfjgsvXChdH+;Gnw$c>0DB*czU@b!RXnW?Pt4=E{NZ!; zhL&UAO8G%w5qZQPH*KjUC*Brf?ZvYC%AByG;rc>H;H$BcZ_YJNO6BqICpR+n_1{?(t38n=WV3%hH59Ay027FE<(=~k2YDR-0bidPlISDB$Fi| zMzOc`fo}9I%~=7VitMyWxQX@BM4bKwjl(Ori^HL()d6p+WZlNgd@apzMRqFpju7vF zlhyOzKU}30w^oXJ9>^rVQT{BLbLM07FT6Hna`~8}^B|6Ui~9a??!7lv+fF>Yef2*b zX+j88`pdRO6R{mXA1lMcpB+nOhgJ{8Pt%ndj4bln{4r%@fNTL+WbyNC!Cj%NM%cGeT3%E)Wo|nGm7e%+lYfK!m>(oM24Zop7x@Ht_{5k$%T7ZDy_` z{RT4Adi;69(fNK0x>hf3TnF^}L)_I^p^EX z_=ac2DOC5X{On=d!BqRGwCViUXphY+xmBrlF59g?tv3KcjKSTexpl#Fu9vr7C)s`; zedpExDHAR&MM@dQ$k9a9|etN>A zcr;!7=&8uZfXi>4;LB%|XSUaNui~$LRY-M<+B{|4XHNidLKRtF=K|nnj^0e7g@hqX zw?#<%G(>^r^|5}rDciS}wK9irDqIc6$pW9%fl;1u+qV_pZBS-dWYnB6s z)%vcA6!s`02A$Xbddzi}aGx-tOtJyH?Y;^=`5$c(MK2<`s^L{$TP` zo|UDwHq*~g&0_f*D=o8e;otJ=9j!lz+Biptx&j{8)U0K+ zNCOaaC_1x?`nh2UoK3oxQ)6D6jj#FHBz0P1-Pz@7^`N1XW%p`|@vp?#nGprx>39Fv zckhvZVh))>AL>;x+2uIp-`7T!0T9eD0Z|6nb=<;XL8D|j%byv@@`EMB#-W<^!Ol|KM z9TsQ%oAr~@iOPpSJ*h|{h(B8s$%sPXKggMw;QiWffB{hc?WcJ>WvJfbN=%frx<0;6 zn3FQg!W5Vv^3%bYs8I~5&pH(J(VH55e4=2-RJx8rOLyw0N`s6o@Zeszc3p{6P(zTzm2y4YPl^T41iSOQ)@Cc6ka22u@*z($OcDp%{_WVGk zZ24~H^QYV29&9Hy$w2|{JD%%CS9Fvqz55LWtNRtzz!D%WOMSn}zRz0cAiq?9hfp9< zAAgx7j-FRs4^z=3+v}`qYF}1OE1z!Gw#|~?APW!I>o^814L!e*Qu2`YGd{M9b0kge zJ6``>5YyPI@3-mJ(7y+Rn_}A^PX0oX0C(p0J{dOe+0h{9TA3$JRxAcSI))|&UN5%V zp|(>_FS>;NB(I%HNY$a9LrA0;+-IlG4dMD2B}k9Mjg-`enU`1;di%_`N$Ac#+LM$V z^0Y%@*TTfS)-)p{BXqMS;*(ZK9lmoX?^W*k*!`p@|DK#7t*k^sH>zYPj096tqY96X zh?LwnxW9d3w{^yAX)aa0NL3;}pqZ(i#hemT-|n12Ru*WhjC-pNoj)gmw8!5tKVL5i z^e1Pqir=ra1lP1b^h+9b@6(T0Zfen>UyAhyw_2WQll+4){4)HkkD4j?H@HNMF009D1CP(Ez9Y(FL-=;rOme(l)oAELA5E z^Gi!DN_KRBBQo5X*{tYt7fT~F9gxA`v$hJ$XGs8r%%3dIaX83HNlS-LK*{_ zZdj+&c9N@sUfjB6*zjk*jZv=a@jOhp-#NYXc4j>#OI#ahCC?QEVj5*=RyQZw257hX@7C?;X;WpP~}s3Q`BGh9AgRmNKKh@m%cr zmh8XU0cG=Nc(=5EdmHd;=i}sWjp~edn!#exZ4D3(fF|@2X#^WNVr+QNJXTYatbLOG z?tyWA>`zQ_EDl?(kLt~~wzh(H64!v% zO7v%TwuEJX)$GN-^2x-mvF^Rxhw3^i|DLy9HEa#fU8v<|drjJ2%BtCvrR zohcL_{?+uMNQaOpvoqdjoBuZOoAk5r9&B->LvG#Jx^ZKRmLCIz(crLzSRH51bpTaV zp(z+_5}R(1y~xoB$ZYfRgY6G=k?_7Tu6%I~i z;+xe?Jo0C=EiLrCu@-vuo&h;RknpEaK0i1Z4GiQ4+~(6-S4^tRpUPE2Qx2v!G>O#T z2B=Hi&bQE{1B%?HNRCMj*^avW)1at~PjT<_@MrMSlm-dE_{JnIkprTkL8~J4A!;}A zxAnE88L)VrerGyXVDZzy_snfVk*L;>tg!+}s1aW|lFHWfG3dcNGR>ikiWn(0gL7Sg zQ*dNvkQ`pEsfGSqf=y|kyTMxnf=a94L2n9^f9CPMidH8RLk6 z7&rt_(|@5fI9s0rqakT%J9ov$lLu$i4k~n*GMM3xtWVi^8R#Y&A5Rf7k)}EY=~)Z9VrU=1Oh-3p>`87dvenrGgZiD{8^z?g^Gd|pXMw{I0UZ;q_{}ujfWo4 z-SbL94Y-pEAr3Irgr6qHKK!a;1GklCso->SwhSRU2EV^OKww(cwVnowh(Hm>gc{y#8+meo z0dzX9A1PuvQ&W>_ap91*QkORV$iBXk4bd-6i%Wpe?d^=6MMWWeY^oioxKbQ^v*cZ~ zXp(M*M&=7Zc-lws$oKgM3Gz~5U|}WGe_mbZ-pUY0{n%oIR;_pF>401`bQbjEQ}gq~ zgivXyxV}WDt2=G7H~DbJ9JF$YU*Bz9SQ&sxCP2QcZ=dqGuV2UbX#UcYDRNd(RimIb za7xYMrF@&kCdf@ukmY*aEMC;V5P$_(yo;47y!>;vadvj5GxhX+r={rC&V7Q?x9ceb zy&Qb}Z$HvBH8`)V1-u&i5I2JOSR#-aWjA2X(lvAMtU7w6Ty+&FxwyGiq@DK3}^!>@)=p0h0n-zU%ygNtHbiCeRL)~ z@hN3GW1wm!AN+kM=?OZSSKq~ypKToSG6Dg+OP#c)A%8EEtg;O}zI9=6I>FF~M1F5F z=D=!Jb*Z;QmQ5^%g@2+;|FgD_sG%lw(9rt^zy*^d)=FrxWT?bJK*abQHpp33iCIdB zM?CrX%W=k55?GFx76ZwZ!u6GXG%>A`t9I=;Nl2>W%h_)FDsCPgYmr(F&DDI0lJP>r zYuY0X2-8?15Czm?dv1S=eM0gsP7>U+{lH^N$|TK-A%?XVKvK1((Z4Qif7?(SJc$3H zfdi&ZY8vpV;Agv(%F?QH6z|X1N!} z?O&qWl)fY2BP3tSkr}FAa1X6lQ0!`EG>qwx2J(;9%BelB~D;I9fzP^UbXKW^EyV)jjjMXwB<1!9ti z0>~_k#h%3o<%^l-EMqlvl9m9G0MK(lBc{jrDqwEz0TrdISpkNhi8S$QG`FMZ)t)@* zC03M|C}Ae6-Rgl7F;>D_gM~mBh|hDND#$VP@IH^YJ*Z$ccj^n6PfZk6nvudLAx1lV z5#aatASRZv1rDI8uJE$O?X~DyagFFt$HIX`@bjdFx!3{kmY}0;8(J8G#l#2+chhGn zUKI!bWEEveFwyA(8^akO+Cn0(i0s3os~_JV&Ua4x;fyu!30hT;Me|}QC=cyyMsECb z6de1yApGBnp@co!czUj`f*QpL797@hpJ~)Blt6fkxZLRq%W{Zsn0Hso+uhnapHe__@USKX+K(@e#Fq1!_jcm*wC}@edK>Sj>)x< zmV>Uvio0pws?PFlIcsuwBM;96))*MeQJ(2PjUdJ_V=&iP^i`-Bs|yFZ=BNyb2=a%q z#5E)Ymgy=;qqLo)&7vKXZ&OSD+%~*%Lz!%{e)4%7;K)~mTyv>M{aD&zyvxgL@{oY& z=}K|&um7nfuiVK!Qw1q!xmNIBKs+I~RG;>ii_UBJTPD{j9?ED%(JM`U73$PtgJ;!| zD$KC-RQ>TFwPm*$Bu>ih%S%C#*_#<({&7;$BmxQnRUv!t>a`RCQcOf2H;#mPw1|$6 z)#?|JpsEYfrNaF7O7IMT$mh(l+YBRB*Yz^IgGPGLA1;Pe#`=nh5=@a~ml=&Ky^tYa zocy{BN}bmT$P+YB3;;u2RfwYNajyuUdY04vo1UIlK5xtcIct-ge>7Bu(T!wRT1ULU zd%kLKRf_?