From 6c941752a6d985b5cca334d1a6cca11af3110d73 Mon Sep 17 00:00:00 2001
From: CallumOrmond <callumormond10@gmail.com>
Date: Sun, 4 Dec 2022 23:01:18 +0000
Subject: [PATCH 1/2] ML work

---
 Tilt Bot.py                                   |   82 ++
 .../ML Coursework-checkpoint.ipynb            | 1303 +++++++++++++++--
 Uni/ML Coursework.ipynb                       |   27 +-
 Youtube Downloader.py                         |    7 +
 4 files changed, 1300 insertions(+), 119 deletions(-)
 create mode 100644 Tilt Bot.py
 create mode 100644 Youtube Downloader.py

diff --git a/Tilt Bot.py b/Tilt Bot.py
new file mode 100644
index 0000000..42b705b
--- /dev/null
+++ b/Tilt Bot.py	
@@ -0,0 +1,82 @@
+import discord
+#Ashdra#2712
+
+client = discord.Client()
+
+
+
+@client.event
+async def on_ready():
+    print('We have logged in as {0.user}'.format(client))
+
+@client.event
+async def on_message(message):
+    if message.author == client.user:
+        return
+
+    msg = message.content.lower()
+
+    if msg.startswith("£save"):
+        fileSave(msg[6:])
+        
+
+    if msg.startswith("£show"):
+        
+        for i in fileRead():
+            if "www" in i:
+                await message.channel.send("<" + i[:-1] + ">") 
+            else:
+                await message.channel.send(i[:-1]) 
+
+    if msg.startswith("£deleteall"):
+        fileClear()
+        await message.channel.send("Cleared") 
+        return
+
+    if msg.startswith("£help"):
+        await message.channel.send("£save - saves the text after the command") 
+        await message.channel.send("£show - shows all the saved text")
+        await message.channel.send("£delete - deletes the line according the number beside the command, e.g £delete 1") 
+        await message.channel.send("£deleteall - clears the all the saved text")   
+
+
+    if msg.startswith("£delete"):
+
+        delint = int(msg[8:]) - 1
+        counter = 0
+        saves = fileRead()
+        fileClear()
+
+        for i in saves:
+            if counter != delint:
+                fileSave(i[:-1])
+                counter = counter + 1
+
+            else:
+                counter = counter + 1
+
+
+    
+def fileSave(msg):
+    f = open("saves.txt", "a")
+    f.write(msg + "\n")
+    f.close()
+    
+
+def fileClear():
+    f = open("saves.txt", "w")
+    f.write("")
+    f.close()
+
+def fileRead():
+    f = open("saves.txt", "r")
+    values = f.readlines()
+    return values
+     
+    
+
+
+
+
+
+client.run('ODU1OTQ3NTIxNjk0ODkyMDQy.YM55Pw.XeISt8dc9f6MQrC0cstO32jlFfE')
\ No newline at end of file
diff --git a/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb b/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb
index 61b8a0a..a4c8910 100644
--- a/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb	
+++ b/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb	
@@ -2,19 +2,11 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "hello world\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "print(\"hello world\")\n"
+    "C1 = create_collection(10_000,100,10,50_000,0.1,0.01)"
    ]
   },
   {
@@ -23,25 +15,136 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Topic models - approach to modelling docs \n",
-    "# assumes document has many topics\n",
+    "# C2 = create_collection(100_000,100,100,50_000,0.1,0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# C3 = create_collection(100_000,10,1000,50_000,0.1,0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import sklearn as sk\n",
+    "import gensim \n",
     "\n",
-    "# latent Dirichlet Allocation (LDA) - Baysian Topic model\n",
-    "#    Defines a \"generative process\" for docs - a recipe for sampling new docs from a complex joint distribution\n",
     "\n",
-    "# Normally - infer posterior distribution from observed text docs\n",
-    "# This case - sample docs from prior distribution \n"
+    "### VERSION --- V x T phi --- each word has a distibution of topics for that word \n",
+    "\n",
+    "# got choice over multinomial - get index instead of list with chosen value\n",
+    "# https://stackoverflow.com/questions/36952419/vectorizing-numpy-random-multinomial\n",
+    "\n",
+    "\n",
+    "C1 = (10_000,100,10,50_000,0.1,0.01)\n",
+    "C2 = (100_000,100,100,50_000,0.1,0.01) \n",
+    "C3 = (100_000,10,1000,50_000,0.1,0.01)\n",
+    "\n",
+    "def create_collection(M, N, T, V, alpha, beta):\n",
+    "    # M = Number of Docs \n",
+    "    # N = Max words in doc \n",
+    "    # T = Number of Topics\n",
+    "    # V = Vocab Size\n",
+    "    \n",
+    "    # GENERATE THETA --- M x T\n",
+    "    alpha = np.full(T, alpha)\n",
+    "    theta = np.random.dirichlet(alpha, M)\n",
+    "    \n",
+    "    #GENERATE PHI --- T x V\n",
+    "    beta = np.full(T, beta)\n",
+    "    phi = np.random.dirichlet(beta, V)\n",
+    "        \n",
+    "    \n",
+    "    # for each doc \n",
+    "    DocTopicForWord = []\n",
+    "    DocWordFromTopic = []\n",
+    "    for i, Doc in enumerate(theta):\n",
+    "        \n",
+    "        if (i % 100) == 0:\n",
+    "            print(\"Done\", i, \"Docs\")\n",
+    "         \n",
+    "        TopicForWord = []\n",
+    "        WordFromTopic = []\n",
+    "        # for each word in doc \n",
+    "        for j in range(N):\n",
+    "            \n",
+    "            # 3. a)\n",
+    "            #For each word in a doc pick a topic from the doc \n",
+    "            topic = np.random.choice(np.arange(T), p=Doc)\n",
+    "            TopicForWord.append(topic)\n",
+    "            \n",
+    "            # 3. b)\n",
+    "            #For each word in a doc choose a word from phi - based on the choosen word above\n",
+    "            # gets vocab for topic \n",
+    "            \n",
+    "            # NEW\n",
+    "            \n",
+    "            phi_normal = phi.T[topic] / sum(phi.T[topic])\n",
+    "            \n",
+    "            word = np.random.choice(np.arange(V), p=phi_normal)  # + (topic * V)\n",
+    "            WordFromTopic.append(word)\n",
+    "        \n",
+    "        \n",
+    "#         # 3. a) --- Faster \n",
+    "#         TopicForWord = np.random.choice(np.arange(T), p=Doc, size=N)\n",
+    "#         # 3. b) --- Faster\n",
+    "#         WordFromTopic = np.random.choice(np.arange(V), p=phi[topic], size=N)\n",
+    "        \n",
+    "        DocTopicForWord.append(TopicForWord)\n",
+    "        DocWordFromTopic.append(WordFromTopic)\n",
+    "        \n",
+    "        D = np.array(DocTopicForWord)\n",
+    "        A = np.array(DocWordFromTopic)\n",
+    "        \n",
+    "    sum1 = 0\n",
+    "    for i in range(T):\n",
+    "        sum1 = sum1 + phi[i][0]\n",
+    "\n",
+    "    print(sum1)\n",
+    "    \n",
+    "    print(theta.shape, phi.shape)\n",
+    "        \n",
+    "    \n",
+    "    \n",
+    "    # A topic for each word, Found by:\n",
+    "    # For each word in a doc, get a topic index by randomly choosing from a list [1...T] using the P in theta for each Doc\n",
+    "    print(\"\\n\\nDocTopicForWord\", D.shape, DocTopicForWord[:1]) \n",
+    "    \n",
+    "    # A Word for each word, Found by:\n",
+    "    # For each word in a doc, get a word index by randomly choosing from a list [1...V] using the P in phi for each topic\n",
+    "    print(\"DocWordFromTopic\", A.shape, DocWordFromTopic[:1])\n",
+    "    \n",
+    "    a = np.random.multinomial(10, theta[0])\n",
+    "    \n",
+    "    print(theta[0])\n",
+    "    print(a)\n",
+    "    \n",
+    "    return DocTopicForWord, DocWordFromTopic, theta, phi\n",
+    "    \n",
+    "\n",
+    "collection0 = create_collection(100, 15, 10, 500, 0.1, 0.01)\n",
+    "# collection1 = create_collection(10_000,100,10,50_000,0.1,0.01)\n",
+    "\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 83,
+   "execution_count": 232,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "hello world\n",
       "Done 0 Docs\n",
       "Done 100 Docs\n",
       "Done 200 Docs\n",
@@ -52,16 +155,21 @@
       "Done 700 Docs\n",
       "Done 800 Docs\n",
       "Done 900 Docs\n",
-      "4.6947910804345103e-07\n",
+      "0.024929452872102225\n",
       "(1000, 10) (10, 500)\n",
       "\n",
       "\n",
-      "DocTopicForWord (1000, 15) [[0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0]]\n",
-      "DocWordFromTopic (1000, 15) [[428, 104, 322, 420, 21, 14, 14, 376, 428, 322, 371, 376, 256, 237, 371]]\n"
+      "DocTopicForWord (1000, 15) [[0, 7, 9, 9, 7, 7, 7, 7, 0, 7, 9, 9, 9, 7, 9]]\n",
+      "DocWordFromTopic (1000, 15) [[315, 33, 125, 144, 408, 35, 245, 164, 338, 418, 292, 125, 278, 448, 125]]\n"
      ]
     }
    ],
    "source": [
+    "print(\"hello world\")\n",