&grf99AK@d+>O!ejzZ-Mouk9>VpjExW2xZEDUp%6ckLYW^kko=YtjY9mEmDS^(gPlI+ z09<34KxC*^6H})RC;8zi1r{cQlqP)as+Y|?x&FRAoy*8VQgK>`cZS;3)Bji@ECq>Yya+0aG!(JU ztaskMQP(~SBTh`W{y6{8qBc6Nqgs5G;(oGy0e_)jR+g6f*SHIZ_Uo!YTgP5?b)LsX zckZuol}Z(5k7jDXEu$B1iZC3K;-hl)qm75# z*9tls)nJ_+7uQ6$e!8e60%=!hF@wT9n@uDH)uIZC>M5`2v)5ga;bGV;{(;lC2D4FV z_Dvk3ZlEC$DOpPwAJEQE&#OiYVu_HSy{RI za!GLg{zGY3ljsjSdU|9Ro#CWX`kRD8Q_ZmoDgu==bVA=m^gg8|z+rZBm1^m1P_1)O zXvWDsbG_l^ityTqPP1igpE$5$RRK)mf0N=F;FtBKsiX^$ZGyi>OZ8DDPLMen{w9h$ zlfytKE}Z08FWvua;X3!Wjg?jGPa&6~dE=Xz#VU9oj=OgQ|1)QpMbiP%E~%t_qVK* zZ@MM33sb|!(GS83IqFU1f(}%rUKDC*AeHWz?~eTaN(Aw3?D2#g%f=}_NMC$~FX+&n z#{fW;_+ms>Ma{xpyN&Wqd4Gz~U*P9*whx96la|k7VtR9GkAFL@PgkB#+62uBFAd7UCIfjCFQ^*ClGCCL=>d)y z=tIUE1H2>$0N~oli0^CO#oAwvNfC z*VfU}XvToGg<|lL%;jL+ZkbCkCUUxKQlGVkGw!= zeim13`FXX)F}kr|R8-WyIPdo9PWYd1|E!kJCc!a4-o|t+>*Snxp^G#WE#=4)T*VF5 zWHS8FW}eRC;LM=`m+B>Hpt3Tt+ISOYXLYm*1XBx()~mt1X^umav&MO!8PYDE%5@$ zP>T472nVM8N(%X}nM7E@PGex_{m61QDTzLVzdQi)n)}bzH^~3rj9Z}nHEKy!duuC( z;kfCYOE5Uv?-4r#w8R+Twe&EqCugMWmICPk_Vfts2-r$)bk4tBWi zFm*PEc$>ViZw));*b70AJg9j@n++7)>O&3HB-7*F z9<{6un1XYA1bq5zma#{;R5c|2v74-NgV*ZI}l+DkVi{+TY>rQXrL$i@g6T zNMY~DGeq&?UwPIdNsUNB`jG_h_Ok#6smc^0S$sXEYJLoC(yRh;^I^1Wx$O9?|Ul~aAt~f7QE@yOVAfnUSb4UHO1ckeOT!D z+EE{&;=1sNaUT-yG}G}^z++0hlGzT3eW?1Uew*K1LjVAxd;-)?>H7EOalJ(kc!5aN z4xU`gmr43ZIdj5G_+8U%HH#(|^Hp+xt66B(p$*rnXOLu!k`NKfQqeII@uQR-$3*5=1{OSCV6GP`zm@39@y3BA74=%~NUjtFL8*}d z<(f3b{Yu`3hR?2Wt#5qRYj)47hT|-xM!~)O>pLZtQJGKgwvnK=w&*E-eZkbX5Q3ii&Unin}FU*3)iS8VrC|f>MN|iZOMVr$x@u zcwf+#rWDn|PiQFpRY=;}-r>H%W5|bv*7l?cn_Mg^5N@K3{7faO^7;&tbCJ9zj_dqQ zcpSbxWx$C7iV?y>Z@>K|JFiJu6(+bIKWse{QT<{aY;R1jiP1+CDy_-Gg&-JTD+wFl zwmYv#X~uK7Xca?itD4ggDX`W(bDXiEZt%T`)!gWURh)KWQ;JmCJ8LE~P-uGE1nZ8< z=BLD<8F9|>{mvF|V~9_mNk@E#9A(LEeErD^v3CdGOLygGPl!?Ng>=D`NCf_ueU$o) z!d+3_vIq%Ie7N$}p8L>cX`1!Yz0QmD2e$oG+PIOh`Ol4S0F!qk&gMAXP~zd0`2JEl zeDFx!b0xLhQN~1ejhE5LCa9uwdg^JUBNmQ=g=>avBwk!b>j;*}htV>19Q}Dw8}hw0 ztMC7iLtz+mFF$?dkbXWu(n7p))tWAa?!SpRcycBkMn+{A$US%WBYO-k|q1*L!1P;im|%NW*V0u z+Z4r;S{WPm1q~7h*xu!%JGJb^Y0^wis8eM&zB>Fb@D(FPivJlyq`HPH^~ebBEkGi? zJPWm^P(9lo*3XYf)ksZk6P3oeikhbfDOF~Xo;koO*K;$w&bBVkcyfIMZE(FEyP;>( z?YZ;@y>pFJBzvEEThf&p0?UflcWxD+4g=EHHMc|}P{F2>H}zG^Cg|qpl6gzgPCN366mL%L zrib&p>DLW=O+|$%XAdEiB?66tEd_s zzu@tl8#f#ZXul-GT%W0PWRWKkJFx5mbL35e_fK6G9G(G$AhH#kv-2o?$d>U6g}RvR zEp7ASXt4JET^hY9)5)3H*hs8p+!6NAHrLI_3CU4+X>!NNu5zntG1)vXZAj#xQ;ACh z=WBw6N{Y3YZdIf3)nmi@Qz-gKswE{iM|#g~#>BU;-)x=j2CM34YD385Jyrw)ZN%G+ zg=Ye8?~V`&2ssHD^kubN94V^;qzy}U)s3$k;{cmI0AVL70Tj434V)v^unRCM!>rF0 zgl?kwd?=6kZ}GYIv?u9!`{;&M(0MT2O%5uIc*n54JF+1%{@`_^LekgdXEtzHSbn*( z9ABaAo3p`$&sh%%+z2$rfP|y!U4?wbS|P2cFE~*!aVYE)p(8Vg&H^LiYHQZl*){U6 zoDJ=a+CaKYmew=~v_F&bF262$v-ahE0w)M)tcCQ|xo$^+lcdHPQ&OEznzO44lx!I? z*=Emx;1tNP?6q7IaI&VUNZ;LdlnO`#*J#$xLd8eOU=uTs`RFdapD$2yP{Nz^6O{77 zf{`P}WlL6C@3#aOEkC2fD|C3#~NOCiZ20GTxF_OEwoL{9#?uhrOLP@I~ALTq#)Khj=}*U)LR3L=Iiw0UoXThk$sY;j%}24p1g$$`lcO7BYfM}8DiRk)%}oUN zlB19%DC!X$Ud3ljFd`dLe#_ouB5smHg*p|0aKuZ|<8C=QS+Vn~^i1HhsjpG-gdBWN zdynspPP$=jB@%oRtI^ppmo-(bG8K=~Ar7qjK1OVUTDvfa7hz^S6Vfcaf#2G!Ys(mH z3D$4|7GwKqg6b!VdiPiwtCc}0F+=dLNNrp{FPE!WOg z`HiV{TDJ&o1+=e2AcMG+o|C@k!%M3L)o)G}R#D6pVPT|Ljlohw|MJ+_f!*Lun5W^d zvycoL$?e*ga!v@tMoTK=Zk!a8ns{S&0G3acU&;h!%xXlPZ;4<%8(dinbo%Tkb=GE4 zvlQ%6fTzGPKf@ag4tcB+wO1Qwtfx3K)kqZ8QKVH^LgLXzl}fDZ)5A+7t*XSkct`__ zI#omJOmShj5X+q@;YuK0ONhK0lJ4K}SO|Fbl}TAq<{gT|0H_g%ggZk4#&iZ z9)7r5K8;8+k@M}eD-W0xtRA6TwZbk@5wjZ&Zu4W)EEIO zhLGTf{gb{wT#stVdSY`UOdZwA#3VK(tzPk*zH9@|W(UAVI#5B-g?9v0{2`HT_|Vh& zLl>IUFQ3Fx&|xHMV#?rX&2CS4dWs~6Jry##&0O!G_Vs>|7*p6o^N>G}f_87&9K%5Z zEZ9ViNPQusx>Y7fe6%M_B`wCKw~O13ao|&t>bISlsHz{Hd5>`KgokTu|J_%FdQ)vl!>iKb2n!;*hC1x5Ck5 z_zjfiv(L|%i%eT`%Vv)PtTm|giPR>Bt0z@g%{Ih{nBr8S7&%0Zf8hQ*E6?CXcbl`O z!}f&9>qwFS1=A7sJ+4I!O`q{V~8WT>OvS7pYpj(&_Q^{my^&<>GdN zk=z5%a2_xib)yheozvE_nDW-&Wd&m>hr`$HJaJxNF%u8NS4;y7V9!_0pKIoU|7MRc z^I{U0!Huwt@O(%(U1sRv+CQK8dXI&?^YyHltlm)|ghD*L3yn-Fp#Mc0i=eazq-dg4 zit(p4Ep2+wIZ|{t%!?sV3`$&cdxeNEuF@7*qtHrAexhUd|ZXItEPVMcJMAqZl9pFO4Ldcv<{AZTDUq&SfR z0EhJ;Q_5vAmBdB zH-G{ZtUNnUw;DVa3bS$~95!vEI*-!#eVkaJsF14{(--ptUARgUV`3N5JQcxWl^L0m zb*x_pVjch|Nt`0}KBv~rK{tU?gG7gh(!D@1 z=Q4tCf0K}i$vP&|J8{RTN5(5WQP?5eP_hU-TVKqPvFVx|pt;9~W4ZfQP>`rWGI>2; zmGyn^h0QW~=*XBcw$j!N9A6ZK^MAZ6%^;qP24K3rGJ79DA90N3j*M$?*%7{jkEilJ z1fKj|tH0oOIA8xzvhYXvdI7g;q{8KUKrty2w|6dsw~LQ-q@jot{s^DX?o`-hpG|l< zg&Xlc)y&kG0PS8HX~b46psQxkysh*0U!&fx1RCsz*Lj!TD(8 zK4OSj<~PwKwMgQ8go0gWT>fJ)DXcFwTKIyKR`;^^x=e~%vo{hN0ReL#yCs@RiEUn|}>-yILgsxbYa{nY&Dy8&YtLvbw;Z}g@F zSYe6+k@i!-i7@h*m)bu6T)gY*;V=cEwJp$j@|WbSg^2i9CMOmN`WJ!g^(|CN6xVQR zF(WL%&VX2#Z1MdkZ5OvXuM9ZF^C39dDk>E6CLb~#h43w(dpVJ!c-ZF4RCK z#2|_qqK`d#eNU4_Qr+S$qAlUk%It~zw=letcl0Zhiu{gT?q6*jUFu!6T~Q;SF!7fg zD*X2m1Vk_V6x!;UdB6^G#Q8%N-H$X_zRg55VhIiBi%h@tlvk#Hkt?42Tm)n2No5cJ zVl-1Cu|stlG_)E$B$uhC#w+D*BTclX@hWj!&S|A2WtK4r>yG z8t$diNIcz>5GqTno)>v*&ddy@2$Y5B|3})P0r$oSzh@c%pftudlz;x=07T|Tj9-Ty z9UXDhk8DJL@?U}q@LmUsw?AOElYMs`ga#z;EVyaMY2eOh9^zHwiIDtvO9-#Z_!~pa z06s7%xf}QR_~#vf{B5@aZN(=-SLfrQ&y@0#^y0I&&n}rRze&OB*t9j>gny*ap7(z> z`>Dagbt4C^js8BzKwur)TBkC@)i|Z7?f_`W<(g@b%y0Na4o+x=ZY=pqKBvLNr)cSS z^jJ~p>0QfX+B}8J!&{!fd+h58uZCgE?|d2A`%ye);F=!C34f) z1WNJTuWoB=Q@$Fkl~|Nu#=-&rDSo6Wjd<|B0swSQAyFQ#sDf-Gcan=3KIMau`cOW1 zk1aPY0^A@?ASQ)cBRyu|RVK%O*A+0xhUT#EV!OJHuJ4sCc|uQriD6k|{}x*e1M_61 z3zS0|uHQ*m>4a@8R=gISrcih+8-TbcI8=?QoccJ$S82(TH>(2Lt%2<70icCo+X7kNn zw~K3JD=XWLWRuOMi)&<*Eqg_jk`dWLMv|gjmlQX$_vYvKFMJ;7{dt^oUeEJ-Er5p- z4FOI|akZ&32OZz7GohdO8=J^7sxPP!2$vrr*lBLR;rkAOY4Xfe0|1C=5&1Y!nWy+t zW*~TDlp=hSJ_9&v7|Z*s?_g?_8c55oW9DQC&A>Rb!1jjz{HG1!DqVhU6jzt0OFc|a z$RPCKHT%JXnsq176F~qL&`{uD%KXWT*^!GVzObvEfAM9@Wol1fM=;-bshF!Eilt&= zKsu+Kp!|#@_m@beQ^nRYS>K}wJA@RTwcNjYky23y&1sOv31uLHzhm4$5XGyEf$$1M zzja2L#HzeNA@Fic@T~^-dWtGqzF#z3cr-#hj-tCS(U$|{q(SEV1xv&`L|iDzW{K6B z)v{%owX)O)1nnLsM+;q?`+SIaO)s7$b9jD$_UOwmFbyF-LT^pTs*;H6U&GquZneq@ z07iGuTIx&Yj~mqO6hH}q!M4-q&CPxFfdpr4DSiEaa>+2VfZW$2xq<{8Ab4BKcrzBcK>m3Yrl_FOCN8rAt?I%Q#t2a_`~`%^{`-MH({cqX#)lx3QXUSj^SA0SRf5> zxF8=imo!?Jc&AUhjylYW1s=LG&Y1 z+g5vmI)EGYki!+ABa2>q zQ>0^PSP>Z5GNq;_?i>=*cB|2^!)r zUN!X`XKiIcn&nO>#4)AK^ZkGkb%g+EGc5@mBo@BXCr(IA4!ea@t5-=!#kT_=RIJl{ zqW&f;!;FN={$^@BUc^%V-gp~e&ky_L@BAdL-crS`LYr?X(1 z3^CtSmoR^d&(jKQ)O*8Nmjz@jU^e|aC*9)}NI!toF=WIbzb^com9lAjEc-j7|E|u! zLIzFNBG>B$AKqP#*tqS$mb2}9rwQ|3K6789h&lmcsuIjHP&tYwL$*f`yr2zKaGAY!Kn0TO$I67 zkLt6bim&E8+n_!#Sb9PUP_1SXj|C5NxU^4@%#uMB?flMw7uf<48|tZB?21gdI}DGb zcazsw9ik&c_@;u+pJ{yH9=&ykU-{L?o;>PnluuJJ`L`aJLK?RWN8~YPi@Y0}Db#xo zytW#*4G9sx90EV;s;zPN!N|85YzQL(htyMU@Ym$ew7Efh*^^+cA1ak}{wMn4+=&E{ z>ru67yqOvZ(F7o`-luB3OPVWr@11w#VI?h{GMtbkoBV@d^B_dJDwsX4|BATDvKHBv zR$g4_q{qPCsz)^brPFgoQ2mysqmz3b)|y{L^rU45x5d#)%a%j5oDnj7FOi@_=r@nw zI2LD1{whQEDjU~Vl&~=3yu*n#8cDjv*`d&4!7m8+=c*?i1bMoU_?h5StE}*zw=Hus zijqZ|oG^`xx{i4e%(Nyq*LPEZpKD(0(PtkApU3Lrvk(DVNeU^UGaDU&0 zgQvr6JXjYgL?8P{{qZMygt#^{TmTIh5a5Q7)!)}K^T7(VaZUtCahW?I7~v^p*Zpvh zWz{%iWDQDf+=!=&*C+CTqH4c>En@+LvM-mNlCA|9cVBLdtG{n=57)Se`un*^>ko~g zYvM*xOkboV+AvXi6!xsBu=9m}FSSc%x@H_8j}l-C&WwX38uawmDTm$P!#M$mUNG}y zr47F4R!_&l6SZh8MW1<2ye4grd`+|cvJ!amk$91#af_sK60!pT^|eM)|p7jM2xCtA>!JOf#01B6~=M5I(T?@&6+Sj+*^uBVhggC zT4w{1ytimz=V$j0)bZ9%$LT9H*?Zi0}ea6=ahtqZP)ljOtnurT;4BEtI z3^~k{^^jzoXCL|S2spN;!RcL%Fn$%ndA2&1st2F2IB+rA=5u_9M#w z<73!>P+x%a0CEGqDUWshhD7sL@K&BmqZ+WQGnzl^cUTh}4trRB(df>n-AA5=CNp{@73u%M?nymH-1H}dBb*aWAN)&+g1I}dxvwkR^M%O&aIt1+FTFG z*fEI~=(Wi|-k*w)7BdxPGa0EFrgTH(Y2Sz?WV_X%zC3CnH8fQx-1}0JM--|2P>kye z8}`~imap`O3Yvqcww9Y;J$`A!1PER9QuykxJU_h~n*TRc>5qkHJvvcSe021t8NPzn zk$dH<0El6n%XDVYNWWM1V6O;IsK}W_qM<+31mwiCLiRX1ScPIM0{hR-#<;4wA*@I@ z(T0}g)HoWswrd~^qv$9`3^gjY#^ywxvu#u)cf8ZM%=&JVHh$Um>T*@%B47B$+a?&V zzPexP#;7>jC8Uai64@NNzcBcZ8j#Uj57cbo2kCP3KNkpSeKz%vIzp16gH2kdoo+#k z75_&rWUaCrY7SPz=#MR;FQ)ERF&b~FiyrJMoI0GZd5O}4>$+pnt%l~x0#r0QX|YfM zXI?INnZfQTH_Ek^ey*L}Ef5Z?|A_FCAAll-YMQw1g%!UR1M}2%QYL1a1xmcirI}rR zaMV9TIVYKjBJqZy@4qRt4uSzr{8-3Us_wGd)L4`I>K7dV6S%@$q0Wq7?w- z>p?td;EjSfh0R*&Y{GN^aaY_)I^gnW@%heGW1Fruz%d^R^Y-qG;3$Rg!)-Qcw30I9 zS&J+X&i1ZK-&dP?dDXE%Q?UA3?W!_2XL`IF8bR~sDdez&BEO-v(gXpM_RN?VXwd4H znTVY#Z&bvDBC7I$Q`T&gr_65T!0p`It;#{QYycgq%wkI?