+    "\n",
+    "\n",
+    "### VERSION --- T x V phi --- each topic has a distibution of words for that topic \n",
+    "\n",
     "import numpy as np\n",
     "import sklearn as sk\n",
     "import gensim \n",
@@ -70,9 +178,9 @@
     "# https://stackoverflow.com/questions/36952419/vectorizing-numpy-random-multinomial\n",
     "\n",
     "\n",
-    "C1 = (10_000,100,10,50_000,0.1,0.01)\n",
-    "C2 = (100_000,100,100,50_000,0.1,0.01) \n",
-    "C3 = (100_000,10,1000,50_000,0.1,0.01)\n",
+    "C1Params = (10_000,100,10,50_000,0.1,0.01)\n",
+    "C2Params = (100_000,100,100,50_000,0.1,0.01) \n",
+    "C3Params = (100_000,10,1000,50_000,0.1,0.01)\n",
     "\n",
     "def create_collection(M, N, T, V, alpha, beta):\n",
     "    # M = Number of Docs \n",
@@ -153,39 +261,96 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 78,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Topic models - approach to modelling docs \n",
+    "# assumes document has many topics\n",
+    "\n",
+    "# latent Dirichlet Allocation (LDA) - Baysian Topic model\n",
+    "#    Defines a \"generative process\" for docs - a recipe for sampling new docs from a complex joint distribution\n",
+    "\n",
+    "# Normally - infer posterior distribution from observed text docs\n",
+    "# This case - sample docs from prior distribution \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 228,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Done 0 Docs\n",
+      "1.1543250549896507\n",
+      "(100, 10) (500, 10)\n",
+      "\n",
+      "\n",
+      "DocTopicForWord (100, 15) [[1, 1, 9, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]\n",
+      "DocWordFromTopic (100, 15) [[104, 301, 180, 120, 368, 423, 185, 425, 292, 234, 412, 368, 18, 351, 458]]\n",
+      "[3.49675567e-06 9.97055694e-01 8.52667030e-06 2.66547341e-06\n",
+      " 6.98583782e-04 1.68824722e-14 3.52534533e-05 1.12931011e-03\n",
+      " 5.30033610e-13 1.06646999e-03]\n",
+      "[ 0 10  0  0  0  0  0  0  0  0]\n"
+     ]
+    }
+   ],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 144,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#want postiror - data from smapling \n",
+    "# want --- plot, topic x-axis, number of occurances overall as y-axis"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 235,
    "metadata": {
     "scrolled": true
    },
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAD+CAYAAADVsRn+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAsSUlEQVR4nO3dfVxUVf4H8M8wxJOI4iAgDz6BCZpa06iJj/yYqK1X/thsf5WkJmprupW+VldUfFjUjTTFyDRXCftp7urPyrJ0xVEMhVR8QAzWZMBnUWLQdFXQYc7vj5a7DjMjQw4XGT7v16vXa+65Z8793sNtvt577j1XIYQQICIiAuDS1AEQEdHDg0mBiIgkTApERCRhUiAiIgmTAhERSZgUiIhIwqRAD6V169bB1dW1qcNoMRQKBTZs2NDUYdh05swZKBQK7N+/v6lDcXpMCk7i9u3bmDNnDrp16wZPT0+oVCr07dsXaWlpTR0aETUj/KeYk3jzzTeRlZWFDz74AH369MH169dx7NgxnDt3rlG3e+fOHbi5uTXqNh4GLWU/nZkQAkajEY888khTh/JQ45mCk9i6dSumT5+OuLg4dOnSBX369MHrr7+OuXPnmtXbtGkTnnzySXh4eEClUuE3v/kNrl69CgC4e/cuEhMTERwcDDc3N/To0QMbN240+75CoUBaWhpGjhyJNm3aID4+HgCwa9cuDBw4EJ6enggODsbYsWNhMBik7xUWFuKZZ55B27Zt0apVK0RGRmL9+vX17pdOp0PPnj3h4eGBfv364ejRowCAGzduoHXr1hbxnTlzBi4uLti7d+992+zVqxc8PDzQu3dvfPfdd2aXT2ovVXz22Wd47rnn0KpVK8yaNQtCCLz//vvo2rUr3NzcEBYWhuXLl5u13blzZyxcuNCsbPz48Rg2bJi0PGzYMCQkJCAxMRF+fn7w8fHB+PHjcfv27fv2xezZsxEZGQkvLy+EhoZi4sSJ+Pnnn6X1tZfccnJyoFar4eXlhb59++LIkSNm7WRlZaF3797S/mdlZd13u7WWL1+OkJAQeHl54ZlnnsH69euhUChw4cIFqc6RI0cQGxsLb29vtG/fHi+++CLOnj0rrZ8/fz7Cw8Px1VdfISIiAq1atUJ0dDRKSkrMtrV582aEh4fDw8MDUVFRKCgosIhHr9djxIgRaNu2LXx9fREbG4sTJ05Y9EdWVhaeeOIJuLu7Y+fOnXbta4smyClERESI559/XhgMBpt1PvnkE+Hq6iqSk5NFYWGhOH78uFi+fLn46aefhBBCTJs2TbRr105s3rxZ/Pjjj2LRokVCoVAInU4ntQFAtGvXTqSlpQm9Xi9+/PFHsXv3buHp6SnS0tLEqVOnxKFDh8SwYcPE4MGDhclkEkII0atXL/Hqq6+KwsJCUVJSIrZv3y62bdtmM9aMjAyhUCjEE088Ifbu3SuOHz8unn/+eREYGChu3rwphBDijTfeEMOGDTP7XlJSkujWrZu03bouXLggPD09xbhx40RhYaHQ6XRCrVYLAGL9+vVCCCFOnz4tAIjg4GCxfv16UVJSIkpLS8WKFSuEh4eHWL16tTh16pRYtWqVcHd3F2vXrpXa79Spk1iwYIHZNseNGyeGDh0qLQ8dOlS0bt1ajB8/XhQVFYmvv/5atG/fXrz11ls2+0MIIRYsWCCys7PF6dOnhU6nE927dxejR4+26LPBgweL7Oxs8c9//lM8/fTTomvXruLu3btCCCEuXrwovLy8xOuvvy4KCwtFZmam6NWrl9n+W/P5558LpVIpli9fLk6dOiUyMjJEhw4dBABx/vx5IYQQhYWFolWrVmLu3Lnin//8pygoKBAvvfSS6Natm7h9+7YQQoh58+YJLy8v8cwzz4jDhw+L/Px88fjjj4shQ4ZI2zp69KhQKBQiMTFRnDx5Unz++eeic+fOAoDYt2+fEEKIy5cvi4CAADFx4kRRUFAgTp48Kf7whz+Idu3aifLycrP+0Gg0Yvfu3aKkpERaR7YxKTiJ/fv3i44dOwoXFxfRq1cvMWHCBLF161azH8fQ0FAxefJkq9+/efOmcHNzEx999JFZeVxcnIiOjpaWAYiEhASzOkOHDhUzZswwKzt79qwAII4dOyaEEMLHx0dkZGTYvT8ZGRkCgFlCqqysFK1atRJr1qwRQghx5MgRAUCcOnVKCCGE0WgUISEhYvHixTbbnTVrlujUqZMwGo1S2Y4dO6wmheTkZLPvhoSEiOnTp5uVTZkyRXTp0kVatjcp1I1h9erVws3NTfzrX/+6b7/c64svvhBubm6ipqZGCPGfPjty5IhU5/vvvxcAxMmTJ4UQQsyePVt07NhRShJCCLFt27Z6k0JUVJR47bXXzMpmzJhhlhTGjBkjXn75ZbM6VVVVwtPTU3z55ZdCiF+SglKpNPtx/tvf/iYUCoWUOOLj48WAAQPM2vnwww/NksK8efNE//79zeqYTCbRtWtXkZqaatYf2dnZNveLLPHykZMYOHAgSkpKsG/fPowZMwZXrlzBiBEjMHz4cAghUF5ejvPnzyM2Ntbq9/V6Pe7cuYMhQ4aYlQ8dOhSFhYVmZf369TNbzsvLw/Lly+Ht7S3916NHDwBAcXExAGDatGnSZZT58+dLl4HqM2DAAOmzr68vIiMjUVRUBABQq9XQaDRYu3YtAGDHjh24cuUKxowZY7O9oqIi9O3bF0ql0uo2bO3n9evXceHCBav9c+bMGdy6dcuu/bm37XtjGDhwIO7cuWNxGeVeX3zxBYYMGYKgoCB4e3sjPj4ed+7cweXLl6U6CoUCffr0kZaDg4MBAFeuXAHwy/7369fP7M6uQYMG1RtvUVERnnrqKbOyuv2Wl5eHL7/80uw4UKlUqKqqko4DAAgKCkL79u3NYqw9Rmu3NXDgQLO268aYl5eHI0eOmG2rdevWOHPmjNm2AKBv37717h/9BweanYirqyuioqIQFRWFP/7xj9iwYQNGjRqF7OxsREZGAvjlR+N+6q4XQliUtWrVymzZZDJhxowZGDVqlEV7gYGBAIA5c+YgPj4e//jHP7Bnzx785S9/wZ/+9CeL6+/1EXUm9Z04cSJmzZqFhQsXYu3atYiLi4O/v/9926i7P7b6pO5+WqtbNx4XFxeLsrt37943Hmvt1HXw4EH87ne/w8yZM7FkyRL4+vriwIEDGDNmDO7cuWO2/XuTTW28JpNJ2o69+19XffVMJhNGjRqFxMREi3UqlUr6XHfA3p4YrW0rJiYGK1assFjXpk0b6bNSqYSHh8d92yJzPFNwYrWJoLy8HP7+/ggJCbE50BYeHg53d3d89913ZuXZ2dno2bPnfbej0WhQWFiI8PBwi/+8vb2lel27dsWkSZOwZcsWJCcnY9WqVfXuw4EDB6TP165dw8mTJ6X9AoBXXnkFVVVVWL16Nb799ltMmDDhvu316NEDeXl5qKmpkcq+//77euPw8fFBSEiI1f7p0qULvLy8AAD+/v64dOmSWZ1jx45ZtGcthtrBa2v2798PPz8/LFy4EP3798ejjz5qNsBrr549e+LgwYNm27bn3v8ePXpY9NO9fxvgl+OgoKAAYWFhFseBr69vg2LMyckxK6u7XHvMBQcHW2zr3rMQ+hWa7MIVOdSQIUPEqlWrRF5enjhz5ozQ6XSiX79+om3bttJA8po1a6SB5qKiIvHDDz+IDz/8UFo/ffp0aaD51KlTNgea61573rNnj3B1dRVTpkwRx44dE3q9XuzYsUMkJCSIW7duiRs3bohJkyaJ3bt3i9LSUnH06FExdOhQMWjQIJv7UztI+OSTT4rvvvtOFBQUiBdeeEH4+/tbXHefNGmScHNzE127drU5wFyrdqB5woQJoqioSOzZs0c8+eSTAoDYsGGDEOI/Ywq1169rffTRR8LDw0P89a9/FadOnRIff/yxxUDz7Nmzha+vr9i5c6c4efKkmDJlivDx8bE60Pz73/9eFBUViW+++UYEBATYHO8R4pfr/gqFQqxdu1aUlJSITz/9VAQHBwsA4vTp01KfKZVKs++dP39eABBZWVlm+5+QkCCKioqETqcTffr0sWug2dXVVaSlpYni4mLx6aefiqCgIAFAXLhwQQghRFFRkfD29hYjR44UBw8eFKWlpWLPnj3i7bffFiUlJUKIX8YCwsLCzNret2+f2X4cPnxYKBQKMWvWLPHjjz+KL774QnTp0sVioLlDhw4iNjZWGnzft2+fmDVrlsjJybHZH1Q/JgUn8e6774pBgwaJ9u3bC3d3dxEaGiri4+NFYWGhWb0NGzaI3r17Czc3N9GuXTvx3HPPiatXrwohhLhz546YMWOGCAoKEo888oiIjIwUn332mdn3bf14ZGdni5iYGOHt7S28vLxERESEeOedd8Tdu3fF7du3xauvvio6d+4s3N3dRfv27cX//M//iHPnztncn9r/oXfu3CkiIiKEm5ub0Gg0Ii8vz6Jufn6+ACD+8pe/2NVXu3btEj179hRubm6iV69e0kDzli1bhBC2k4LJZBKLFy8WnTt3Fq6urqJLly7SoGat69evi9dee020bdtWtG/fXsybN8/qQPPYsWOlu728vb3F2LFjpbuqbElKShL+/v7Cy8tL/OY3vxEbN25scFIQQgidTicee+wx4ebmJnr27Cl2795db1IQQohly5aJoKAg4eHhIWJjY8Xq1asFAFFRUSHVKSgoEMOHDxdt27YVHh4eIiwsTEyYMEG6K86epCDEL4PPXbt2FW5ubqJfv35i69atFn+TM2fOiJEjRwo/Pz/h5uYmOnbsKOLj40VpaanN/qD6KYTgm9eoedu+fTvi4uJw7tw5aQyjIbKzszF06FAUFBSgV69ejRChuWHDhiE8PFwaIG+ukpOT8cEHH5g9j0LNHweaqdm6desWzp07h+TkZIwcOdLuhLBq1Sr06dMHQUFBKCoqwtSpU9G/f39ZEkJzdffuXSxdulR6mC8rKwtLlizB5MmTmzo0cjAONFOztXjxYjz22GNwcXHB4sWL7f7e2bNn8corr6B79+548803MXjwYHz77beNGGnzp1AosHfvXsTExKBnz55YunSpdNcXORdePiIiIgnPFIiISMKkQEREkmY/0Fz3QSEiIrq/oKAgm+t4pkBERBImBSIikjApEBGRpNmPKdQlhEBVVRVMJpPdsz86CyEEXFxc4OHh0eL2nYgcw+mSQlVVFR555BGz+eJbEqPRiKqqKnh6ejZ1KETUDDnd5SOTydRiEwLwyzsVauelJyJqKKdLCrxswj4gol/P6ZICERH9ek5/naVmwnCHtqdc83W9dUJDQxEREQGj0QilUonf/e53mDBhAlxcHiwHb9u2DcuWLUNxcTG+/fZbs3fxEhE5gtMnhabg4eGBXbt2AQAqKiowefJk3LhxA9OmTXugdiMiIrBmzRqr78Bt6fz1M62Wl4e/K3MkRM0bLx81Mj8/PyxevBgZGRnS7bJTp05FTEwMYmNjpXfP1tTUIDk5GTExMdBqtfjkk08s2urWrRvCw8Pl3gUiakF4piCDTp06QQiBiooKfP755wCA3bt3Q6/X49VXX8W+ffuwadMmnD9/Hjt37oSrqyuuXr3axFETUUvEMwWZ1L62Ii8vDyNGjAAAhIeHIyQkBKWlpdi/fz9GjRol3U7r6+vbZLESUcvFpCCDs2fPwsXFBX5+frD1TiO+64iIHgZMCo3MYDAgMTERY8eOhUKhQP/+/fHll18CAEpKSnDx4kWEhYVhyJAhWL9+PYxGIwDw8hERNQmnH1Ow5xZSR6uqqsLTTz8t3ZL60ksv4Y033gAAjBkzBomJiYiJiYFSqURqairc3d0xcuRIlJaWQqvVwtXVFfHx8Rg7dqxZuzt27EBSUhIqKysxevRo9OzZExs3bpR9/4jIeTX7dzTXfcnOrVu34OXl1UTRPBxaYh/wllQi+/ElO0REZBcmBSIiksg2ppCfn4+MjAyYTCbExMQgLi7ObP2tW7eQlpYGg8GAmpoavPDCC4iOjpYrPCIigkxJwWQyIT09HUlJSVCpVJg5cyY0Gg1CQkKkOv/4xz8QEhKCxMREXL9+He+88w4GDx7coqfBJiKSmyyXj/R6PQIDAxEQEABXV1dERUUhLy/PrI5CoUBVVZU0FYS3t/cDTyBHREQNI8s/wysrK6FSqaRllUqF4uJiszrPPvssFi9ejN///ve4ffs2pk6dajUp6HQ66HQ6AEBKSgr8/PzM1l+5cqXFn124u7tb9IvT01svbnH9QPSAZPn1tHbXa90XwRw/fhydOnXC3LlzceXKFSxYsAAREREWt1ZqtVpotVppuaKiwmx9dXU1lEqltPzfn510xC5IvoqPqLdOY02dvWDBAuzatQtubm7o1KkTli1bhjZt2ljUq66utugXZ+dvo7yl9QORPZr8llSVSgWDwSAtGwwGi7l9srKy0L9/fygUCgQGBsLf39/iGYTmonbq7KysLPz973/Hnj17sGzZsgdud8iQIdizZw90Oh26du2KFStWOCBaIqL/kCUphIWFoaysDOXl5TAajcjNzYVGozGr4+fnhxMnTgAArl27hkuXLsHf39a//5oPR06dPXToUOnSmFqtRllZmaz7QkTOT5bLR0qlEgkJCVi0aBFMJhOio6MRGhqKzMxMAEBsbCxGjBiBlStX4o9//CMAID4+Hj4+PnKE1+gaY+rsv//97xg+3LFvlSMikm1EVq1WQ61Wm5XFxsZKn9u1a4ekpCS5wpHdvVNn185p9Gunzv7ggw/g6uqKF198sfEDJ6IWpWXfpiMTR06dvXnzZuh0OmzevNlisJ6I6EHxQYBG5sips7OysrBy5UqsW7cOnp6esu4HEbUMnCW1EdS9JbV26mwXFxdUVVUhMTERJ06cgFKpxLx58zBw4EAYjUYsXLgQe/futTl19sCBA1FdXS1dWlKr1Xjvvfcstv8w9IHcOEsqkf3ud0sqk4ITaol9wKRAZL8mf06BiIiaByYFIiKSMCkQEZGESYGIiCRMCkREJGFSICIiidM/0bxt0zWHtvfCy23rrdNYU2cvXrwYmZmZUCgU8PPzQ2pqKgIDAx+oTSKie/FMoRE01tTZb775JnQ6HXbt2gWtVovU1FQHREtE9B9MCo3MkVNnt27dWvp869Ytzn1ERA7n9JePHgaOnDo7JSUFW7ZsgY+PD/7v//5Pzt0gohaAZwoyuXfq7BEjRgD4dVNnJyYm4vDhw/jtb3+LjIwMeYInohZDtjOF/Px8ZGRkwGQyISYmBnFxcWbrv/76a+zbtw8AYDKZcOHCBaSnp8Pb21uuEBuNI6fOrvXb3/4Wo0ePxrRp0xwRIhERAJnOFEwmE9LT0zFr1iykpqYiJycHFy5cMKszfPhwLFmyBEuWLMGrr76KHj16OEVCcOTU2aWlpdLnzMxMhIWFybMTRNRiyHKmoNfrERgYiICAAABAVFQU8vLyEBISYrV+Tk4OBg4c6JBt23MLqaNVVVXh6aeftpg6GwDGjBmDxMRExMTEQKlUIjU1Fe7u7hg5ciRKS0uh1WptTp397rvvoqSkBC4uLggODkZKSors+0bkDGzdqt4UvxcPG1mSQmVlJVQqlbSsUqlQXFxstW51dTXy8/Mxbtw4q+t1Oh10Oh2AXwZd/fz8zNZfuXJFui7fVMrKymyu8/b2xooVKyzKXV1dsXDhwvu2a+8Ygru7u0W/OD299eIW1w9kp2tWS3m8yJQUrF0vt3U75ZEjR9C9e3ebl460Wi20Wq20XFFRYba+uroaSqXyAaJt/qqrqy36xdn52yhvaf1AD6alHC9N/j4FlUoFg8EgLRsMBpt31+Tk5GDQoEFyhEVERHXIkhTCwsJQVlaG8vJyGI1G5ObmQqPRWNS7desWioqKrK4jIqLGJ8vlI6VSiYSEBCxatAgmkwnR0dEIDQ1FZmYmACA2NhYAcOjQIfTp0wceHh5yhEVERHXINiKrVquhVqvNymqTQa1hw4Zh2LBhcoVERER18IlmIiKSOP3cR2lpaQ5t7+233663TmNNnV3r448/xoIFC3DixAm0a9fOIW0SEQEtICk0hdqps4FfbnGbPHkybty44ZApKS5evIjs7GwEBwc/cFtERHXx8lEjc+TU2QAwf/58zJ49m9NmE1Gj4JmCDBw1dXZmZiY6dOiAnj17yr0LRNRCMCnI5N6ps2vnNGrI1Nm3b99GWloaNm7cKG/gRNSi8PKRDBwxdfaZM2dw7tw5PP300+jfvz/KysrwzDPPoLy8vDFCJqIWikmhkTlq6uzIyEgUFBTg4MGDOHjwIDp06ICdO3fC39/WrD9ERA3n9JeP7LmF1NEaa+psIqLGphANfeXXQ+bSpUtmy7du3YKXl1cTRfNwaIl94K+fabW8PPxdmSOh5qClv0+hyWdJJSKi5oFJgYiIJEwKREQkYVIgIiIJkwIREUlkuyU1Pz8fGRkZMJlMiImJQVxcnEWdwsJCrFu3DjU1NWjdujX+/Oc/yxUeERFBpqRgMpmQnp6OpKQkqFQqzJw5ExqNBiEhIVKdmzdvYu3atZg9ezb8/Pzw888/O2Tbtm5V/LXsucWxsabOXrp0KTZu3ChNl137vAMRkaPIkhT0ej0CAwMREBAAAIiKikJeXp5ZUti/fz/69+8PPz8/AECbNm3kCK1RNObU2RMmTMDEiRMfuB0iImtkSQqVlZVQqVTSskqlQnFxsVmdsrIyGI1GzJ8/H7dv38Zzzz2HoUOHWrSl0+mg0+kAACkpKVISqXXlyhVpUrnGYG/btfUCAwOxdOlSPPvss5gxYwaqq6sxY8YM5Ofnw9XVFX/+858xaNAg1NTUYMGCBcjKyoJCocBrr72G8ePHm7Xp4uICFxeXemNwd3e36Benp7de3OL6gex0zWopjxeZkoK1h6brvg+gpqYGp0+fxpw5c3Dnzh0kJSWhW7duFk/eabVaaLVaabmiosJsfXV1NZRKpQOjN1c7N1FD6oWEhMBkMuHy5cv4/PPPYTKZrE6dffbsWbOps+tuq/Yy3ObNm9G7d2/MnTsXbdu2tdh2dXW1Rb84O1szQLW0fqAH01KOlyZ/olmlUsFgMEjLBoPBYmpolUqFPn36wMPDAz4+PoiMjMTZs2flCE8W906dPWLECAANmzobAEaPHo3c3FxkZmbC398fycnJ8u0AEbUIsiSFsLAwlJWVoby8HEajEbm5udBoNGZ1NBoNTp48iZqaGlRXV0Ov1zvNKycdMXU2ALRv3x5KpRIuLi6Ij49Hfn6+gyMlopZOlstHSqUSCQkJWLRoEUwmE6KjoxEaGorMzEwAQGxsLEJCQvD4449j2rRpcHFxwX/913+hY8eOcoTXqGxNnT1o0CCrU2dHRUVJl4/qni1cuXJFGqzfsWMHunfv3hS7REROjLOkNoK6t6TWTp3t4uKCqqoqJCYm4sSJE1AqlZg3bx4GDhwIo9GIhQsXYu/evTanzn7rrbdQVFQEhUKBkJAQvPfee1KSuNfD0Adya+gsqf/92Umr5V/FRzgsJnp4cZZU22MKTv8+haZw/vx5m+s8PDywfPlyi3JXV1fMnz//vu1++OGHDxgZEdH9cZoLIiKSMCkQEZHE6ZJCMx8icQj2ARH9Wk6XFFxcXOx+wMwZGY3GB55jiYhaLqcbaPbw8EBVVRWqq6stnpp2dkIIuLi4wMPDo6lDIaJmyumSgkKhgKenZ1OHQUTULPE6AxERSZgUiIhIwqRAREQSJgUiIpIwKRARkYRJgYiIJEwKREQk+dVJ4c6dOy36yWEiImdk98Nr//u//4uoqCiEh4fj6NGjWLp0KRQKBaZMmWLxFjVr8vPzkZGRAZPJhJiYGMTFxZmtLywsxOLFi+Hv/8vbdvv374+XXnqpYXtDREQPxO6ksH//frz88ssAgC1btuCtt96Cl5cXPv3003qTQu0L55OSkqBSqTBz5kxoNBqEhISY1YuMjERiYuKv2A0iInIEu5NCdXU13N3dcePGDVy5cgVPPfUUAKCioqLe7+r1egQGBkpvCYuKikJeXp5FUiAioqZld1IICgrCvn37cPnyZfTu3RsAcP36dbi5udX73crKSqhUKmlZpVKhuLjYot6pU6cwffp0+Pr6YtSoUQgNDbU3PCIicgC7k8K4ceOwbt06KJVKvPnmmwCA48ePSwnifqzN7193BtMuXbpg5cqV8PDwwNGjR7FkyRKkpaVZfE+n00Gn0wEAUlJS4OfnZ+8ukDPTWy9u6PHB46mluGa1lH//BiSF8PBwLFy40Kxs8ODBGDx4cL3fValUMBgM0rLBYICvr69ZnXtfNK9Wq5Geno7r16/Dx8fHrJ5Wq4VWq5WW7bl8Rc7P30Z5Q48PHk8tW0v5+wcFBdlc16BbUgsKCrBq1SqkpKQAAEpKSvDDDz/U+72wsDCUlZWhvLwcRqMRubm5FoPT165dk84o9Ho9TCYTWrdu3ZDwiIjoAdl9prBjxw5s374dMTExOHDgAADAzc0NGRkZFmcQdSmVSiQkJGDRokUwmUyIjo5GaGgoMjMzAQCxsbE4cOAAMjMzoVQq4ebmhilTprS4l+QQETU1u5PC9u3bMWfOHPj7++Orr74CAAQHB+PSpUt2fV+tVkOtVpuVxcbGSp+fffZZPPvss/aGQ0REjcDuy0e3b9+2GIQxGo1wdXW6l7cREbVYdv+iR0ZGYuvWrXjxxRelsh07dqBnz56NEhgRkdys3fEIAG+//bbMkTQdu88UEhIScOjQIUyePBlVVVV45513cODAAYwZM6Yx4yMiIhnZdaZgMplw8eJFJCcn49y5c/jpp5+gUqkQHh4OFxdOtEpE5Czs+kV3cXHB4sWL4ebmhvDwcAwYMACPPvooEwIRkZOx+1c9MjISp06dasxYiIioidk90Ny+fXu8++670Gg0UKlUZs8Q1M6eSkREzZvdSeHOnTvo27cvgF8muCMiIudjd1KYNGlSY8ZBREQPgQY9eVZWVoacnBxUVlaiXbt2GDhwIDp06NBYsRERkczsHmg+fPgwEhMTcfHiRXh7e+PSpUtITEzE4cOHGzM+IiKSkd1nCn/7298wffp0PPbYY1JZYWEhPvnkE7ve0UxERA8/u88UKisrERkZaVYWERFh9p4EIiJq3uxOCp07d8a2bdvMyr755ht07tzZ0TEREVETsfvy0fjx4/Hee+9hx44d0pvU3N3d8ac//akx4yMiIhnZnRSCg4ORmpqK4uJi6e6j8PBwTp1NRORE7L58dObMGVy9ehURERGIiopCREQErl27hjNnztj1/fz8fLzzzjt46623sHXrVpv19Ho9Xn75ZentbkREJB+7k8KHH36ImpoaszKj0YgVK1bU+12TyYT09HTMmjULqampyMnJwYULF6zW++yzz/D444/bGxYRETmQ3UmhoqICAQEBZmWBgYH46aef6v2uXq9HYGAgAgIC4OrqiqioKOTl5VnU27FjB/r37w8fHx97wyIiIgeye0CgXbt2KC0tRdeuXaWy0tJS+Pr61vvdyspKqFQqaVmlUqG4uNiizqFDhzBv3jysWrXKZls6nQ46nQ4AkJKSYvGKUGqh9NaLG3p88HhqKa41qHZLOi7sTgrPP/88lixZguHDhyMgIACXL1/GN998Y/Z6TluEEBZl986yCgDr1q1DfHx8ve9o0Gq10Gq10nJFRYWde0DOzN9GeUOPDx5PZI2zHRdBQUE219mdFLRaLVq1aoU9e/ZI//IfPXo0nnrqqXq/W3sLay2DwWBxhlFSUoIPPvgAAHD9+nUcO3YMLi4u6Nevn70hEhHRA6o3KZSWlsLV1RUdO3bEgAED0KNHD6xbtw7nz59HQUEBHn/8cXh4eNy3jbCwMJSVlaG8vBzt2rVDbm6uxYuwP/roI7PPTz75JBMCEZHM6h1oXrduHa5duyYtr169GpcvX4ZWq8X58+exYcOGejeiVCqRkJCARYsWYerUqRgwYABCQ0ORmZmJzMzMB9oBIiJynHrPFC5evCjNeXTz5k0cO3YMS5cuRVBQEDQaDebMmYPx48fXuyG1Wg21Wm1WFhsba7Xu5MmT7YmdiIgcrN4zhZqaGump5eLiYrRt21YapPDz88PNmzcbN0IiIpJNvWcKoaGh+P777xEVFYWcnBz06tVLWldZWQkvL69GDZCIGqZmwnCr5co1X8scCTVH9Z4pxMfHY82aNRg7diyOHj2KuLg4aV1ubi66d+/emPEREZGM6j1TiIiIwMqVK1FWVoYOHTrA09NTWqdWqxEVFdWoARIRkXzsek7B09PT7EnmWvd7AIKIiJofu+c+IiIi58ekQEREEiYFIiKSMCkQEZGE79IkInoIpKWlWS2vO09cY+OZAhERSZgUiIhIwqRAREQSJgUiIpIwKRARkUS2u4/y8/ORkZEBk8mEmJgYs4n1ACAvLw+bNm2CQqGAUqnE66+/joiICLnCo0bGmTuJmgdZkoLJZEJ6ejqSkpKgUqkwc+ZMaDQahISESHV69eoFjUYDhUKBs2fPIjU1FcuXL5cjPCIi+jdZLh/p9XoEBgYiICAArq6uiIqKQl5enlkdDw8PKBQKAEB1dbX0mYiI5CPLmUJlZSVUKpW0rFKpUFxcbFHv0KFD2LhxI37++WfMnDnTals6nQ46nQ4AkJKSAj8/v8YJmhzqio1yh/399I5p3xmOp0bva6dwrUG1m7Lv5N62LElBCGFRZu1MoF+/fujXrx+KioqwadMmzJkzx6KOVquFVquVlisqKhwbLMnKUX8/fwe178zHkzPvW2Nryr5rjG3f77UHsiQFlUoFg8EgLRsMBvj6+tqs36NHD3z00Ue4fv06fHx85AiRrNi26ZrNdS+83Fa2OIhIPrKMKYSFhaGsrAzl5eUwGo3Izc2FRqMxq3P58mXpjKK0tBRGoxGtW7eWIzwiIvo3Wc4UlEolEhISsGjRIphMJkRHRyM0NBSZmZkAgNjYWBw4cADZ2dlQKpVwc3PD1KlTOdhMRCQz2Z5TUKvVUKvVZmWxsbHS57i4OItnF4iISF58opmIiCRMCkREJGFSICIiCZMCERFJmBSIiEjCpEBERBImBSIikjApEBGRRLaH14iagq2X+2DYYnkDIWomeKZAREQSJgUiIpIwKRARkYRJgYiIJEwKREQkYVIgIiKJbLek5ufnIyMjAyaTCTExMRbvTti3bx+++uorAICHhwfGjx+Pzp07yxUeERFBpjMFk8mE9PR0zJo1C6mpqcjJycGFCxfM6vj7+2P+/Pl4//33MWLECPz1r3+VIzQiIrqHLElBr9cjMDAQAQEBcHV1RVRUFPLy8szqdO/eHd7e3gCAbt26wWAwyBEaERHdQ5bLR5WVlVCpVNKySqVCcXGxzfp79uzBE088YXWdTqeDTqcDAKSkpMDPz8+xwdI9rtlc09B+v+KgdmzSO6YZZzieGr2vncK1BtVuyr6Te9uyJAUhhEWZQqGwWveHH35AVlYWkpOTra7XarXQarXSckVFhWOCpAZxVL87qh1/h7Ti3MeTM+9bY2vKvmuMbQcFBdlcJ8vlI5VKZXY5yGAwwNfX16Le2bNnsXr1akyfPh2tW7eWIzQiIrqHLEkhLCwMZWVlKC8vh9FoRG5uLjQajVmdiooKvP/++/jDH/5w3yxGRESNR5bLR0qlEgkJCVi0aBFMJhOio6MRGhqKzMxMAEBsbCy2bNmCf/3rX1i7dq30nZSUFDnCIyKif5PtOQW1Wg21Wm1WFhsbK32eOHEiJk6cKFc4RERkBZ9oJiIiCZMCERFJmBSIiEjCpEBERBK+o5ma1H9/dtJq+VfxETJHQkQAzxSIiOgeTApERCRhUiAiIgmTAhERSZgUiIhIwqRAREQS3pJKRORgtm61Bh7+2615pkBERBImBSIikjApEBGRRLYxhfz8fGRkZMBkMiEmJgZxcXFm6y9evIiVK1fi9OnTeOWVVzB8+HC5QiMion+TJSmYTCakp6cjKSkJKpUKM2fOhEajQUhIiFTH29sbY8eORV5enhwhEVEz5K+fabW8PPxdmSNxXrJcPtLr9QgMDERAQABcXV0RFRVl8ePfpk0bhIeHQ6lUyhESERFZIcuZQmVlJVQqlbSsUqlQXFwsx6aJ6N84Iy3ZQ5akIISwKFMoFL+qLZ1OB51OBwBISUmBn5/fA8VG93PN5pqG9vuVBm65wX9XfQM34KjtPoQc1ddz5861Wp6cnNzALTiQjb9zw/9u1xpU25HHRUPbkvuYlCUpqFQqGAwGadlgMMDX1/dXtaXVaqHVaqXlioqKB46PGq6x+72h7fs30XadQUP3uSn7yNbf+WE7Hh3ZVmPsW1BQkM11sowphIWFoaysDOXl5TAajcjNzYVGo5Fj00RE1ACynCkolUokJCRg0aJFMJlMiI6ORmhoKDIzMwEAsbGxuHbtGhITE3H79m0oFAps374dy5Ytg5eXlxwhEhERZHxOQa1WQ61Wm5XFxsZKn9u2bYuPP/5YrnCIiMgKPtFMREQSJgUiIpIwKRARkYRJgYiIJEwKREQkYVIgIiIJkwIREUmYFIiISMKkQEREEtmeaH7Y1Eyw/WY35ZqvZYzk10tLS7Na/vbbb8scCRE5C54pEBGRpMWeKRA1FzwjJDkxKRBRg/A9yc6Nl4+IiEjCMwWy+e7e8a6BMkfiXGzdzNBcbmSglolJoQnwx4KIHlayJYX8/HxkZGTAZDIhJiYGcXFxZuuFEMjIyMCxY8fg7u6OSZMmoWvXrnKFR82ErUHXhc/JHAiRk5JlTMFkMiE9PR2zZs1CamoqcnJycOHCBbM6x44dw+XLl5GWloY33ngDa9eulSM0IiK6hyxnCnq9HoGBgQgICAAAREVFIS8vDyEhIVKdw4cPY8iQIVAoFHj00Udx8+ZNXL16Fb6+vnKEaMbWNfav4iMa1I6tuzTKGhzRw8f2v9it750z7DM53rZN16yWv/ByW4e0z/GyhlMIIURjb+TAgQPIz8/HxIkTAQDZ2dkoLi7GuHHjpDopKSmIi4tDRMQvP7zJycmIj49HWFiYWVs6nQ46nU76DhEROY4sZwrW8o5CoWhwHQDQarXQarWOC46IiCSyjCmoVCoYDAZp2WAwWFwWUqlUqKiouG8dIiJqXLIkhbCwMJSVlaG8vBxGoxG5ubnQaDRmdTQaDbKzsyGEwKlTp+Dl5cWkQEQkM1nGFADg6NGj+PTTT2EymRAdHY0XX3wRmZmZAIDY2FgIIZCeno7jx4/Dzc0NkyZNshhPICKixiVbUiAioocf5z4iIiIJkwIREUmYFIiISMKkQEREEiYFIiKSMCkQEZGESYGIiCT/D6FY/JecIT7vAAAAAElFTkSuQmCC\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3de7xVdZ3/8dcbDggoiArkBRC1RBJQ6VQ2NeZoMqZWU9NkXhi7iaQ4aYppl1Err2PWTPbLSAVMo7zxi6wU1BpzKgkUuQSiCMIxjCMakIYD+pk/1jq2OZzLOmfvtfbh7Pfz8dgP9rrs7+ez9z589trf9d3fpYjAzMxqR49qJ2BmZsVy4TczqzEu/GZmNcaF38ysxrjwm5nVGBd+M7Ma48Jv1g4lpkl6SdK8aufTGkkPSTq52nlY1+fC381I+oSkxZJekfS8pO9KGljtvHZy7wGOA4ZGxDs624ik4ZL+UnILSS+XLP99OUlGxDER8eMO5tSnWR4Nkq6RVJHaIOlHkl6VtDm9LZL0NUm7VaJ96xwX/m5E0gXANcAUYHfgSGB/YK6k3gXlUJdl3U5mf2B1RLxcTiMRsSYidmu6pasPK1n36/JT7bSRaU7/CHwa+NeONtDG+/y1iOgPDAbOBP4B+LWkPp1N1srjwt9NSBoAXA6cGxH3RcTWiFgNfIykcJ2e7tdT0hclrUyPwBZIGpZuO1TSXEkvSvqTpC+m66dL+npJrKMlNZQsr5b0BUmLgJcl1bWybl9Jd0tqlLRK0r+VtHGZpDsk3ZrmtVRSfcn2YZLuSR+7QdINJds+JWlZ2hVzv6T90/WS9E1J6yVtTI82R7fy+u0raXb63J+WdGa6/tPATcC70iPiy1t47EFpN8sGSS9Iur2z37Ik7SnphyWv0UWSlG6blMb5nqRNkv4g6aiSx/5O0ukly2dLWp6+nosljWkvfkQsBX4LjE7bGCbpJ+nzekbSpJL2r05z/bGkzcDH22l7S0Q8CnwAGMr2f5OXS1qT/t3dIql/SZyj0+e2Md3n1GyvprUqInzrBjfgeGAbUNfCthnAzPT+FGAxMBIQcBiwF9AfWAdcAPRJl9+ZPmY68PWS9o4GGkqWVwMLgWFA35bWkRxkLAD+HegNHAg8A/xjuv9lwBbgBKAncBXwu3RbT+AJ4JvArml+70m3/RPwNDAKqAO+DPwm3faPacyB6XMdBezTyuv338D/S9s+HGgEjk23fQJ4pI3X/s0kXUG7kBzVPgx8K8N7FsCbm627A7gT2C1tdxVwWrptUvoenw30IjkqfxEYkG7/HXB6en8C8CxwRPrcR5J0VTXPoU+ax9B0eQzwAnBa+rovBr6QvmcHA2uA96b7Xg28mr5nPZre+2bt/wj4cgvr7wBmpPfPBpaRHKAMAO4Fvl/y2v4F+Of0/R1M8i2p6v/nduZb1RPwrUJvZHL09Hwr264G5qb3nwQ+1MI+pwCPt/L46bRf+D/V7DHbrQPeCaxpts8lwLT0/mXAAyXb3gr8Nb3/rrQQt/Sh9gvg0yXLPYBX0iJyDLCCpMurRxuv3TDgNaB/ybqrgOnp/U/QRuFvob1/au21bLbfdoWf5IPjNeDAknWfA+5L708CVjVrYxHwL+n90sL/38BZGXJoKvwbgZeAp4BLST4s3gs81Wz/y4HvlvxdzWmn/dYK/7eAn6b3/6fZ38ph6XuoNN7Mov4f1cptZ+97tb95ARgkqS4itjXbtk+6HZIit7KFx7e2Pqu17azbH9hX0p9L1vUESvu1ny+5/wrQJ+03HgY828Lzamr3PyV9o2SdgP0i4qG0S+g7wHBJs4ALI2JTszb2BV6MiM0l654F6slA0hDgv4C/J/mm1IOkiHbU3ulj1zTLY7+S5Qa29yxJ/s119P08NCK2azvtMhvRwnv2QMlyS+97FvuRfFuBJP9nS7Y9S/ItcU/K/7u0FriPv/v4LcnX7o+UrpS0K/B+4MF01VrgoBYe39p6gJeBfiXLe7ewT0vTvJauW0tytDqw5NY/Ik5oJWbz3Ia3cvJwLcmRbWm7fSPiNwAR8V8R8TbgUJKuiikttPFHYM/SfmVgOPBchtwg+XYQwNiIGEDy7UsZH1vqeeD1NHZreQxt9pjhJPk319b7mdVaYHkL79mHS/bp8PS+6fmPo/nbh/4fST7AmwwH/krywVCJ52HNuPB3ExGxkeRr8bclHS+pl6QRJP3FDcAP0l1vAr4m6S3pyc+xkvYi6VfdW9J5knaR1F/SO9PHLAROSE887g2c14kU5wGb0hO+fdMTeqMlvT3jY9cBV0vaVckQxHen224ELpF0KICk3SX9S3r/7ZLeKakXyYfXFpKulO1ExFrgN8BVadtjSUa23J7xufUn6Yf+s6T9aPnDpV0R8SowC7gyfZ4HkXT13Fay27D0JG9deiJ3ODCnheZuAi6WdFj6Ph8sqfmHRnseAUj/JvqkMcdKGtfxZ/fG0NF3AD8hKfZNz2smcKGS4a79ga8DP4yk3+dW4CRJH07/Zgan74+VwYW/G4mIa4EvAtcBm4BHSY6Yjk2LCsD1JCfW5qT73ExyUm4zyQnKD5AceT5FMuwOkg+NJ0j67ecAHRornub2Wtr24SQnLF8gKU67d+CxbybpBmkATk63zSIZwvojSZuAJSTfcCA5Ufh9km6XZ4ENJK9NS04BRpAUpFnApRExN+PTuxwYR9JP/jPgnoyPa8lZ6b/PAg+RvEalH0APk5ywfRH4EvDh9EN/OxHxA5L3+i6S9/kukpPcmUXEVpITt3+X5tMIfJfkxHNHfCUd9fMCcAtJn/7fR8SWdPt3SV6z35B067wIfD7NYSXwIZK/65eA+STf3qwMSk+mmFkXlw6l/GhEvK/audjOzUf8ZmY1xoXfzKzGuKvHzKzG+IjfzKzG7BQ/4Bo0aFCMGDGi2mmYme1UFixY8EJEDG6+fqco/CNGjGD+/PnVTsPMbKci6dmW1rurx8ysxrjwm5nVGBd+M7Mas1P08ZuZtWXr1q00NDSwZcuW9nfuhvr06cPQoUPp1atXpv1d+M1sp9fQ0ED//v0ZMWIE6QXLakZEsGHDBhoaGjjggAMyPcZdPWa209uyZQt77bVXzRV9AEnstddeHfq248JvZt1CLRb9Jh197i78ZmY1xn38ZtbtjLj4ZxVtb/XVJ7a7T8+ePRkzZgxbt26lrq6OM844g/POO48ePco7vr7zzju57LLLWLZsGfPmzaO+PtMVQdvkwt8B35n00A7rzrnxmCpkYmZdTd++fVm4cCEA69ev59RTT2Xjxo1cfvnlZbU7evRo7rnnHs4666z2d87IXT1mZhU2ZMgQpk6dyg033EBEsGXLFj75yU8yZswYjjjiCH75y18C8Nprr3HhhRcyZswYxo4dy7e//e0d2ho1ahQjR46saH4+4jczy8GBBx7I66+/zvr167nttuTywosXL2b58uWMHz+eFStWMG3aNFatWsXjjz9OXV0dL774YiG5+YjfzCwnTdc7eeSRR5gwYQIAhxxyCPvvvz8rVqzggQceYNKkSdTVJcfge+65ZyF5ufCbmeXgmWeeoWfPngwZMoTWLngVEVUZhurCb2ZWYY2NjUyaNInJkycjiaOOOorbb78dgBUrVrBmzRpGjhzJ+PHjufHGG9m2bRtAYV097uM3s24ny/DLSvvrX//K4Ycf/sZwzgkTJvD5z38egLPPPptJkyYxZswY6urqmD59Orvssguf+cxnWLFiBWPHjqVXr16ceeaZTJ48ebt2Z82axbnnnktjYyMnnngihx9+OPfff39Zue4U19ytr6+PrnAhFg/nNOuali1bxqhRo6qdRlW19BpIWhAROwz8d1ePmVmNceE3M6sxuRV+SbdIWi9pSQvbLpQUkgblFd/MzFqW5xH/dOD45islDQOOA9bkGNvMzFqRW+GPiIeBlsYmfRO4COj6Z5XNzLqhQvv4JX0QeC4inigyrpmZ/U1h4/gl9QO+BIzPuP9EYCLA8OHDc8zMrHtoaSriaoxn7xIu273C7W1sd5e8pmWeMmUKP/3pT+nduzcHHXQQ06ZNY+DAgWW1WeQR/0HAAcATklYDQ4HHJO3d0s4RMTUi6iOifvDgwQWmaWbWcU3TMi9dupS5c+fy85//vOwpmQGOO+44lixZwqJFizj44IO56qqrym6zsMIfEYsjYkhEjIiIEUADMC4ini8qBzOzIlRyWubx48e/MYnbkUceSUNDQ9n55dbVI2kmcDQwSFIDcGlE3JxXPDOzriSPaZlvueUWTj755LJzy63wR8Qp7WwfkVdsM7OuoHRa5nPPPRfo/LTMV1xxBXV1dZx22mll5+VJ2szMclDJaZlnzJjBvffey4MPPliRaZw9ZYOZWYVVclrm++67j2uuuYbZs2fTr1+/iuTnI34z634yDL+stLymZZ48eTKvvvoqxx13HJCc4L3xxhvLytWF38ysAl577bVWt/Xp04fp06fvsL6uro7rr7+e66+/vtXHPv3005VIbzvu6jEzqzEu/GZmNcaF38ysxrjwm5nVGJ/ctdrQfNKuKoz6MOsqfMRvZlZjfMRvZt3OmBljKtre4jMWt7tPXtMyf+UrX+EnP/kJPXr0YMiQIUyfPp199923rDZ9xG9mVgF5Tcs8ZcoUFi1axMKFCznppJP46le/WnabLvxmZhVWyWmZBwwY8Mb9l19+uSJz9birx8wsB5WclvlLX/oSt956K7vvvvsbHxrl8BG/mVlOSqdlnjBhAtC5aZmvuOIK1q5dy2mnncYNN9xQdl4u/GZmOajktMxNTj31VO6+++6yc3PhNzOrsEpOy/zUU0+9cX/27NkccsghZefnPn4z63ayDL+stLymZb744ot58skn6dGjB/vvv3/ZUzKDC7+ZWUXkNS1zJbp2msutq0fSLZLWS1pSsu4/JC2XtEjSLEkD84pvZmYty7OPfzpwfLN1c4HRETEWWAFckmN8MzNrQW6FPyIeBl5stm5ORGxLF38HDM0rvpmZtayao3o+BfyitY2SJkqaL2l+Y2NjgWmZmXVvVSn8kr4EbANub22fiJgaEfURUT948ODikjMz6+YKH9Uj6QzgJODYaO1XDWZmlpt2C7+kK4FrI+LP6fIewAUR8eWOBpN0PPAF4L0R8UpHH29mlsWyQ0ZVtL1Ry5e1u09e0zI3ue6665gyZQqNjY0MGjSorLayZPT+pqIPEBEvASe09yBJM4HfAiMlNUj6NHAD0B+YK2mhpPJ/iWBm1gXkNS0zwNq1a5k7dy7Dhw+vSHtZCn9PSbs0LUjqC+zSxv4ARMQpEbFPRPSKiKERcXNEvDkihkXE4eltUjnJm5l1RZWclhng/PPP59prr63IlMyQrY//NuBBSdOAIBmNM6Mi0c3MuqlKTcs8e/Zs9ttvPw477LCK5dZu4Y+IayUtBo4FBHwtIu6vWAZmZt1U6bTM5557LtCxaZlfeeUVrrjiCubMmVPRvDKN6omIX9DGmHszM9teJaZlXrlyJatWrXrjaL+hoYFx48Yxb9489t57707n1mofv6RH0n83S9pUctssaVOnI5qZdXOVmpZ5zJgxrF+/ntWrV7N69WqGDh3KY489VlbRhzaO+CPiPem//cuKYGZWsCzDLystr2mZ85Cpq0fSOOA9JCd3H4mIx3PNysxsJ5PXtMylVq9e3cnsttfucE5J/04yimcvYBAwXVKHf7xlZmZdQ5Yj/lOAIyJiC4Ckq4HHgK/nmZiZmeUjyw+4VgN9SpZ3AVbmko2ZmeUuyxH/q8BSSXNJ+viPAx6R9F8AEfFvOeZnZmYVlqXwz0pvTX6VTypmZlaELL/cnSGpN3BwuurJiNiab1pmZpaXLNMyH00yqmc1yZQNwySdkV5a0cysy/nOpIcq2t45Nx7T7j55Tct82WWX8f3vf5+mC1JdeeWVnHBCuxMktylLV883gPER8SSApIOBmcDbyopsZtaNNE3LDLB+/XpOPfVUNm7cWJGpmc8//3wuvPDCsttpkuWjqFdT0QeIiBVAr4plYGbWzVR6WuZKy3LEP1/SzcAP0uXTgAX5pWRmtvOr1LTMADfccAO33nor9fX1fOMb32CPPfYoK7csR/yfBZYC/wZ8DvgDcFZZUc3MakDptMwTJkwAOjYtM8BnP/tZVq5cycKFC9lnn3244IILys4rS+GfFBHXR8RHIuLDEfFNkg8DMzNrRSWmZQZ405veRM+ePenRowdnnnkm8+bNKzu3LIX/jBbWfaLsyGZm3VSlpmUGWLdu3Rv3Z82axejRo8vOr9U+fkmnAKcCB0iaXbKpP7ChvYYl3QKcBKyPiNHpuj2BHwMjSIaHfiy9eLuZWcVkGX5ZaXlNy3zRRRexcOFCJDFixAi+973vlZ1rWyd3fwOsI5mR8xsl6zcDizK0PR24Abi1ZN3FwIMRcbWki9PlL3QkYTOzriivaZl/8IMftLqts9q6EMuzwLPAuzrTcEQ8LGlEs9UfAo5O788gmf7Bhd/MrECZLsRSQW+KiHUAEbFO0pDWdpQ0EZgIMHz48ILSMzOrvKUvLN1h3aGDDq1CJonyfkuco4iYGhH1EVHf9FNlM7PWtDZyphZ09LkXXfj/JGkfgPTf9QXHN7NuqE+fPmzYsKEmi39EsGHDBvr06dP+zqksk7S9BbgKeCslF2SJiAM7keNskuGhV6f//qQTbZiZbWfo0KE0NDTQ2NhY7VRa9Pxfnt9hXY/Gyh139+nTh6FDh2beP0sf/zTgUuCbwD8AnySZpbNNkmaSnMgdJKkhbeNq4A5JnwbWAP+SOVMzs1b06tWLAw44oNpptOpjMz62w7rFZyyuQiaJLIW/b0Q8KEnpSJ/LJP2apJC3KiJOaWXTsR1N0szMKidL4d8iqQfwlKTJwHNAq6NxzMysa8vSyXQe0I9kkra3AafT8jQOZma2E8hy6cXfp3f/QtK/b2ZmO7EuO47fzMzy4cJvZlZjWi38kq5J//WQSzOzbqStI/4TJPUCLikqGTMzy19bJ3fvA14AdpW0ieRHW9H0b0QMKCA/MzOrsFaP+CNiSkTsDvwsIgZERP/SfwvM0czMKijLcM4PSXoT8PZ01aMR0TUnxDAzs3a1O6onPbk7j2RenY8B8yR9NO/EzMwsH1mmbPgy8PaIWA8gaTDwAHBXnomZmVk+sozj79FU9FMbMj7OzMy6oCxH/PdJuh+YmS6fDPw8v5TMzCxPWU7uTpH0EeA9JEM5p0bErNwzMzOzXGS62HpE3APck3MuZmZWAPfVm5nVGBd+M7Ma06HCL2kPSWPLDSrpfElLJS2RNFNS9svDm5lZWbL8gOtXkgZI2hN4Apgm6frOBpS0H8nVvOojYjTQE/h4Z9szM7OOyXLEv3tEbAI+AkyLiLcB7yszbh3QV1IdyWUd/1hme2ZmllGWwl8naR+S6RruLTdgRDwHXAesAdYBGyNiTvP9JE2UNF/S/MZGTw1kZlYpWQr/V4H7gZUR8XtJBwJPdTagpD2ADwEHAPuSTPt8evP9ImJqRNRHRP3gwYM7G87MzJppt/BHxJ0RMTYiPpsuPxMR/1xGzPcBqyKiMSK2kvw+4O/KaM/MzDogy8ndgyU9KGlJujxW0pfLiLkGOFJSP0kCjgWWldGemZl1QJaunu+TXH5xK0BELKKMUTgR8SjJzJ6PAYvTHKZ2tj0zM+uYLFM29IuIecnB+Ru2lRM0Ii4FLi2nDTMz65wsR/wvSDqI5Hq7pBdhWZdrVmZmlpssR/znkHTFHCLpOWAVsMMoHDMz2zlkmZb5GeB9knYluSjL5vzTMjOzvGQZ1XOlpIER8XJEbE7n6/l6EcmZmVnlZenjf39E/LlpISJeAk7ILyUzM8tTlsLfU9IuTQuS+gK7tLG/mZl1YVlO7t4GPChpGsnInk8BM3LNyszMcpPl5O61khaT/MJWwNci4v7cMzMzs1xkvebuL4Bf5JxLl7LskFE7rjz6O8UnYmZWYVlG9XxE0lOSNkraJGmzpE1FJGdmZpWX5Yj/WuADEeGJ1MzMuoEso3r+5KJvZtZ9ZDniny/px8D/B15tWhkR9+SWlZmZ5SZL4R8AvAKML1kXJBdQMTOznUyW4ZyfLCIRMzMrRjWuwGVmZlVU+BW4zMysurIU/n4RMa/ZurKuwGVmZtXjK3CZmdWYzl6B67RygkoaCNwEjCad+C0ifltOm2Zmlk2bhV9SD6A+Iip9Ba7/BO6LiI9K6g30q0CbZmaWQZtdPRHxOjA5vf9yJYq+pAHAUcDNabv/W3qhFzMzy1eWPv65ki6UNEzSnk23MmIeCDQC0yQ9Lumm9NvEdiRNlDRf0vzGxsYywpmZWakshf9TJP38DwML0tv8MmLWAeOA70bEEcDLwMXNd4qIqRFRHxH1gwcPLiOcmZmVytLHf3pE/E8FYzYADRHxaLp8Fy0UfjMzy0eWPv7rKhkwIp4H1koama46FvhDJWOYmVnrsnT1zJH0z5JUwbjnArdLWgQcDlxZwbbNzKwNWcbxfx7YFdgmaQvJdXcjIgZ0NmhELATqO/t4MzPrvCyzc/YvIhEzMytGu4Vf0lEtrY+IhyufjpmZ5S1LV8+Ukvt9gHeQDOk8JpeMzMwsV1m6ej5QuixpGMkF2M3MbCeUZVRPcw0kk6uZmdlOKEsf/7dJp2Qm+aA4HHgiz6TMzCw/Wfr4S6dn2AbMrPAvec3MrEBZCv9dwJaIeA1AUk9J/SLilXxTMzOzPGTp438Q6Fuy3Bd4IJ90zMwsb1kKf5+I+EvTQnrfF04xM9tJZSn8L0sa17Qg6W3AX/NLyczM8pSlj/884E5Jf0yX9wFOzi8lMzPLU5YfcP1e0iHASJIJ2pZHxNbcMzMzs1y029Uj6Rxg14hYEhGLgd0knZ1/amZmlocsffxnll4MPSJeAs7MLyUzM8tTlsLfo/QiLJJ6Ar3zS8nMzPKU5eTu/cAdkm4kmbphEnBfrlmZmVlushT+LwATgc+SnNydA9yUZ1JmZpafLIW/N/AI8GtgZURsqUTgtMtoPvBcRJxUiTbNzKx9rfbxS6qTdC3JNMwzgNuAtZKuldSrArE/ByyrQDtmZtYBbZ3c/Q9gT+CAiHhbRBwBHAQMBK4rJ6ikocCJuMvIzKxwbRX+k0iGcm5uWhERm0j6+k8oM+63gIuA18tsx8zMOqitPv6IiGhh5WuSdliflaSTgPURsUDS0W3sN5HkpDLDhw/vbLjMxswYs93yHblHtK5m2SGjdlg3arl7I637aeuI/w+S/rX5SkmnA8vLiPlu4IOSVgM/Ao6RdFvznSJiakTUR0T94MGDywhnZmal2jriPwe4R9KngAUkY/jfTjIf/4c7GzAiLgEuAUiP+C+MiNM7256ZmXVMq4U/Ip4D3inpGOBQkjH8v4iIB4tKzszMKi/L7JwPAQ/lETwifgX8Ko+2zcysZVnm6jEzs27Ehd/MrMa48JuZ1RgXfjOzGuPCb2ZWY1z4zcxqjAu/mVmNceE3M6sxLvxmZjXGhd/MrMa48JuZ1RgXfjOzGuPCb2ZWY1z4zcxqjAu/mVmNceE3M6sxLvxmZjXGhd/MrMa48JuZ1ZjCC7+kYZJ+KWmZpKWSPld0DmZmtazdi63nYBtwQUQ8Jqk/sEDS3Ij4QxVyMTOrOYUf8UfEuoh4LL2/GVgG7Fd0HmZmtaoaR/xvkDQCOAJ4tIVtE4GJAMOHDy80r1xdtnsL6zYWn0cXteyQUTusG7V8WRUyMeu+qnZyV9JuwN3AeRGxqfn2iJgaEfURUT948ODiEzQz66aqUvgl9SIp+rdHxD3VyMHMrFZVY1SPgJuBZRFxfdHxzcxqXTWO+N8NTACOkbQwvZ1QhTzMzGpS4Sd3I+IRQEXHNTOzhH+5a2ZWY1z4zcxqjAu/mVmNceE3M6sxLvxmZjXGhd/MrMa48JuZ1RgXfjOzGlPV2Tmtto2ZMWaHdXdUIY+uoDvMSjri4p9tt7z66hMzPa75c9/ZnvfOyEf8ZmY1xoXfzKzGuPCbmdUYF34zsxrjwm9mVmNc+M3MaowLv5lZjXHhNzOrMS78ZmY1xoXfzKzGVKXwSzpe0pOSnpZ0cTVyMDOrVYUXfkk9ge8A7wfeCpwi6a1F52FmVquqccT/DuDpiHgmIv4X+BHwoSrkYWZWkxQRxQaUPgocHxGfSZcnAO+MiMnN9psITEwXRwJPFpDeIOCFAuI4dteJ79i1Fbva8YuOvX9EDG6+shrTMquFdTt8+kTEVGBq/un8jaT5EVFfZMxaj13t+I5dW7GrHb/az71JNbp6GoBhJctDgT9WIQ8zs5pUjcL/e+Atkg6Q1Bv4ODC7CnmYmdWkwrt6ImKbpMnA/UBP4JaIWFp0Hq0otGvJsbtEfMeurdjVjl/t5w5U4eSumZlVl3+5a2ZWY1z4zcxqjAs/1Z1CQtItktZLWlJk3DT2MEm/lLRM0lJJnyswdh9J8yQ9kca+vKjYJTn0lPS4pHurEHu1pMWSFkqaX3DsgZLukrQ8fe/fVVDckenzbbptknReEbHT+Oenf2tLJM2U1KfA2J9L4y4t8jm3KiJq+kZygnklcCDQG3gCeGuB8Y8CxgFLqvDc9wHGpff7AyuKeu4kv+fYLb3fC3gUOLLg5/954IfAvVV47VcDg4qOm8aeAXwmvd8bGFiFHHoCz5P8wKiIePsBq4C+6fIdwCcKij0aWAL0IxlQ8wDwlmq89003H/FXeQqJiHgYeLGoeM1ir4uIx9L7m4FlJP9BiogdEfGXdLFXeitspIGkocCJwE1FxewKJA0gOdi4GSAi/jci/lyFVI4FVkbEswXGrAP6SqojKcJF/X5oFPC7iHglIrYB/w18uKDYLXLhTwrd2pLlBgoqfl2JpBHAESRH3kXF7ClpIbAemBsRhcUGvgVcBLxeYMxSAcyRtCCdnqQoBwKNwLS0m+smSbsWGL/Jx4GZRQWLiOeA64A1wDpgY0TMKSj8EuAoSXtJ6gecwPY/Yi2cC3/GKSS6M0m7AXcD50XEpqLiRsRrEXE4ya+33yFpdBFxJZ0ErI+IBUXEa8W7I2IcySy150g6qqC4dSRdi9+NiCOAl4Giz2v1Bj4I3FlgzD1IvskfAOwL7Crp9CJiR8Qy4BpgLnAfSXfytiJit2h8XIoAAAGCSURBVMaFv8ankJDUi6To3x4R91Qjh7Sr4VfA8QWFfDfwQUmrSbr2jpF0W0GxAYiIP6b/rgdmkXQ5FqEBaCj5dnUXyQdBkd4PPBYRfyow5vuAVRHRGBFbgXuAvysqeETcHBHjIuIokq7dp4qK3RIX/hqeQkKSSPp6l0XE9QXHHixpYHq/L8l/zOVFxI6ISyJiaESMIHm/H4qIQo7+ACTtKql/031gPEl3QO4i4nlgraSR6apjgT8UEbvEKRTYzZNaAxwpqV/6d38syTmtQkgakv47HPgIxT//7VRjds4uJao8hYSkmcDRwCBJDcClEXFzQeHfDUwAFqd97QBfjIifFxB7H2BGemGeHsAdEVH4sMoqeRMwK6k/1AE/jIj7Cox/LnB7eqDzDPDJogKnfdzHAWcVFRMgIh6VdBfwGEk3y+MUO33C3ZL2ArYC50TESwXG3oGnbDAzqzHu6jEzqzEu/GZmNcaF38ysxrjwm5nVGBd+M7Ma48JvZlZjXPjNzGrM/wEp3jDVNSE/GgAAAABJRU5ErkJggg==\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
@@ -200,37 +365,52 @@
     "\n",
     "for i in range(3):\n",
     "    \n",
-    "    c = collections[i]\n",
-    "\n",
-    "    indexs = np.arange(len(c[2]))\n",
-    "    indexChoices = np.random.choice(indexs, size=5, replace=False)\n",
+    "    DocTopicForWord, DocWordFromTopic, theta, phi = collections[i]\n",
+    "    \n",
+    "    vocabLen = len(phi[0]) \n",
+    "    topicLen = len(phi)\n",
+    "    numDocs = len(theta)\n",
     "    \n",
-    "    theta1 = c[2][indexChoices[1]]\n",
-    "    theta2 = c[2][indexChoices[2]]\n",
-    "    theta3 = c[2][indexChoices[3]]\n",
-    "    theta4 = c[2][indexChoices[4]]\n",
-    "    theta5 = c[2][indexChoices[0]]\n",
+    "    DocTopics = []\n",
+    "    for j in range(5):\n",
+    "        \n",
+    "        targetDoc = np.random.choice(np.arange(numDocs))\n",
+    "        \n",
+    "        Doc = DocTopicForWord[targetDoc]\n",
+    "        topicOccrs = Counter(Doc)\n",
+    "        sparse = []\n",
+    "        for k in range(topicLen):\n",
+    "            sparse.append(topicOccrs.get(k, 0))\n",
     "\n",
-    "    # theta1 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta2 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta3 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta4 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta5 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "        DocTopics.append(sparse)\n",
+    "    \n",
+    "#         print(sparse)\n",
     "\n",
-    "    N = len(theta1)\n",
+    "    theta1 = DocTopics[0]\n",
+    "    theta2 = DocTopics[1]\n",
+    "    theta3 = DocTopics[2]\n",
+    "    theta4 = DocTopics[3]\n",
+    "    theta5 = DocTopics[4]\n",
+    "    \n",
+    "#     theta1 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "#     theta2 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "#     theta3 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "#     theta4 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "#     theta5 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
     "\n",
-    "    ind = np.arange(N) \n",
-    "    width = 0.2\n",
-    "    plt.bar(ind, theta1, width, label='Doc 1')\n",
-    "    plt.bar(ind + width, theta2, width, label='Doc 2')\n",
-    "    plt.bar(ind + width * 2, theta3, width, label='Doc 3')\n",
-    "    plt.bar(ind + width * 3, theta4, width, label='Doc 4')\n",
-    "    plt.bar(ind + width * 4, theta5, width, label='Doc 5')\n",
+    "    X = np.arange(topicLen) - 0.27\n",
+    "    width = 0.15\n",
+    "    \n",
+    "    plt.bar(X + width * 0, theta1, width, label='Doc 1')\n",
+    "    plt.bar(X + width * 1, theta2, width, label='Doc 2')\n",
+    "    plt.bar(X + width * 2, theta3, width, label='Doc 3')\n",
+    "    plt.bar(X + width * 3, theta4, width, label='Doc 4')\n",
+    "    plt.bar(X + width * 4, theta5, width, label='Doc 5')\n",
     "\n",
-    "    plt.ylabel('Scores')\n",
-    "    plt.title('Scores by group and gender')\n",
+    "    plt.ylabel('Occurrences of a topic')\n",
+    "    plt.title('Occurrences of a Topic Per Doc')\n",
     "\n",
-    "    plt.xticks([])\n",
+    "    plt.xticks(np.arange(topicLen))\n",
     "    plt.legend(loc='best')\n",
     "    plt.show()\n",
     "\n",
@@ -241,32 +421,48 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 111,
+   "execution_count": 240,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[ 12 217 252 390  56]\n",
-      "[1.48406824e-043 3.25022340e-029 3.65485458e-035 8.72419832e-029\n",
-      " 1.01325918e-002 2.39844031e-107 6.48373859e-156 4.69666155e-043\n",
-      " 6.83613358e-008 1.71417346e-019]\n"
-     ]
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
    "source": [
-    "# # Q1 - Pick 5 Words\n",
+    "# Q1 - Pick 5 Words\n",
     "\n",
     "import matplotlib.pyplot as plt\n",
     "import numpy as np\n",
@@ -276,45 +472,49 @@
     "\n",
     "\n",
     "# P of word in Vocab \n",
-    "for i in range(1):\n",
-    "    \n",
-    "    c = collections[i]\n",
-    "    \n",
-    "    #c[3] = phi    \n",
-    "    indexs = np.arange(len(c[3][0]))\n",
-    "    indexChoices = np.random.choice(indexs, size=5, replace=False)\n",
-    "    \n",
-    "    print(indexChoices)\n",
-    "    \n",
-    "    \n",
-    "    phi1 = c[3].T[indexChoices[1]]\n",
-    "    phi2 = c[3].T[indexChoices[2]]\n",
-    "    phi3 = c[3].T[indexChoices[3]]\n",
-    "    phi4 = c[3].T[indexChoices[4]]\n",
-    "    phi5 = c[3].T[indexChoices[0]]\n",
+    "for i in range(3):\n",
+    " \n",
+    "    DocTopicForWord, DocWordFromTopic, theta, phi = collections[i]\n",
+    "\n",
+    "    vocabLen = len(phi[0]) \n",
+    "    topicLen = len(phi)\n",
     "    \n",
-    "    print(phi1)\n",
+    "    wordOccurs = []\n",
+    "    for i in range(5):\n",
+    "        targetword = np.random.choice(np.arange(vocabLen))\n",
     "\n",
-    "    # theta1 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta2 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta3 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta4 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
-    "    # theta5 = [0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,]\n",
+    "        count = np.zeros(topicLen)\n",
     "\n",
-    "    N = len(phi)\n",
+    "        # for each topic\n",
+    "        for i, doc in enumerate(DocWordFromTopic):\n",
+    "            for j, word in enumerate(doc):\n",
+    "                if word == targetword:\n",
+    "                    count[DocTopicForWord[i][j]] = count[DocTopicForWord[i][j]] + 1\n",
     "\n",
-    "    ind = np.arange(N) \n",
-    "    width = 0.2\n",
+    "        wordOccurs.append(count)\n",