_zK;ULT^1sou) zo61|AY|}yzq2zpXIA)`ZacWmn^S7D529wjulzT$MC?cbN*LRSkXkny3)6a_vj9BOk zw-R+!Q@x6ikdrgd7R%}uXLznkr~SLFIr7lS&vPX`H}NVqkE2ajqL*22we#Q|PbB`D z<@vc!XVhjqylnG)kB1+63n6p2BoS>*NTd~hTCLI-5Wolo;>rjFS{Ue=c=GSvmgApd zOmkzhv$_fmzcjCoCQ|n1@G$=F>)~O0hPs-ji}!mekwR~{#do^@RD@xl!WKB@3Y?s) z;uv9H&8#vkEk2gL{MA-q_>A`D3t=PeqS841vSnC~M}GA)fq0F(2WT+W=Si*f?f2=2 zYO2laR0}{ER^L_V8--h{{`H(I=)L>J2OjwhTkBVycKdwWCaFqamu6V{2BnwcZ^z!L z=NzgC7*LDDxUrw$#E=n-)|O-7Yia=&1)4;`o~1WPHvY^_3VUi{EtwUUKpk7EP~==>~M%6_&uS(XlvAx$^A1qT^02DGG<+%(LS^_mYW z#7PefZ6sP~Q_3*{!bE$MXpqCT{RGY;YC`D*(L>S7W0Wars`;Y+<&UK0=tL>;)}d z$Lmi?1SfJ{r6c0M5lx*+7DFA~bhC_fIvBGF>Lh0A6CC$q*WzD50)%e2A1xAh7|E(* znmTd1aZ_YR$sQptMjgk;l&(q>!%t&fElSEO3uZ=plBviGKy43F3jwT;L2I_ScT^v@ z<=q$S;}{I5U%46aZYn-c+Jc{s8U-|&h(5XB72$Is)416QCf{#1osGK8 zCciZUQOPC?uNwpMLxt*q{>Ks;encClCYLewE75IjMKgH z;+RgoVG_&%y$L$o(s`1QkT7v@bB3KUjg39P-E;Q5EE-;bdiTM&tCV4M9j9 zJ91epj3r}%bAr?CNIHH|>$zsqXwgbtKg&=$j3uc&DJXrGb z>TRu`I+!bfIWgsO0p2)rZP3V|IQ2y2qV^X@cDN%W!op6T`4n+z?Q!1x+OFMd`&7F8 ziTbeJAr+x@l#<>Aj^O2MRuu`vgTaI*3ykAcFa;EWQ5r%t4|>8J5iO<$K-5Iddkyw1 zAd_icgUCxU8)cJt{!)>oJN3Ci?*v%i0nJ6-1 zLb^7*AN7kiafD>o`bB*E8i3aM*1Kz$&CKuD+SMwO1+7Xcf{Fi_mCP|Od?`|;!oOoN z91#yehV)Dy%5@*V70-G0fQ6N-eLb8NixT|soZQ~Nv=Q|xV8Sd}g!xY8q*=9fwOAPd zVz>D7XV?!^mP52}XZZVX$y4u*nDv15O7|SqSLC2@%;14M}bK^Wk2R4cDgaS39H{pw`h9J zIlDi#Yhmm2X%2&#QG}fZY0A+p0VM!|9=Q6aJ?FI)ac)*81d6zlx&Y&^( zF64`>p8T@cjC5*xd>CXhw2Lg1{>R*41^H{Tl7CZ%&E zcKFD&vAwE}J)XT6H42D6p47#Vnn|-bC_ww~{K*F2#lUpe6GIY6h6ki=BL7D5Y3M;J z5H4Q(ps%xGa$Xj+ee9i=eRVZcIBp~M@09ezoHbPT{ zlM}e{nH?b%pZF=v@u>nIn@ua>`F(AWTdgIDhd+Vx4g4`gfM0+L@zm$^XwT{LayQDp zIF-8!+FX>1He}OE8ZHy#Aj^n{x6Ij4YeMFyctodp(Stk17FJ0t3+4E)OgT2<&j9{K zip(}GOCO-3FH@$Bb13i0TH%v21C?oOpRG2J_y&TvI`0+Bf=Ca4@I^=JT;0e9V15;! zSMFc?c+HgcfLBulHOv7V%>9=}m-K`K-=<+Q)Mi`T;Qf6fhTsmTeX zzX??6x408=r;s({OUxu~rL%MvrpQwNQ!pb>;y;VHowsm4D90TZ(65a@+cOe!PWWK1 z4YtUm8EPlU#QJ(rxXs~eP_)e{hB072!iYjIUn(&Rscbfqk{Rv3@##nap># zCA_NBVa^3sEEO}htTLCNMJnqFZD2jqc)J~_V&wu-{A|ym`qNkj%T-kWWFX&XtgF^2 zGpfe!a=q((BzpU5*S~ogeYx}_cn`>8XjXgG@_|iZ|4}sIzES#Nu}+e9vy*-NilzrT z;R;BUk9ycq$!KTVlpM1zm&n6Dm=4!%A;{?9pl$%Qmf9SRAxhP5F6(`oCeE!kod!%F zh|g8#ym_WFGpsIhUkI+}w6Ww`Y+WzApanl_qh z%gV{(wukuJsRP5U2MX#E)HVjYLJm9lqAzs2&i#b=HOK8BrM_Xjs88D~9mfOO;mOr+ zYCeM>&c%i9$=u*O?~`#X5}K;UTU2?xs`U3sT<}t;Z`)82x{iFO1!0&_@wbt-Z8S7? zLKm@1#Eo6Eq)AMiP058%JiIS_=lgpo3dbz#1St;%Q^n{y5X~P;BRE}5>es0%FdTf@ zAwR~C+{_2?uySIy7}}7Vg$2mQK(N5;1iJ&%%iBYp$-kD{$Fv7#92b`fCu?pVrwu{E zsax_s*a5?$IivMQL_8By{(VqI8i|Uo82brui1U@CEg4d}E!#Ho91v?#;fvOUn;{2G z+XrIfwx;TW#7G%hnKlai`UEGM7zQYI{Bs<=f}+I8pUHhC2~GNRbT0IYl6mu>K2%ol zba`P+%PcM&sUt0?&ufwzeZCtk<2#HWXQI%OI7Ny?=K`iik9cjC6Gjk*awdqq4 zb1fE}xW0}BKKmA)An-4||FzQK5jnzze8?XYeRg@_7vdexHI|HNbmU*eJxqu3LSyi! z>fG-02_kL`iut|>SZmr`d*bi*;7$9^56q%LWIZKrY5vhHU!SR9_E?*34=)=$JfRP( zuEVmu+uIKSn`<3LL|i965Mx?M&xP{x{y#cE0=0tgu4NHTQoM~3(f%)Ej2}n3*W@c& z`PIvFHAsWLba3+Wv4ht{LQ@8(*Ix4`Z>ViMb@CVK$C_MZ-pIZE6>@oAc>9xb?R68q z6@w8^>=h7C&Tvso|0ORYeY#wprrFqD zdV|7=&1~f5X7-~iMxQ4d`u=#q0_+9F5@OsHq0()S2PewqZqvtSE?q=WJMwnxwuvxG z7v4`vX;?7NqF0~}oIS+O_3@1DTK{`IQFN+y`FmP@e0LkPcBLpq9l8neazhn@$Mv9) z!V>9F^ybg-5Wea>%P+mvNiZ^1i!E~xMDZ#Z>RTP1k);?&sNG2dB_fUt5?gH(!fx^w zb$g$RwKF2X$T-O$wON`M$J$qDo9wV)w1#u$VqHPX0~75NtL#MS7Tt5`VhVaSg46We-pf^8xrRF1qje(SzD;q+cY-kkT#!i)Nhm&^A2sl26AYbXcustSlbZAvt)h- z<&_NAg%EF9P!#?Wk!Vo8ZNS03UVG}oD3=x(jC}$bx)KJanE&H}xwak#5%10)4#=Ny zm>~r&KkJ(4EB)@3vj1?@b-vrxKLopy+~H`LAMPEXH=z~7Z+M7ttwidBEnT$$_j}Yq zU)2CTqnbD#guoaNh+*!UV4Scl0U?(OazToQhOP3@pj~~O)J}%Po;7?n?{Oq24YuA^ zRLuhq(Wvq~d%M4!F&3~{h3xzCG+}x-Su8+wQoh(y?6zFw>Ej)B)HY|Qp97x9Bsa(b{bq~u zS3Ftv`j1N>pd#3rmJz&V$R@7eGI*P%mJn`J!PF*z<-TtKbrzv9xAdhL>bQn|mjb@a z+WkU50AcW9x3V(#3^S$K!JOsCzW!-ma)Dx% zGaQz5&RP15xvh} zMEq=DCpyA_+~KE37e9G}H7lZzUK0DA$xQ&~E1RF`8mbcQcM~v~u~;)aVL~@4lcH8O zo{p+SFwWv~zkR%R7#Iywy))&z$1E+)>I7%{SI;AQ+RNSDkzh^R2w^7!$LcihB|!Ch zhTai;Ap+VQyrY;z&t!TqE>yHOka%j=(9~JK`sYtjMw^vH6X|YvTTl+?Ubj$jjkDKHGQ}$wp|LrmB|;a4fE{aorp1<%!QO7nqS} z?*=o7G8HYncm|j4op=sln<(%(?R>3*qlmK(!G}LeR#%07^7FaIkx}w(IG_nz8)qWo zT=eEznv0{vb({c5P)x>kl=*P((0dLDVtxu!H3xTIYEI@fIn2M8BcT0NX{B&h41+h_ znBHPo`E+nA4Tc{n!ji#&66KfDYZI^1j62%7gW7Zh_~Eg#I9f20-&FJ|TkXk~x0`8J zUaE?@2?*J$Zw6;#&@Cy8v#zE5J{g#g4^kd?>wKQ{bP81G@#jz@g6-qyzq?n6-sQPN zl*a6#C%c{j{I&vQg1cA0fBhZkpbFId^PZAl zx^y%hfcV{)=MhXh9!izA)|Xrs-tdd`UOdO9sjeUkk=r+$2xqpeG}0GKw+j@r)O%qI zkfobO*fy#=zH{n3?bPTDJ>5E3-!Dz`@;vx4c(FmE;L*e)z?%PZKT^oTuj7gP%8>4u zQ>^QwBACnN%dV3J;o_(__E&+|552f-dS^X*<_bIO29pUIXJhSC2~Aq!P1&c7*?*~a zjy_fwXCfj4Rznx9Buc)SiA5eguraDd;B~Q9K!%UQb+_`e-$OtlNexuTrNL@_$sopj zDZdOGslQ@2+3$<8QKJ=xHb1UGb9TRZUwKjB-t$*&&czGDD&A8AVSa*R3!jEX@p)`1 zzRFpF5Cy*N`O9Am(wFF~yP@cCed-p zt)>X34kdBp-?DMCVRftL)5Z9t&L%kLgg;(Lhq9kuhpknR=Wt`fk0q$Qmb) zYkYX1NA8l3(kM`x5Eh&|l`NDlER5lXw%Uq?Vj5F8EYs|t$9zEW2d@V21JdicE~$jG zz%1V#9Q|5q%_0rj(J1_CdL8jNat#!o{Pd>C{(~xMKHR#wm1AndIhEMl(FIz0mnj=i zXsurpgqQtyyixhj?jHz~`u65#eCj(d6=%fPhQ01kv2=cPpG$7ssfDj6m7}=0zK4?% z_~_TTAFRg}0|TMCBsnqwF!OrDP(z0}W_YzVFK;u_QRfcWNQizhR7^ZY^H-iDKxvl1 z6qgPYu~oQpIJzKoCAV~pueDyL$aF@|vAg|7W7u3#Wng@Vph4IHq(z-vgY@T18QMiA z;y=}{c-ui@9+b|VZ~F;?AgZ|j!3v?cFup_)JP-%@aPa7QHf2YFG#nC8b!fPQd(*78 z9KpuwDAdxAn_FKZiJnMsH+|V}yZ!8Qb=$@&UsQkbz02y>Eyn1`)2ReYtKO^lb+Pz{ zhZ7S5&_rb7Ny!M^&x1Q~@sxB^d~brSDWbo#>9jqxV}qAEUv|_hP^s_*lUmaN5$+TR z%~tSyNe{NjwDjxRCz1_dxsBmMOfmPJkvN{`;VBDKD2RbX+BDTGeGAH>gYB%GoKqEa zNsZ*y;*93B2dt_J-8}c*hzeE)ezb72Fx2r+>E7$`l^<+Qn=57tso7ns7>T$}QQH5` zj)U0OE;TS;o8Uw)Yn-ChVw1!Z+roR>J=;k-o1UxXw(g=|M%y|fe;h`>j$iHi!HN^l9Me`VwlRH&omJAfK+thh z2Wg)5J)JX>Q8TyNW3B$aMGORogO9ipzLjmb%o`H2j;4OQu3+7qrBB(T&p*Dr`x?Ed z>GALlt5Uf^4KG=2_A#Np{xjrNX#T%89iWL5uB6`!%xnUWZ$RGQo@%!e@oKK^$`>n~m`$VoXZt$3 zm%gkH8OaxSTuvADZc$1{sd#+k?b&qn$?oNO^cm3rg=lx##DKy=b`Zm%QM<<{zujbe zEr!>XWj&ad>thI-sq~cgK2h(-DT4D>t|*OB%xF;2z>Eluh^33|p_`IRtn9j)@sM$g zJx2c;!fIO4*e%e}DHx9)1b#w%bm@5YXI~Hc``tH#CxfeDZC!i3%xSeFf%d9x|IIhw zY)(burAI&faiUgc(=uKicwrbxjIV|*w5XH4%V1<#J z%iZ#kp_C_!cQq0mwQ|{eD^@36=3OAE9XV#k)&D4F%lCWqE3cLMdoW2~k z0)v>)c^Q``x!_+^XaiaW=xRm&z(>J594oT%#sNIxQ#Nci3`MieUYA>G_D9~2or>g` zCw4{FzoCE`sYAUY zVa`R=`{+<=mQ;sd&!rM9k$zGXDcp9WKNIO9kD6%~p==3C=?+~mOxCQ~Iq+Z}aGTP& zsEus?H&M0SSln?wA=MJVs+n%a!4q>>zfNJCv$Wlu(`%VXkc)wloo@#pYgaPq3BWmy-i~nx_Hv>(l2<^r3mrxLpXbSnf8ml zXKW*QJV6VSD8mXYvrcEdvWatUr2uitCTMwGb#L&4%(&(lYlAAD zn0y=-?&V{Bvhq&JQAOX*$j&t(i2WdF7ThYv-Y zo00+F+sz|GH424QB_PowFf~@Q7=ik6mUNwkB$ukxH`PxXCJMZ!v%L}Bc-U0S)!x5j zb9Fs(dvtI&SJ+p5~^W1X87iUC|uC)E>+b3d?V2;@ZM5?gU-w4M8X50r?Nu zbZxb$027ZaxXty;z62~WgV&6ST}K7uQaK|F=MfPRXU}{_%ji=^OSR7dUEyWC+T41} zr&vNdLkbvJ3mU@UpO2s_X_?$h19SiS+yBcrF{HNK4@7~GY!iAl^&N+nA=uv4ndZTQ zCRm2k%9L99wJpaw!mD~+h^DK;t?y>WeP0mT5Hq97X4|fUFQ92s+fSb@;w_eTJIP|z znL`a?Mz6TYz)jhBpAI!I&k=fLHk$U%+o)ewIL~XEPQ5m`(0rM08q*O}f0Kww*svSTiAr zQ{%-+=Z6S>=7cs@1}c@}f24c&*z(x0-4Km;4FUsBeF$p0!^n5u?YauHVa*e-+tXmV zrr^mpJ|r*c$rUs5N`=J*xpBXPHanxGz$g5}Y43dJ<5lzdI|A`*SoJw|`+56A!iWz00>|Fwkw0aRxYo1e~tlyUUlI zxWXozC`=;-OV0%y|K9jsWB6Pv<&?KwNzUzSh4gPI5uB)Y0#osfhg}>`UW4jj54V;R zkMDP-xSakBzx;ep7{EqVWmB5K&!?^`iD;trf12kgeHu|yuJXnR<$X^ivvOW%5%{ae zsd>YjLca!>!BF@18>dJFp!~TM^GYV*XESkGCGINk@O0T-Xz*zyp7Tnud))+&|EVv~ z$|#QT3ZI$oLAvmu_l{bx0g#HMv8-zE<6!lq(4w?kzb-B=x{j`z8Ic~}V0$AboSCPf z8A9^;@F@M<$yDhYJTGQNX9AG!F=%3lh(-CWf+v3Z;k7Z)WD}nh&u{-3NTlyu?;WmZmcREKj}Ppq}x*af}$i!Jxyr9+p8 z1R+Uh;T`v{ayBkqtk}mQz#Kx#YDbb8S4_^aeLNJ@TwlX6hUdzz0 znQ0d&eTx2}`q&6z`TlDU^QJn7Yt8&-n||NCK6r_lZcQaVy28DLsCmjF~h`W56Br8GjSqxHIzcO!c5 zBo2W4W8wU&)o!soM^2_ls8!?BejCb$K7i4*(1?{u^?WJWBTq1<)xxppZ)`eedTtRF zwSpD&10@xc3gYh1DIx7RCRLqAfr=RMj^IKTQ6`q&yi%h{Ux8l3BDw;>fV~C7bb2`E zc|u195SE9G>iF#xb+d6kbfV8m^j|7Ylg8NdH9{m$#7q&cm!5~9Xn0p-gi4%l;wSft z^eeLtVQ>J+%C^$a*V8ed>Tre;4;NqngdRcQrii7k@qciN-|4Y+kkm9=p=XAE(Nz3l zAv5;T=kL0>`s}{N_dt-bnVDiO9k-jfd6F{cX70#F-pitzGpxpF#^EofhEeXAY63EndL$qMKQiv$_Gsl1&wZ9(n%_P7J-l#vbT&9s#??o*Buv*Vr;G3#= ziQ8Qkth(YKa)nI`@nl;tDk}03+7=hy;JcBcPo%2e4Mh}PVaFlaq0V_3y{8+^!Opxl zX9b}DBHF@veiYT0xJ7)f9^u&f?U|F%(lS&?syf~~x$*iyEqSM5!=<$ssSP$ZX7t(< zJ9K;M+HdkbRy@Em4)0O`#iUNwg#Efke=UPY&~iu(N1)T4k);42lq3u#9s^`1-%^WL zyS*#5oVsHC3vLIyX-ZR21HjCW0=VD={7xO}_CGIM9`qNci-PF$l{=k}#FliQ}_o=|vFzd6_9*0b}9BKqC>PDDyh}`k* ze0~_5);-B_mDxE4=4@9}Z{mndI2-7=*hwaS@mDUq+NnnGzK*^a>^i39K42}OP1_>4 z<%!