+    "        \n",
+    "#         print(count)\n",
+    "  \n",
+    "    phi1 = wordOccurs[0]\n",
+    "    phi2 = wordOccurs[1]\n",
+    "    phi3 = wordOccurs[2]\n",
+    "    phi4 = wordOccurs[3]\n",
+    "    phi5 = wordOccurs[4]\n",
+    "    \n",
+    "    \n",
+    "    X = np.arange(topicLen) - 0.27\n",
+    "    width = 0.15\n",
+    "    \n",
     "    plt.bar(ind, phi1, width, label='Word 1')\n",
     "    plt.bar(ind + width, phi2, width, label='Word 2')\n",
     "    plt.bar(ind + width * 2, phi3, width, label='Word 3')\n",
     "    plt.bar(ind + width * 3, phi4, width, label='Word 4')\n",
     "    plt.bar(ind + width * 4, phi5, width, label='Word 5')\n",
     "\n",
-    "    plt.ylabel('Scores')\n",
-    "    plt.title('Scores by group and gender')\n",
+    "    plt.ylabel('No. Word Occurrences')\n",
+    "    plt.title('Distibution of words by topic')\n",
     "\n",
-    "    plt.xticks([])\n",
+    "    plt.xticks(np.arange(topicLen))\n",
     "    plt.legend(loc='best')\n",
     "    plt.show()\n",
     "\n"
@@ -375,27 +575,29 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 219,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "0.009198579418597084\n"
+      "collection 0 Doc sim is  0.09090280663511088\n",
+      "collection 0 word sim is 0.09806215084607471\n",
+      "collection 1 Doc sim is  0.0719097993113821\n",
+      "collection 1 word sim is 0.12935430487177807\n",
+      "collection 2 Doc sim is  0.0989470298162901\n",
+      "collection 2 word sim is 0.09090839904749326\n"
      ]
     }
    ],
    "source": [
     "# Q3 \n",
     "\n",
-    "doc1 = collection0[2][0]\n",
-    "doc2 = collection0[2][0]\n",
-    "\n",
-    "def cosine_similarity(vec1, vec2):\n",
+    "def cosine_similarity(doc1, doc2):\n",
     "    productTop = 0\n",
     "    for i in range(doc1.size):\n",
-    "        productTop = productTop + doc1[i] * doc2[i]\n",
+    "        productTop = productTop + (doc1[i] * doc2[i])\n",
     "\n",
     "    productBottum = len(doc1) * len(doc2)\n",
     "\n",
@@ -403,12 +605,40 @@
     "    return cosine_sim\n",
     "\n",
     "\n",
-    "cosine_similarity(doc1, doc2)    "
+    "collections = [collection0, collection0, collection0]\n",
+    "\n",
+    "# Docs \n",
+    "for col in range(3):\n",
+    "    \n",
+    "    DocTopicForWord, DocWordFromTopic, theta, phi = collections[col]\n",
+    "    numDocs = len(DocTopicForWord)\n",
+    "    vocabLen = len(phi)\n",
+    "    \n",
+    "    \n",
+    "    docs = []\n",
+    "    cosSim = 0\n",
+    "    for i in range(100):\n",
+    "        doc = np.random.choice(np.arange(numDocs), replace=False, size=2)\n",
+    "        cosSim = cosSim + cosine_similarity(theta[doc[0]], theta[doc[1]])\n",
+    "\n",
+    "    print(\"collection\", col, \"Doc sim is \", cosSim)\n",
+    "\n",
+    "\n",
+    "\n",
+    "    #Words\n",
+    "    docs = []\n",
+    "    cosSim = 0\n",
+    "    for i in range(100):\n",
+    "        doc = np.random.choice(np.arange(vocabLen), replace=False, size=2)\n",
+    "        cosSim = cosSim + cosine_similarity(phi[doc[0]], phi[doc[1]])\n",
+    "\n",
+    "    print(\"collection\", col, \"word sim is\", cosSim)\n",
+    "\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 289,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -472,27 +702,880 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 361,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\cluster\\k_means_.py:442: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  max_iter=max_iter, verbose=verbose)\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n",
+      "C:\\Users\\callu\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\pairwise.py:56: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.\n",
+      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+      "  dtype = np.float\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10\n"
+      "[0 2 0]\n",
+      "[[2, 0, 0, 0, 0, 0, 0, 7, 0, 6], [0, 0, 1, 0, 0, 14, 0, 0, 0, 0], [0, 0, 0, 1, 2, 0, 0, 0, 12, 0]]\n",
+      "[[-8.78345062  2.10410324]\n",
+      " [-0.70777457  0.02019144]\n",
+      " [-9.7354944  -0.42997697]]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "purity : 0.38\n"
      ]
     }
    ],
    "source": [
     "# Q4 Clustering the Documents of the Synthetic Collections\n",
+    "# Task 1 - implement K-means clustering \n",
     "\n",
-    "\n"
+    "# from sklearn.cluster import KMeans\n",
+    "import sklearn as sk\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "\n",
+    "# NOTES\n",
+    "# clustering like documents\n",
+    "\n",
+    "\n",
+    "DocTopicForWord, DocWordFromTopic, theta, phi = collection0\n",
+    "\n",
+    "# get the correct class/topic for each document - one with highest P \n",
+    "def get_class(doc_porbabilities):\n",
+    "    class_labels = []\n",
+    "    for doc in theta:\n",
+    "        cls = np.argmax(doc)\n",
+    "        class_labels.append(cls)\n",
+    "        \n",
+    "    return class_labels\n",
+    "\n",
+    "#Caclulates purity\n",
+    "def purity(k, cluster_Set, j, class_lables):\n",
+    "    \n",
+    "    # make confusion matrix\n",
+    "    confusion_matrix = np.zeros((k, k))    \n",
+    "    for cluster in range(cluster_Set.size): \n",
+    "        \n",
+    "        actual_topic = class_lables[cluster]        \n",
+    "        confusion_matrix[cluster_Set[cluster]][actual_topic] = confusion_matrix[cluster_Set[cluster]][actual_topic] + 1\n",
+    "    \n",
+    "    # get purity\n",
+    "    row_max = 0\n",
+    "    for row in confusion_matrix:\n",
+    "        row_max = row_max + max(row)\n",
+    "        \n",
+    "    purity = row_max / j\n",
+    "    return purity\n",
+    "        \n",
+    "    \n",
+    "\n",
+    "targetDoc = 0\n",
+    "vocabLen = len(phi[0]) \n",
+    "topicLen = len(phi)\n",
+    "numDocs = len(theta)\n",
+    "class_lables = get_class(theta)\n",
+    "\n",
+    "\n",
+    "k = 10\n",
+    "\n",
+    "\n",
+    "# simplify data d=10 ---> d=2\n",
+    "pca = sklearn.decomposition.PCA(n_components=2)\n",
+    "pca_data = pca.fit_transform(DocTopicForWord)\n",
+    "\n",
+    "cluster = sk.cluster.KMeans(n_clusters=k)\n",
+    "cluster_set = cluster.fit_predict(pca_data)\n",
+    "\n",
+    "\n",
+    "print(cluster_set[:3])\n",
+    "print(DocTopics[:3])\n",
+    "print(pca_data[:3])\n",
+    "\n",
+    "\n",
+    "\n",
+    "x = pca_data.T[0]\n",
+    "y = pca_data.T[1]\n",
+    "colors = cluster_set\n",
+    "area = 3  # 0 to 15 point radii\n",
+    "\n",
+    "plt.scatter(x, y, s=area, c=colors, alpha=1)\n",
+    "plt.show()\n",
+    "\n",
+    "p = purity(k, cluster_set, num_classes, class_lables)\n",
+    "print(\"purity :\", p)\n",
+    "    \n",
+    "    \n"
    ]
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },
@@ -506,7 +1589,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.12"
+   "version": "3.7.3"
   }
  },
  "nbformat": 4,
diff --git a/Uni/ML Coursework.ipynb b/Uni/ML Coursework.ipynb
index a4c8910..a8d6d0b 100644
--- a/Uni/ML Coursework.ipynb	
+++ b/Uni/ML Coursework.ipynb	
@@ -15,7 +15,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# C2 = create_collection(100_000,100,100,50_000,0.1,0.01)"
+    "C2 = create_collection(100_000,100,100,50_000,0.1,0.01)"
    ]
   },
   {
@@ -24,7 +24,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# C3 = create_collection(100_000,10,1000,50_000,0.1,0.01)"
+    "C3 = create_collection(100_000,10,1000,50_000,0.1,0.01)"
    ]
   },
   {
@@ -174,6 +174,8 @@
     "import sklearn as sk\n",
     "import gensim \n",
     "\n",
+    "\n",
+    "\n",
     "# got choice over multinomial - get index instead of list with chosen value\n",
     "# https://stackoverflow.com/questions/36952419/vectorizing-numpy-random-multinomial\n",
     "\n",
@@ -256,6 +258,8 @@
     "\n",
     "collection0 = create_collection(1000, 15, 10, 500, 0.1, 0.01)\n",
     "# collection1 = create_collection(10_000,100,10,50_000,0.1,0.01)\n",
+    "\n",
+    "\n",
     "\n"
    ]
   },