*eI`wP5a$8ey>?Q-)q&EJlgWRa^(2hVpBl#&Q_x7jkBSU`P(OlC!=ov|dcZRlR zx%JY+A*(IeFrw%r*Rz55_C?;O$s>!vjiQu(Sg#|0j~A(m^7X2s@y7`gjqmKdG2i6? zPmuB2l!JSmA+hLSmqeN@kU|gcy^Y%OPLSeilrlIzDyNzByIy7(|5Z&sYwP=64ewJn zZ?GZ#h~8jI+h`<74cL>odY;@pzLopwXT=+yo_|)nMOgV=SxI zzjkv8$5)&TtMG-q~@T(#5NQ9J9_i=6Nf_ zt}GmLrqAZmG`dIwIB~lheiba*87bZy@gTI~_>z0ZlkTAT!^l}uv050GTj5W$m#srg zgLVop+Y^A!IQKh-zRKAy#k@cMHoV0%vUmIA@O|)mF)`+6FXp~Kw~(gc=Ghqw+6Qk^ zWW@ix+7}t8l`{=ucHbzZ*Y*$F8B00nf@lhQthgo{*&b~zKs{_Fv_KWseJiaSf{g)4 zSlr(}qHUSL_vLs125FOoi0vu{&YmnEFO0lC+t`e(HWEYf%Cjz+>8(tK&4+vTAjY;m z7zH_^>XdMGAXG~S?57DUi|OK&EHGn$*g}+2zc)?BT7{HB;qk~#^#QVN zicWQVIAYe;O5W2}wDLbugN4N09#d5lSkq!#908iNK6i2@TPY%no0uB8F21X%4{3B9 z&RrKvCJG>-;^|u}6DWndV_Q`_k%z6}G&wdw6x#ZJv6Bzk0Fp)r&oT9VKM>*IEK- zGEhQ+$<2cq^%HEW^7O!dwKg1@iBD%GCcAa1niqE8YASlFlcq+_Y0{?(o;cMi!)Si{ z-WLTcW`Ss`M^q7sgMZqy3{Z-Q*1vP%xq16IZms9w(dsJ>Vn}n`p|iv28KTCh(Ves7 z)b#X$1Wxu%FyGK7@12vKdB!NO5DU9X!c(J0Hnn@7%b%QwCBIuzz4zJemH)lN2Jn!w z6yVcbzVt$HOC&OM80yVQJo$Wvdg42s~#3XJD%#3<5?I`3R3sU?ay zFd6h@6dT#TZY2g1{$p1Wo=FRQb2y>EMOFFBcWd-9GpC0xUlX+zHjgq9Euwirsy@w# zRBA51+}-TNiUx_HdWR;(wWWL2;Az+rX{zBrd2ga>1Mm5acr<5Xe7T-wFY7Su+O?3N ziMR)*rhXu>=P7xR@S7o#m$%P@r~;>(n2Sfd4$SoM3x5(*5$2@i&npOp57j(y{EBQ* z@Jv}Cy(oju@&=E?V~9c2-OXIAbINIO*vZ&VV0xou&dB)Olg;R`LbS9X_h?~;1U;O| ziWaFE?9h^1%UxCSURto9BT#xxm#>38JH;^VJtm|j5>h&)k3;UXZvGnm@t~gYzC|*n zo&sJjW`0mlbmw&4SRj*t%c`ZHWi1m&r_Q$h;FATO&&i>k4%w$U!49F8e?N|?ed3v| z4|4oI`Tk=3;`}x5BRImD36y8F->d}XPm1{#8nu7Y#P)p{5g0x@{&&gS{(%#hk&TKoQA#?P*iZ;F%@H4 z%v1m5td=N4|0R50hpG7uf57)PM!-`nQXlp}K#iV>i6$22p3w`}AdX~2u5?5nRs#8D zIODDNQv1`hEeWE%yCNB5+?mp&P=c&z>63T}q$G|;&*1&F5isA)aks#ia%yVc0hVdy z39Paa57%*Gz&5zK=-UFH4u=<~M&To_&Z8dhZ*|MEQs=Ytz8t&HFgTU(ANe-8q?NHK zyZv2!rgZHx>x4a`*ipavh8pd9kn+>Z#KSR_t(s#Ue`!A zpiKagIv1fCtg~0|(xK@XX3POhfINJ2)eAH{u|2%|SSNswOx1raRY!1X|IvsDd<9F2 zXaz2+N*Mca`L8tJEx{M@a^CW|KK+WBjGOCIE5xxhXMjwz4qJ#&BNJcJwku31U=(BUr?A+osBwQ8J9B`>t~MGTPVCm*B2u@X8nL?;S1<-#FyiX#pv_T z&u~qQcVP=y9#;SVy5hqnIenF~5~_b3@>CwAJT2dRfke<`?8;I&nm)pLOuhN}&Q~#Y zu3UF_D(tSLSesYJopf2J;Ndl_Wt_-sqT}zNMntO@Ht#Tp#>M1!b+N|9OXHbjb_dJv z1gq(NjiYn(u=sRn%a8h7qbv(WdTco7bIt`0r%aCepM=%sJB5u^!mX4#!u*h2gV4j) z&%3T0tM!dLTMJadhf61l@7P9OxxjsezN!|(oNb?|k>o@buJFgY$-bc1|EWgK8$Qhl z3Fi(U-=8U(VvPg)Ho9`A8HaEFI6R-)T_;xCCd-_J@gUDOlp$r|T`dmDLo0#G#p4>$ z+{F*yG`;z`-XI7P1eC~vp>h)y$zcWCx3b#Y8|aBvfIQ~K_qv}QLb#9|<5-FdP-qPZPoxOa50y^=iGeEn7- zIY*Xcs^%3h<8!j4_~p?l@bppRpr^}%M-8X^tj+kG4I?q1?n`JDR%qnys6nd>h=T~I2Fjm6 zN@jHJ57qi?D+VI3v{{q0g2&h*zStD zo44~vYteX4_3P-WeUWGAc-b}q6u>{syHE4WD1Y}LrrT_L!D+U5h(7*t$A!`36Y^L= z@#WdyN8zVk$9?n58&_u8M%;aR+WI{Z#`ae-cd;m6-KtR0RbFj4f~5BQXO&;@`_td+ zMCBS-!kY~W&K$j5C$IAlhT)7cU}^!n8CC zm#K?8&yW7xeJ#a|2G|H`qg;P%SvN8YyCV4aHed7Q&vk?!ZcG|$oL~K)-HNK;td|F4 z-@(x$O<>#;`y_-DYHeC5HWIC7$r`sOwqzWAws<)2RUf(YNw_OKbhkcjoZJn{SS0g( z2-`mx&>Km@KmlXDp!mVAvy0c4f4UA;(nCfGqZz;!k`0%E6WdJu{Jh6?bO>1W z-jfLUEh_J^u|&+fe74(VW(=t8!yDI9l(rm?HAEEDyS)(B-t}xrK2#y#I)gYt;F%c*Z_!ETE(ul5&U}@22nONC z2aj5ci&Nh%8h0K2-n^JtZE3mKIopX6g;D?@T4axV)6E-*p!_7Gr$d}Bkq;*R1m>A| z&QHzh+^emP*i|r!4CbH^$-qq<#7vOP|2*voQg{pt4pNs$5oWP0CC`Ko>c9)s;m*8%K`STvk&Vy*SIOe=YOL#CW%r8v71tp zK2cT(J;>Of6{io1DnXsmp~A6ETLlio5iwqj?)9VQ1}6#{Ay;K!1X)SAIjFywcI6}? zG*Kqr;B`L6UeA)p{w1x-Hz@_HB;M+)Rm8S7b`TWtvl`vI#;~h;d9te@rBO2Crq`&x z1@ASn7_Vb3e&s?Os0e9$jhLlg+9)Rz!@24XU`=bPrR64NWJJM<#0T0G@fjdW!gmv==;2RHC$Gh>^*)v1u;71c}b z7qw>KjI>|^a*605nrrtBht914QCDtH-GLXWL-&@pZ+fpjdaoi-`Z{tiX7zhrp&*bHL zq>InhZ?>%N*XAL^+MX}T$K2vlKu{w1^k#vyn(%aj5WZT5ZT0wwOgbN-8;h{Y0%V!*C-&H`hp<&yBD>>}RSEB@Z zB+m(a6%KQ}f4uvUFRHTh>n^*Gj++Rihzebe^69E@zA~8Gc@p5g+F-W7eVy)L*J8ur zc~I3za==7$=g%&`Hq~Duhi*kV)wa>06ZGLqHy&if8XK=$?Eh(3miW?QRXhW6*}lss zH_^joNQTrk0c6q>3pCb=J3V^+I)e#=td`lzN&}?i-2?+wwO8vN4)$y>XfS#w8y!tY_?a<5hH7_FYB&3*TY3IO4zW)=_r$*-$5I#5_?Zhoe7Fp8Xi zM(2vg9d7UE1Q&gV<7F!PRnB=}zTxUzO>$?x^TQuj<`=UPkx1&~jdM+ja8ao<9Hces_cf{;PlU%m3s5_C@qP zeSrJjX{nPI#?4>s)j?N#*c_>^aCIOV!9oj2JncPQ!sQPOt8$DiiDhm&^RC zPu|#$!omU^d0xuXFR$IzqmcuMgE%~y#0f3oT}nfc=T4Lxj$ST zG5bJ;qzcLN#Br!Gj48^Pks(OGTq~U5Sc9Hs*dIVk+1KGC?q0rvY*{%i5(2n<<%5qS zJ4Sze>{1MX@6Y;`(?S(XEi5@>TyJA{f4JeSc{9(z%Q6>=?qT6&seAYBwqCF3U~Wi6 z#2Rg<-7Q$>N7P5gfSxH_G*@#RhZUw2Pa1@rOu~$bf0FPm$^`D zHvWenTwlNbxYp|Kx7+O>eMi^pO_@FMG#}7ATt@0QKa>h`(;kgJLb#lGZKf} z7!5+s7lpvl+|m#UQCb>cs({SoUaVQ)hDgy_Ju8Cc2=}3tsYnJJU}lu5j1jGkT7*jd z2j9&fesAi%-`(B){O6B9dE37AjgK&Zg*@C7))7(=LRC|gkwuhQg_54GYg?NN)=H6R z)*l`WfAZ3bs`}jgVe`> zX_}L~)QUwZ%>u&W{9+F15IF+t342fCc73`6s7Jnh!S5ckwfzW%9paAdWx6bn>(g9S zw7GY16c#`<3INwhFk$pjm5>pzac4kaA^S37e{|cA-do2J1gKMmw)K`VZCwe&cPgn9 zkyeNpJq1ZGb(vv(+`}XTPv@$&ntR{+S3fv=LWGaj7F9y=^b5=R#W|5!bHDY>D5_f6 zwlo~>$wZdPLVeq(nbLJwjNJfnTEx5p1(>;+R3-p-0f2qjJV_W_*KwYORhM(jV@AZ* z+V8$vOF2ocOHm?9F!u-`A}$l~aAk4MDusjyxwrB3bXy9SiPFK{9v;Wi1616dxTFIS zolX;13Af&?H|kr*1estKi8QkW2=J0Pz*!tR1ZZI-Knf5pb#9wQf-shX6hg`XS+cri zx1R1kGP4(3lw)&*y*q|N6l{ z_=B&2Q!;8@kh2XyqNE)W5=e?>W(4e`v9Q)<-&+rw&ji8KM8I(W;xdR57{UnC$AF+x ziXz0_sc%CFO3}=rry2$bgbb7=!yySE-AAIDyArd?ZXMkLn86bK@!Au|X~I%)y*e?B zW>l6DG|#hW8KbY8oo4>#3;NL)_?!yB^d8~e!-0y{lDRUwZ;qO7>1IbSdq#*TVhX6q z4oL5jSP`N1bd9y&B1^fyucgT7S0&#J-3?QEkL6scTN`niY8|aAJ+3QY>3al@X#oJV zv5rpRW8V$%OZb=@u^}BBI0A4$oX;gZY(xO)Oyum0Bv>%P2LV7>L}mn$2{EJ8s@;ZT zdLI4pnQ(+wB0Z;5ITN*J-}~vqUwm=*!3QNGZnxXt|GoO=o!@S45-?|}TuM!fF_MTp z)4~`k5Qj%#MCR$TAV9X}&h^SgEsS+Kvde^F`yi@QB_$z2A5K`bXhfH(991w( z3?%aKW*3sUHUyqd=tclZ1xFuXd6cYT9|k5^2+5=09-RwlDa_=7!Yp9Ia`X$DW2VSV zbDcyv(nkkj)fdV*J-z$lx900x|JLD*_{mSc_#gl3^Z)1%zAp0wJooLYN|*`EPfu5( zWS6<}L{Z9QFtQt_$`rzdDo*D`>-_rl>(9RUJS@3PW*s1d*$JZeOe$LHOgb;y!wyDR zP>wRYWr=8gHz5L+Oazh0%nTMvMnsgk#%Rkd;lpj5&mUNXh@}r7BMdUbyj9i5)~6EN zZb!Ylh-eX}=naZ0QI6Zb@B2$;JeS@&p%E6qLCjJnKx@}cv?-|p9ep#I`Sp8e&RKZP zU>I{fA=tyMUFQ0>**!7UX}QM(;8aVU>O71lrH8jq7kqs_l{PjFZtJiZh>+l-G#r>E z2%_6<#|(ocVvLA7b2t(}5Ddt)z<@;3&QyRI(gGm2>tGFF;VNO749YlZS)qkP9{@>M zaH>^`WLoboET8XqKA%-}+qUmso?@>^@a{TF;nPw`YjPdEGarQ>aocuptBj?n%q6tc zK=(0(G)gc_B1kDxW@7U2bP(0yjslVBs$hde!m&=owTxZB5-Z`cUOK6~yeNVr7eG+u z9Ae(w{mrLO%Zph`#Yj?`MS4r~0SQ_ujk1?y(~=LLKeir#J})vH#+EwU*b0WlQ1u+7! z4+QLp6=}+OXiz;@@AmH3H$Bn5{^HHs2irT>3h9rJ9mj2Xaba{s#E{$Djt^cq_8nNX zZBLM_mq&t>(|MlM1_EY(x|V4|R9(8vW7eHM1_IMwI2%0>jj~XYz|EP%sgEL{PdRBx6#UE*PyP z0wH9OiXb9K-KaV$ULlF&Xb2l7;_*5S=lyWYa$X`YJ|6pF(0dg}(LM8Of7x1O_F z^AtMa6l@ME^Q7z38V`>$*D?0KHycqeOS`SxI+U|43g%q-wt_va%!Fy&ibq&`B39wD zP!U!Lhlj@}-v)|KwSM&N6Of08_wdMVA2}$@kxQLPnGIcF^7Q$VClE)h4A<4m1c?ff zLI9e`6wzP4tn=yQt>4x)JAU`_@W#G+B*lL6lQ)0;(_j70H(xT0%dF%2V39cNeSC1PRsskjZ31kqvJYSc1ONZEbS^u#W!YIC zw>js!?H&7^IC<(gGnIu!mCJ>VjRg%rwuF$NfrM!JF)bdzJp>Y>VIx_%0a#TS%2kz> zc_L!R-fPXvxX=i{!E6j^4h{YP_k};4r;=xdgOCwmISBwoFfyb>F*GxR0T#re+CvI| zey&eXuUXRNaizc|^KJCws7VP|UQaxl!$401*`D^i~B}&h(H2s_hC))U|z!ech~E- z$@TN`@Tr)qHmfgDKU+eSZ~68TfM_ z%Q}DfL=GVq5k8u7!YOfZBk$9aFbq(a4c8>C4N?U{s0tycEvGaYR0Ki*j}$=Cs-k0m zb`d9B{=bYTtcy2=+b(wP#xi^vwg$u+0 z;%NZ`3J?>TR)Q7(=yZPJ??(epZUZbbJHQ*HMCKIXrqwCAM+d+X&w1*c&eR`@M4&Y3k z*f}ujz+6V{B$-!2K@3J3VAAe&j6Mb=S=I+440qRNqgya=VT$ae2o9>(>qt3GnKifv zNM^z?r@#7EzWQn*qJ7^x>U6qb;Zl~TYm-rznLU^&rbG~sGm&E(e!JI_Xqh~fa{0lW zgfRvhNJgd-0PbOT?Uw?2V}6v_qcb__$+PDtk^MOg3Syk?OEP(6s*S;)JV zB7zzQl2A(g(UJZLEXl~5Zl-*2+j;`oPf=&iQQ$`QYX^15RxZQ1?-onr377&;xQPpE_DVH+m zc||0i=B&VpF~BbmQv@0?GEXZ?5`$obhli;jwYM>?%ieTbp;auU9LGUK065GDJ=iG( z0YfMpQADO*IC8c7hGNo7E%Fk#O`u<&S|2qY6_Q9}%`y-SIb*?dTn zoWhiqfuj&DxBWQwSHd}Bw62K|gS1;Do)*4en;Dpmbs}O6?GbUBGcyGEfBdh%KA*qL zIlnni|J@(IBM4-dtRF7h-@e{6$k9!N3P?;6;W4yK^y#;RJl4}R&r2Uin5*L(*3)CO z+vr}+A%YP>O%nq~@;5bAHx2>;M*}qx@E9pG4IrSGeXK=fJEsk$j~K_OM~xZ>>pY`& z1t$nt=7JIe0+eK8NnuVBMi|y@-U=va59UllfN0o?w^1BDlze84oo^anx~uy-ck>N=xtZWe6*ekgQ-X&jwCriA>y=^ z+G?Ui4oL(P)S;yWA`gpP5&$t{=3ohcAQBWN@OhhD?X(iWtlHe74K>4Ic+>IS=bMCu z=`!a$&4jy9LILaD%*g3)zkB_i?WBj^Q5onIPZeh(`k%i%{_T$om@(04z4w^Q0|0AN z?adioIUGmlB=Z;!f#`15_df3T`!r4N-s`BxF@}{Xwc3{DZNH+9(ffGd8PF38cpKm) zCpI5r`a=hapsI*#LaPS@Vg z;ei-Y&00)zx_n3vLol4!rwoaOgpjhDMIS5(;Yj45Xk2TL;BJBL!I&fiTS5WK;f{f7 znj|p_gh`feM#-6Y3`KquLhzgs2;}?U+z}FjGUMf8Y3gQR2j1?E{wvH(#g z7}l4W(9Bd*;pYQ zXiPX>iO4mwL_g{qA~Pb)!wt-1)RB;;d28+N964DV3L3l$wxFQ!C^;ixE=-i~&DPUo z+DGq85td>el;>$#dL6wsKVDn!)4Cw}{ds@+?zU$noCpxWO5s&<8!*(V%}@Kjt`t|GSGKsfkn-CK~?