@@ -319,7 +323,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -331,7 +335,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -343,7 +347,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -426,7 +430,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -438,7 +442,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -450,7 +454,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deZwV9Znv8c9XQFFZlMWFRdBEDVxxQWJAJqhRJzFREjIa9xBjQmaiGRUjMYkZnTt6RzOGJI5OvIwayOaGMaLmJnGJoreJAoICQcfoEG1EbQgqEDfwmT+qGo9t9+nqperQ1Pf9ep3XqapT9fs9dbr76Tq/qvOUIgIzMyuPbWodgJmZFcuJ38ysZJz4zcxKxonfzKxknPjNzErGid/MrGSc+O09JF0r6Tvt2G4PSesldUvnH5D0pU6M6/9JmtxZ7bWh30slrZb0YtF9N4ljhaSj2rjN4ZLq84qpI9Lflb1qHUdZda91AFYcSSuAXYGNwCbgj8BPgBkR8Q5ARPx9G9r6UkTcm273HNCrk+K8BPhgRJzWuCwijumMttsYx1DgfGBYRLxcdP9bEkkzgfqIuKgz2ouITvldsfbxEX/5HBcRvYFhwOXAN4DraxvSFmsYsKbIpC/JB2OWOyf+koqIVyNiDnAiMFnSfpAc2Um6NJ0eIOkuSa9I+oukhyRtI+mnwB7AnelH9mmShkuKJonrA5IelfSqpDsk9Uvbfd8QRONQhqRPAN8CTkzbfjx9ffPQURrDRZL+LOllST+R1Dd9rTGOyZKeS4dpvt3S+yCpb7p9Q9reRWn7RwH3AIPSOGY2s+2Dkv4unf6btN9PpvNHSVrchnjPlPQccH+6/PR0/TVN45d0iKQFkl6T9JKk6dV+1pK+lb4PKySdmi77cLpt94r1/q4x5ibbTwFOBaal78Wd6fIR6c/lFUnLJE2s2GZmOmx4j6R16Xs1rOL1kPTBdHp7Sd9L9/dVSQ9L2r7aPlnHOPGXXEQ8CtQDH23m5fPT1waSDBF9K9kkTgeeI/n00CsivttC858HvggMIhleuipDPL8B/g9wc9r2Ac2s9oX0cQSwF8kQ09VN1vkbYF/gSOCfJI1ooct/B/qm7RyWxnxGOoR1DPBCGscXmtn2QeDwdHoC8GzaRuP8g22I9zBgBPBxSSOBHwGnk7x3/YEhFev+EPhhRPQBPgDc0sK+AewGDAAGA5OBGZL2jYj5wBrg6Ip1TwN+2rSBiJgB/Bz4bvpeHCepB3An8DtgF+BrwM8l7Vux6anAv6T9L07baM6VwMHAoUA/YBrwTpV9sg5y4jeAF0j+4Jp6G9idZIz77Yh4KNpW3OmnEbE0IjYA3wE+p/TkbwedCkyPiGcjYj3wTeCkJp82/jkiXo+Ix4HHgff9A0ljORH4ZkSsi4gVwPdIEm4WD/LeRP+vFfOH8W7izxLvJRGxISJeB44H7oqIuRHxJsl7V5kI3wY+KGlARKyPiD+0Eud3IuLNiHgQuBv4XLp8FkmyJ/009nHgFxn3fSzJP7DLI+KtiLgfuAs4uWKduyv24dvAuPS8yWaStiE5ODgnIlZGxKaIqEu3sZw48RskR4N/aWb5vwF/An4n6VlJF7ax3ecrpv8M9CA5+uuoQWl7lW13J/lU0qjyKpy/0vyJ5wHAts20NThjHPOAfSTtChxIcqJ8qKQBwCHA3DbEW/leDaqcT/9xrql4/UxgH+BJSfMlHVslxrXp9pV9D0qnfwYcJ6kXyT+DhyJiVZW2Kg0Cnm+8KKCi7cr3rnIf1pP8jg3ivQYAPYFnMvZrncCJv+QkfZjkj/Xhpq+lR8HnR8RewHHAVElHNr6cofnKo7s9SI5UVwMbgB0qYuhGMpy0uetW2n2B5MRrZdsbgZcyxFRpdRpT07ZWZtk4Iv4KLATOAZZGxFtAHTAVeCYiVrch3sp9XkXFeydpB5LhnsZ+n46Ik0mGWK4AZkvasYUwd27y2h5pPETESpJ/XpNIPuW8b5inhfga92loesRe2Xble1e5D71IPlW+0KSd1cAbJENWVhAn/pKS1Cc9UrwJ+FlELGlmnWMlfVCSgNdILgHdlL78Esl4dTWnSRqZJq7/DcyOiE3AfwE9JX0qHSu+CNiuYruXgOFNkkqlG4HzJO2ZJpTGcwIbs+x7ozSWW4DLJPVOTz5OJTkSzupB4GzeHdZ5oMl8e+KdDRybnjDeluS92/xeSDpN0sD0aPuVdPGmZtpp9M+StpX0UeBY4NaK135CMqY+Cri9ShtNf96PkPwDnyaph6TDSQ4ObqpY55MV+/AvwCMRUfnJhnQfbgCmSxokqZukcZIqfx+skznxl8+dktaRfAz/NjAdOKOFdfcG7gXWkxwZ/kdEPJC+9q/ARekVHV9vYfufAjNJhl16Av8IyRVFwFeB60iOEDeQnERu1JiY1kh6rJl2b0jbngv8N8kR49da3OPqvpb2/yzJp55fpO1n9SDQm3eHdZrOtzneiFgGnJXGsgpYy3vfn08AyyStJznRe1JEvNFCcy+m279AcnL17yPiyYrXbyf5NHJ7kyGhpq4HRqY/71+ln24mkpwAXw38B/D5Jm3/AriYZIjnYJJzHc35OrAEmJ+uewXOTbmSb8RiVm6SngG+0vhlvE5qcyad+IUv61z+r2pWYkq+hxCk3x+wcvC3BM1KStIDwEjg9CZX59hWzkM9ZmYl46EeM7OS6RJDPQMGDIjhw4fXOgwzsy5l4cKFqyNiYNPlXSLxDx8+nAULFtQ6DDOzLkXSn5tb7qEeM7OSceI3MyuZXBO/pJ0kzZb0pKTl6Vex+6U1up9On3fOMwYzM3uvvMf4fwj8JiKOT+t17EBS0/2+iLg8rfZ4IcldoMzMWvT2229TX1/PG2+0VJ2ivHr27MmQIUPo0aNHpvVzS/yS+pDUKP8CQFrb4y1Jn+bdm1fMIilq5cRvZlXV19fTu3dvhg8fTlI30AAigjVr1lBfX8+ee+6ZaZs8h3r2AhqAH0taJOm6tDzsro01v9PnXZrbWNIUJbeXW9DQ0JBjmGbWFbzxxhv079/fSb8JSfTv379Nn4TyTPzdgdHAjyLiIJIKiJlv5BERMyJiTESMGTjwfZehmlkJOek3r63vS56Jv56kOt8j6fxskn8EL0naHSB9fjnHGMzMrIncxvgj4kVJz6c3dn6K5KbXf0wfk4HL0+c78orBzLZewy+8u1PbW3H5p6q+ft555zFs2DDOPfdcAD7+8Y8zdOhQrrvuOgDOP/98Bg8ezNSpU9vV/yWXXEKvXr34+tffe3uLuXPncu655/LEE09w0003cfzxx7er/Up5X9XzNeDn6RU9z5Lc8GMb4BZJZwLPASfkHIPVQNM/ytb+qMy2dIceeii33nor5557Lu+88w6rV6/mtdde2/x6XV0dP/jBDzK1tWnTJrp165Zp3T322IOZM2dy5ZVXtivu5uR6HX9ELE7H6fePiM9ExNqIWBMRR0bE3ulzczf5NjPboowfP566ujoAli1bxn777Ufv3r1Zu3Ytb775JsuXL+eggw4iIrjgggvYb7/9GDVqFDfffDMADzzwAEcccQSnnHIKo0aNAuCyyy5j33335aijjuKpp55qtt/hw4ez//77s802nZeuu0StHjOzWhs0aBDdu3fnueeeo66ujnHjxrFy5UrmzZtH37592X///dl222257bbbWLx4MY8//jirV6/mwx/+MBMmTADg0UcfZenSpey5554sXLiQm266iUWLFrFx40ZGjx7NwQcfXMi+OPGbmWXUeNRfV1fH1KlTWblyJXV1dfTt25dDDz0UgIcffpiTTz6Zbt26seuuu3LYYYcxf/58+vTpwyGHHLL5WvuHHnqISZMmscMOOwAwceLEwvbDtXrMzDI69NBDqaurY8mSJey3336MHTuWefPmUVdXx/jx44HkC1Ut2XHHHd8zX6vLU534zcwyGj9+PHfddRf9+vWjW7du9OvXj1deeYV58+Yxbtw4ACZMmMDNN9/Mpk2baGhoYO7cuRxyyCHva2vChAncfvvtvP7666xbt44777yzsP3wUI+ZdUm1uFJs1KhRrF69mlNOOeU9y9avX8+AAQMAmDRpEvPmzeOAAw5AEt/97nfZbbfdePLJJ9/T1ujRoznxxBM58MADGTZsGB/96Eeb7XP+/PlMmjSJtWvXcuedd3LxxRezbNmyDu1Hl7jn7pgxY8I3YulafDmndbbly5czYsSIWoexxWru/ZG0MCLGNF3XQz1mZiXjxG9mVjJO/GZmJePEb2ZWMk78ZmYl48RvZlYyvo7fzLqmS/p2cnuvVn25VmWZp0+fznXXXUf37t0ZOHAgN9xwA8OGDWtXH418xG9mlkFjuQZgc1nmyi9SVZZtaM2mTZsy93vQQQexYMECnnjiCY4//nimTZvWtsCb4cRvZpZBrcoyH3HEEZsLuY0dO5b6+voO74uHeszMMtgSyjJff/31HHPMMR3eFyd+M7OMalmW+Wc/+xkLFizgwQcf7PB+eKjHzCyjWpVlvvfee7nsssuYM2cO2223Xft3IOXEb2aWUS3KMi9atIivfOUrzJkzh1122aVT9sNDPWbWNbVy+WUealGW+YILLmD9+vWccMIJQHLz9Tlz5nRoP1yW2XLhsszW2VyWuTqXZTYzsxY58ZuZlYwTv5lZyTjxm5mVjBO/mVnJ5Ho5p6QVwDpgE7AxIsZI6gfcDAwHVgCfi4i1ecZhZmbvKuI6/iMiYnXF/IXAfRFxuaQL0/lvFBCHmW1FRs0a1antLZm8pOrrtSrLfO2113LNNdfQrVs3evXqxYwZMxg5cmS7+mhUi6GeTwOz0ulZwGdqEIOZWZvUqizzKaecwpIlS1i8eDHTpk1r9z+WSnkn/gB+J2mhpCnpsl0jYhVA+tzsd5AlTZG0QNKChoaGnMM0M6uuVmWZ+/Tps3l6w4YNmev7VJP3UM/4iHhB0i7APZKebHWLVETMAGZA8s3dvAI0M8uilmWZr7nmGqZPn85bb73F/fff3+F9yfWIPyJeSJ9fBm4HDgFekrQ7QPr8cp4xmJl1lsqyzOPGjWPcuHGb51srywy0WJa5T58+Vcsyn3XWWTzzzDNcccUVXHrppR3ej9wSv6QdJfVunAb+FlgKzAEmp6tNBu7IKwYzs85Uq7LMjU466SR+9atftT3wJvI84t8VeFjS48CjwN0R8RvgcuBoSU8DR6fzZmZbvFqUZX766ac3T999993svffeHd6P3Mb4I+JZ4IBmlq8BjsyrXzMrh9Yuv8xDLcoyX3311dx777306NGDnXfemVmzZjW7Xlu4LLPlwmWZrbO5LHN1LstsZmYtcuI3MysZJ34zs5Jx4jczKxknfjOzknHiNzMrmVav45c0HlgcERsknQaMBn4YEX/OPTozsxYs/1DnXto54snlVV+vVVnmRrNnz+aEE05g/vz5jBnzvis02yTLEf+PgL9KOgCYBvwZ+EmHejUz62JqVZYZYN26dVx11VV85CMfadN2LcmS+DdG8i2vT5Mc6f8Q6N0pvZuZdRG1KssM8J3vfIdp06bRs2fPTtmXLCUb1kn6JnA68FFJ3YAendK7mVkXUauyzIsWLeL555/n2GOP5corr+yUfcmS+E8ETgG+GBEvStoD+LdO6d3MrAupLMs8depUVq5cSV1dHX379m21LHOfPn1aLMsMNFuW+Z133uG8885j5syZnbofrQ71RMSLwG3Adumi1SS19c3MSqXosszr1q1j6dKlHH744QwfPpw//OEPTJw4kY7WLms18Uv6MjAb+L/posFAxwtCm5l1MUWXZe7bty+rV69mxYoVrFixgrFjxzJnzpwOX9WTZajnLJI7Zz0CEBFPp7dSNDOrmdYuv8xDLcoy56HVssySHomIj0haFBEHSeoOPBYR+xcTossyd0Uuy2ydzWWZq+vssswPSvoWsL2ko4FbgeZvFWNmZlu8LIn/QqABWAJ8Bfg1cFGeQZmZWX6yjPFvD9wQEf8JkF7Hvz3w1zwDMzOzfGQ54r+PJNE32h64N59wzMwsb1kSf8+IWN84k07vkF9IZmaWpyyJf4Ok0Y0zkg4GXs8vJDMzy1OWMf5zgVslvZDO705SxsHMrGau+fv7O7W9s679WNXXa1WWeebMmVxwwQUMHjwYgLPPPpsvfelL7eqjUZaSDfOBDwH/AHwVGBERCzvUq5lZF1PLsswnnngiixcvZvHixR1O+pD9DlwfBvYHDgJOlvT5DvdsZtaF1LIsc2fLcgeunwIfABYDjf+mAt+MxcxKpFZlmQFuu+025s6dyz777MP3v/99hg4d2qF9yTLGPwYYGa3VdmhBet3/AmBlRBwraU/gJqAf8BhwekS81Z62zcyKVHRZZoDjjjuOk08+me22245rr72WyZMnc//9HTu/kWWoZymwWwf6OAeorKZ0BfD9iNgbWAuc2YG2zcwKU3RZZoD+/fuz3XZJVfwvf/nLLFzY8VOsWRL/AOCPkn4raU7jI0vjkoYAnwKuS+cFfIykzDPALOAzbQ/bzKx4RZdlBli1atXm6Tlz5nRKobosQz2XdKD9H5DcoL3xHr39gVciYmM6X09S3/99JE0BpgDsscceHQjBzLZGrV1+mYdalGW+6qqrmDNnDt27d6dfv36dcjeuVssyA0gaBuwdEfdK2gHoFhHrWtnmWOCTEfFVSYcDXwfOAOZFxAfTdYYCv46IUdXaclnmrsdlma2zuSxzdW0py5zlqp4vkxx59yO5umcwcC1wZCubjgcmSvok0BPoQ/IJYCdJ3dOj/iHAC1XaMDOzTpZljP8skiT+GiR34AJavQNXRHwzIoZExHDgJOD+iDgV+D1wfLraZOCOdsRtZmbtlCXxv1l5uWV6B652XdqZ+gYwVdKfSMb8r+9AW2ZWIu28qnyr19b3JcvJ3aZ34PoqbbwDV0Q8ADyQTj9Lcg9fM7PMevbsyZo1a+jfv3+myyDLIiJYs2YNPXv2zLxNlsR/Icm19pV34LquXRGambXTkCFDqK+vp6GhodahbHF69uzJkCFDMq9fNfGn37qdFRGnAf/ZwdjMzNqtR48em7/1ah1TdYw/IjYBAyVtW1A8ZmaWsyxDPSuA/59+W3dD48KImJ5XUGZmlp8sif+F9LEN734D18zMuqgsY/y9IuKCguIxM7OcZRnjH11tHTMz61qyDPUsTsf3b+W9Y/y/zC0qMzPLTZbE3w9YQ1JOuVEATvxmZl1Qq4k/Is4oIhAzMytGluqcP6aZ2jwR8cVcIjIzs1xlGeq5q2K6JzAJl1I2M+uysgz13FY5L+lG4N7cIjIzs1xlKcvc1N6A74VoZtZFZRnjX8d7x/hfJKmpb2ZmXVCWoR6XaTAz24q0OtQjaZKkvhXzO0n6TL5hmZlZXrKM8V8cEa82zkTEK8DF+YVkZmZ5ypL4m1sny2WgZma2BcqS+BdImi7pA5L2kvR9YGHegZmZWT6yJP6vAW8BNwO3AK8DZ+UZlJmZ5SfLVT0bSG64bmZmW4EsV/XcI2mnivmdJf0237DMzCwvWYZ6BqRX8gAQEWuBXfILyczM8pQl8b8jaXOJBknDaKZap5mZdQ1ZLsv8NvCwpAfT+QnAlPxCMjOzPGU5ufsbSaOBsYCA8yJidWvbSeoJzAW2S/uZHREXS9oTuInkzl6PAadHxFsd2AczM2uDqkM9kraVdAbJVT2HAQOAdRnbfhP4WEQcABwIfELSWOAK4PsRsTewFjizvcGbmVnbtZj4JY0E/ggcDjwH1KfTy9LXqorE+nS2R/oIknv3zk6XzwJc98fMrEDVhnr+HfiHiLincqGko4BrgCNaa1xSN5Jv+X4w3eYZ4JWI2JiuUg8MbkfcZmbWTtWGegY3TfoAEXEvsFuWxiNiU0QcCAwBDgFGNLdac9tKmiJpgaQFDQ0NWbozM7MMqiX+bSRt13RhetK2TUXa0u8BPEBygngnSY3bD6GF+/dGxIyIGBMRYwYOHNiW7szMrIpqif8nwG2ShjcuSKdvAX7aWsOSBjZ+41fS9sBRwHLg98Dx6WqTgTvaHraZmbVXi0fuEXGppLOBuZJ2SBdvAK6MiH/P0PbuwKx0nH8b4JaIuEvSH4GbJF0KLAKu79gumJlZW1QdsomIq4GrJfVO57NeyklEPAEc1MzyZ0nG+83MrAYyjdW3JeGbmdmWLUutHjMz24o48ZuZlUyLQz2SPlttw4j4ZeeHY2Zmeas2xn9c+rwLcChwfzp/BMk1+U78ZmZdULXLOc8AkHQXMDIiVqXzu5OUXzAzsy4oyxj/8Makn3oJ2CeneMzMLGdZLud8IL3H7o0kdXVOIvn2rZmZdUFZbsRytqRJJHfeApgREbfnG5aZmeWlauJPyy38NiKOApzszcy2AlXH+CNiE/BXSX0LisfMzHKWZYz/DWCJpHtIirQBEBH/mFtUZmaWmyyJ/+70YWZmW4EsJ3dnSdqWdy/hfCoi3s43LDMzy0uriV/S4SQ3RV8BCBgqaXJEzM03NDMzy0OWoZ7vAX8bEU8BSNqH5Jr+g/MMzMzM8pHlm7s9GpM+QET8F9Ajv5DMzCxPWY74F0i6nnfvs3sqsDC/kMzMLE9ZEv8/AGcB/0gyxj8X+I88gzIzs/xUq8f/GaAuIl4GpqcPMzPr4qqN8Z8GLJL0tKSZkqZI+l9FBWZmZvloMfFHxPERMRg4GvgdsD/wE0kNkn5dVIBmZta5snyBa4WknsD26aNx2szMuqBqY/zfAsYBA4GngD8AVwNT0uJtZmbWBVU74v88sB64C6gDHomIVwuJyszMclPtnrsfktSP5EbrhwMXSuoFPE5ytc+PiwnRzMw6U2v1+P8SEXcB/wR8E7gVOAK4rrWGJQ2V9HtJyyUtk3ROuryfpHvSq4XukbRzJ+yHmZll1GLilzRR0uWSHgJeBq4EBgDnA7tlaHsjcH5EjADGAmdJGglcCNwXEXsD96XzZmZWkGpj/F8gGdufBiyMiLfa0nBErAJWpdPrJC0HBgOfJhk6gqTq5wPAN9rStpmZtV+1Mf7PdlYnkoYDBwGPALum/xSIiFWSdumsfszMrHVZqnN2SHpC+Dbg3Ih4rQ3bTZG0QNKChoaG/AI0MyuZXBO/pB4kSf/nEfHLdPFLknZPX9+d5PzB+0TEjIgYExFjBg4cmGeYZmalklvilyTgemB