}O?@D^4WNcYOCC4)x_Awr$Gp(uYX`_auS3+x0%E82WGq4CoCGg@5pv|N1*s zozLgbpTaF0?tgj&J!Y}c|JXc z4b!F;ipeR65=Vrp??+>dJj-|AK9dj^LCjx&bsmE`At4~BH*)~w1tlW^r@VN;|Neh# zN{NdOwcFiI`!MDd>cD8AQ<+bX#Rea@*wxg=TN>i`zFC3sy1)4rcpN)Kb6X@yd+kYl zUh}v?+gl{1w8~-uotZAD4Zr>SIv7$2QchWiuIGg^?R(v?Efw#UZnrwElgHkA%+uu7 z`Yt)m7(wQHKnjH@u`0p-Mj?RFr=S};6y_22s@@65+> z{K5PE^~$h6W0bdH(2k}!4EpN`bQp$20^HdFd%c;CA(S-Ymru84&7VHa=l3TJHf-(H zbwr230&l}mcmja%01CIzrWS_c0ulfKFw|=ugelMb{kP9Dueh$;7%{KK-P+!N{e7DX zH&s>Vq&YEje|omOc+SV=yeyY@X&s^a%hR#51*VlNFfJ1liUJTrM#R_*wYEcY`Rd~t z!fX!*Gl1W`g1;IQ{K-Fr|Mm;{VtV(d|6+Y3Al+|At-W@$h=+&u>#x_8QHbi5zf>V~GHnq^Zo!)PCSRpR!8sO%hO;Z}*zxKy>GRx%45JX7!yW6)<@ow72 zfB4;pK8!KAOduX(A5F`IvMe@yKMvqr*2&Zv^;TnDeY9Jm<>j?6TY>1VJ+V|r19-+s3f6Oc>g`J^$kR1{}Y>9LPawWRj-ZRXNv(mDdmN zWp3`;j{DP14RFSTnMx)haR6W_g#Y>{>)&1b|M$R1s zKu{k;MF&I>>Yr`hV^0FIfMk>T$S#U%b?rw(&~VhCT7GmKd$lbe@ff-)%Ek| zJMQ3g+_jrOJeHzN8+H;=HGUb#@eL2u$&h+Dt2be9xFqVPB4sKWT-&Yn-r7<3`^b!vr!t2b4(++5OtA00MMSvFls9@fO@J_E z*)BA?>X=Ys2@L>Z2}iFr6KNn#izE#QNkz31SW{F5ARfc#^7@Z|EC1vj)E%uiEekK{ zxQ({=s-Pj>Ni_opA-M6ab|ZJgV|NdH_b@N(l=9@RZf54g?{y!*5@_=RGN0BDZ0c?X zKF%W(#>4qy@OF4(2WwRs&;fL)>BuRWnsm@K;W8HpVip_j9wre$sxW#FRUvT4aT{EO z0gsm(l?>pBn)BQmV375I5&(qmB9JC2C7XL$G9yPDeHdXpd~wbK+Xe)04hYk`;& zyyiMa-PY1fQ8fej{itRmb!1o9Q6T_RPDO0g>MicqzJH5MNXg&atzpJWPfyPPAQ8yP z96cO?K!)7WaUcMSL^v^;22aHd$6h0_cbymU9<|=ev`H?@D&T(G`@`cFuBMLNkcf~N z(MFv&PIY6Cz56f@6r z5iS89Bn+`^X`QEasR*va2!R=xxE*zvg;1GNnGyvOWdOinb1*{)F&#FBFht-`1p*@~ diff --git a/binPicking/visual/FetchPreprocessing.py b/binPicking/visual/FetchPreprocessing.py new file mode 100644 index 0000000..dd58ed7 --- /dev/null +++ b/binPicking/visual/FetchPreprocessing.py @@ -0,0 +1,101 @@ +import os +import pickle +from PIL import Image + +import numpy as np +from torch import nn + +from detectron2.engine import DefaultPredictor +import torch + +################################################################ +''' +README + +# constructor +preprocessor = Preprocessor( + MODEL_ROOT = '{MODEL_ROOT}' +) + +# use + +preprocessed_feature_vector = preprocessor( img ) +# img should have shape => ( Height, Width, Channel ) + + +''' +################################################################ +""" +MODEL_ROOT + L model_cfg.pickle + L {cfg.OUTPUT_DIR} + L model_final.pth +""" + +class Preprocessor(nn.modules): + def __init__( + self, + MODEL_ROOT = './', + mask_size = (128,128) + ): + + # Load the config and weight of model and construct the predictor + with open(os.path.join(MODEL_ROOT, 'model_cfg.pickle')) as f: + cfg = pickle.load(f) + + cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth") # path to the model we just trained + cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set a custom testing threshold + + self.predictor = DefaultPredictor(cfg) + + self.mask_size = mask_size + + + ## img should have shape of ( Height, Width, Channel ) + def forward(self, img): + instances = self.predictor(img)["instances"] + ''' + instances.pred_boxes + Boxes object storing N object + instances.pred_boxes.tensor return => (N, 4) matrix + + instances.pred_classes + shape: (N) + + instnaces.pred_mask + shape: (N, H, W) + + instances.score + shape: (N) + ''' + info = torch.cat( + ( + instances.pred_boxes.tensor, + instances.pred_classes.unsqueeze(1), + instances.scores.unsqueeze(1) + ), dim = 1) + + masks = [ + np.asarray( + Image.fromarray( + m.detach().numpy() + ).resize( self.mask_size ) + # tiny decision + # .convert("RGB") can convert the mask into a RGB + ) + for m in instances.pred_masks + ] + masks = torch.tensor( np.asarray(masks) , dtype = torch.uint8) + ''' + N + number of instances idenify in the image + HS, WS + pre-defined number of the resized mask, default (128,128) + info + tensor shape: (N, 6) <- the six dim are : (x1, y1, x2, y2, classes_id, score) + masks + tensor shape: (N, HS, WS) + ''' + + + diff --git a/binPicking/visual/__init__.py b/binPicking/visual/__init__.py new file mode 100644 index 0000000..e69de29