5RFQWeJsDTE6nJwN35BWDmZm9X7sSv6RLMqw2Hjgd+Jikxenjk8DlwNGSniapA3R5e2IwM7P2yVKPvzmt3oglIh4mqd/fnCPb2a+ZmXVQu474I+LOzg7EzMyK0WrilzRE0u1pOeaXJN0maUgRwZmZWefLcsT/Y5ITsruTfAHrznSZmZl1QVkS/8CI+HFEbEwfM0lKNZuZWReUJfGvlnSapG7p4zRgTd6BmZlZPrIk/i8CnwNeJKm9c3y6zMzMuqAst158DphYQCxmZlaAarde/Kcq20VE/EsO8ZiZWc6qHfFvaGbZjsCZQH/Aid/MrAuqVpb5e43TknoD5wBnADcB32tpOzMz27JVHeNP77k7FTiV5KYpoyNibRGBmZlZPqqN8f8b8FlgBjAqItYXFpWZmeWm2uWc5wODgIuAFyS9lj7WScp8QxUzM9uyVBvjz/3uXGZmVjwndzOzknHiNzMrGSd+M7OSceI3MysZJ34zs5Jx4jczKxknfjOzknHiNzMrGSd+M7OSceI3MysZJ34zs5Jx4jczKxknfjOzkskt8Uu6QdLLkpZWLOsn6R5JT6fPO+fVv5mZNS/PI/6ZwCeaLLsQuC8i9gbuS+fNzKxAuSX+iJgL/KXJ4k+T3MKR9PkzefVvZmbNK3qMf9eIWAWQPu/S0oqSpkhaIGlBQ0NDYQGamW3tttiTuxExIyLGRMSYgQMH1jocM7OtRtGJ/yVJuwOkzy8X3L+ZWekVnfjnAJPT6cnAHQX3b2ZWenleznkjMA/YV1K9pDOBy4GjJT0NHJ3Om5lZgbrn1XBEnNzCS0fm1aeZmbVuiz25a2Zm+XDiNzMrGSd+M7OSceI3MyuZ3E7umplZYtSsUe9btmTykhpEkvARv5lZyTjxm5mVjBO/mVnJOPGbmZWME7+ZWck48ZuZlYwTv5lZyTjxm5mVjBO/mVnJOPGbmZWME7+ZWck48ZuZlYwTv5lZyTjxm5mVjBO/mVnJOPGbmZWME7+ZWck48ZuZlYwTv5lZyTjxm5mVjG+2bluU5R8a8Z75EU8ur1EkZvlq+rsOxf2+1+SIX9InJD0l6U+SLqxFDGZmZVV44pfUDbgGOAYYCZwsaWTRcZiZlVUtjvgPAf4UEc9GxFvATcCnaxCHmVkp1WKMfzDwfMV8PfCRpitJmgJMSWfXS3qqg/0OAFZ3sI2u2n/N+9YV7dxa6nDfHWmgi/Zd6/7ddwbNDnN07Pe9OcOaW1iLxN/cnsX7FkTMAGZ0WqfSgogY01ntdaX+3Xe5+q51/+57y1eLoZ56YGjF/BDghRrEYWZWSrVI/POBvSXtKWlb4CRgTg3iMDMrpcKHeiJio6Szgd8C3YAbImJZAV132rBRF+zffZer71r37763cIp43/C6mZltxVyywcysZJz4zcxKphSJv5YlIiTdIOllSUsL7neopN9LWi5pmaRzCu6/p6RHJT2e9v/PRfafxtBN0iJJdxXc7wpJSyQtlrSg4L53kjRb0pPpz35cgX3vm+5z4+M1SecW2P956e/aUkk3SupZYN/npP0uK3Kf2y0ituoHyQnkZ4C9gG2Bx4GRBfY/ARgNLC14v3cHRqfTvYH/Kni/BfRKp3sAjwBjC34PpgK/AO4quN8VwIAi+6zoexbwpXR6W2CnGsXRDXgRGFZQf4OB/wa2T+dvAb5QUN/7AUuBHUgumLkX2LsW73vWRxmO+GtaIiIi5gJ/Kaq/in5XRcRj6fQ6YDnJH0dR/UdErE9ne6SPwq4kkDQE+BRwXVF91pqkPiQHGtcDRMRbEfFKjcI5EngmIv5cYJ/dge0ldSdJwkV9P2gE8IeI+GtEbAQeBCYV1He7lCHxN1ciorAEuCWQNBw4iOSou8h+u0laDLwM3BMRRfb/A2Aa8E6BfTYK4HeSFqalR4qyF9AA/Dgd4rpO0o4F9l/pJODGojqLiJXAlcBzwOZmb7gAAAINSURBVCrg1Yj4XUHdLwUmSOovaQfgk7z3S6pbnDIk/kwlIrZWknoBtwHnRsRrRfYdEZsi4kCSb2cfImm/IvqVdCzwckQsLKK/ZoyPiNEkFWjPkjShoH67kwwr/igiDgI2AIWXPU+/mDkRuLXAPncm+SS/JzAI2FHSaUX0HRHLgSuAe4DfkAwnbyyi7/YqQ+IvbYkIST1Ikv7PI+KXtYojHW54APhEQV2OByZKWkEytPcxST8rqG8i4oX0+WXgdpLhxiLUA/UVn6xmk/wjKNoxwGMR8VKBfR4F/HdENETE28AvgUOL6jwiro+I0RExgWRo9+mi+m6PMiT+UpaIkCSSsd7lETG9Bv0PlLRTOr09yR/mk0X0HRHfjIghETGc5Od9f0QUcvQnaUdJvRungb8lGQrIXUS8CDwvad900ZHAH4vou4mTKXCYJ/UcMFbSDunv/pEk57UKIWmX9HkP4LMUv/9tstXfejFqVyICAEk3AocDAyTVAxdHxPUFdD0eOB1Yko6zA3wrIn5dQN+QXFU0K73xzjbALRFR6GWVNbIrcHuSe+gO/CIiflNg/18Dfp4e5DwLnFFg36Rj3EcDXymy34h4RNJs4DGSYZZFFFtC4TZJ/YG3gbMiYm2BfbeZSzaYmZVMGYZ6zMysghO/mVnJOPGbmZWME7+ZWck48ZuZlYwTv5lZyTjxm5mVzP8AJ5t2GCSaMEAAAAAASUVORK5CYII=",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1471,7 +1475,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1590,6 +1594,11 @@
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
    "version": "3.7.3"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "6195794cbcdce20f0e270d374b1f7d69304b590c495656fadb23ef4a134eb729"
+   }
   }
  },
  "nbformat": 4,
diff --git a/Youtube Downloader.py b/Youtube Downloader.py
new file mode 100644
index 0000000..7ab898b
--- /dev/null
+++ b/Youtube Downloader.py	
@@ -0,0 +1,7 @@
+import pytube
+
+link = input("Paste Link Here - ")
+print("Downloading...")
+
+pytube.YouTube(link).streams.first().download(output_path=r"C:\Users\callu\Pictures")
+print("Video Download Done!")
\ No newline at end of file

From 1c128aa00624027be3661285e93973c7a372ad4a Mon Sep 17 00:00:00 2001
From: CallumOrmond <callumormond10@gmail.com>
Date: Sun, 4 Dec 2022 23:16:25 +0000
Subject: [PATCH 2/2] Desktop

---
 .../ML Coursework-checkpoint.ipynb            | 245 +++++++++++++++++-
 Uni/ML Coursework.ipynb                       | 115 +++++++-
 2 files changed, 350 insertions(+), 10 deletions(-)

diff --git a/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb b/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb
index a4c8910..1304ad4 100644
--- a/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb	
+++ b/Uni/.ipynb_checkpoints/ML Coursework-checkpoint.ipynb	
@@ -3,6 +3,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
+<<<<<<< HEAD
    "metadata": {},
    "outputs": [],
    "source": [
@@ -139,6 +140,144 @@
    "cell_type": "code",
    "execution_count": 232,
    "metadata": {},
+=======
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "C1 = create_collection(10_000,100,10,50_000,0.1,0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# C2 = create_collection(100_000,100,100,50_000,0.1,0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# C3 = create_collection(100_000,10,1000,50_000,0.1,0.01)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import sklearn as sk\n",
+    "import gensim \n",
+    "\n",
+    "\n",
+    "### VERSION --- V x T phi --- each word has a distibution of topics for that word \n",
+    "\n",
+    "# got choice over multinomial - get index instead of list with chosen value\n",
+    "# https://stackoverflow.com/questions/36952419/vectorizing-numpy-random-multinomial\n",
+    "\n",
+    "\n",
+    "C1 = (10_000,100,10,50_000,0.1,0.01)\n",
+    "C2 = (100_000,100,100,50_000,0.1,0.01) \n",
+    "C3 = (100_000,10,1000,50_000,0.1,0.01)\n",
+    "\n",
+    "def create_collectionVxT(M, N, T, V, alpha, beta):\n",
+    "    # M = Number of Docs \n",
+    "    # N = Max words in doc \n",
+    "    # T = Number of Topics\n",
+    "    # V = Vocab Size\n",
+    "    \n",
+    "    # GENERATE THETA --- M x T\n",
+    "    alpha = np.full(T, alpha)\n",
+    "    theta = np.random.dirichlet(alpha, M)\n",
+    "    \n",
+    "    #GENERATE PHI --- T x V\n",
+    "    beta = np.full(T, beta)\n",
+    "    phi = np.random.dirichlet(beta, V)\n",
+    "        \n",
+    "    \n",
+    "    # for each doc \n",
+    "    DocTopicForWord = []\n",
+    "    DocWordFromTopic = []\n",
+    "    for i, Doc in enumerate(theta):\n",
+    "        \n",
+    "        if (i % 100) == 0:\n",
+    "            print(\"Done\", i, \"Docs\")\n",
+    "         \n",
+    "        TopicForWord = []\n",
+    "        WordFromTopic = []\n",
+    "        # for each word in doc \n",
+    "        for j in range(N):\n",
+    "            \n",
+    "            # 3. a)\n",
+    "            #For each word in a doc pick a topic from the doc \n",
+    "            topic = np.random.choice(np.arange(T), p=Doc)\n",
+    "            TopicForWord.append(topic)\n",
+    "            \n",
+    "            # 3. b)\n",
+    "            #For each word in a doc choose a word from phi - based on the choosen word above\n",
+    "            # gets vocab for topic \n",
+    "            \n",
+    "            # NEW\n",
+    "            \n",
+    "            phi_normal = phi.T[topic] / sum(phi.T[topic])\n",
+    "            \n",
+    "            word = np.random.choice(np.arange(V), p=phi_normal)  # + (topic * V)\n",
+    "            WordFromTopic.append(word)\n",
+    "        \n",
+    "        \n",
+    "#         # 3. a) --- Faster \n",
+    "#         TopicForWord = np.random.choice(np.arange(T), p=Doc, size=N)\n",
+    "#         # 3. b) --- Faster\n",
+    "#         WordFromTopic = np.random.choice(np.arange(V), p=phi[topic], size=N)\n",
+    "        \n",
+    "        DocTopicForWord.append(TopicForWord)\n",
+    "        DocWordFromTopic.append(WordFromTopic)\n",
+    "        \n",
+    "        D = np.array(DocTopicForWord)\n",
+    "        A = np.array(DocWordFromTopic)\n",
+    "        \n",
+    "    sum1 = 0\n",
+    "    for i in range(T):\n",
+    "        sum1 = sum1 + phi[i][0]\n",
+    "\n",
+    "    print(sum1)\n",
+    "    \n",
+    "    print(theta.shape, phi.shape)\n",
+    "        \n",
+    "    \n",
+    "    \n",
+    "    # A topic for each word, Found by:\n",
+    "    # For each word in a doc, get a topic index by randomly choosing from a list [1...T] using the P in theta for each Doc\n",
+    "    print(\"\\n\\nDocTopicForWord\", D.shape, DocTopicForWord[:1]) \n",
+    "    \n",
+    "    # A Word for each word, Found by:\n",
+    "    # For each word in a doc, get a word index by randomly choosing from a list [1...V] using the P in phi for each topic\n",
+    "    print(\"DocWordFromTopic\", A.shape, DocWordFromTopic[:1])\n",
+    "    \n",
+    "    a = np.random.multinomial(10, theta[0])\n",
+    "    \n",
+    "    print(theta[0])\n",
+    "    print(a)\n",
+    "    \n",
+    "    return DocTopicForWord, DocWordFromTopic, theta, phi\n",
+    "    \n",
+    "\n",
+    "collection0 = create_collection(100, 15, 10, 500, 0.1, 0.01)\n",
+    "# collection1 = create_collection(10_000,100,10,50_000,0.1,0.01)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+>>>>>>> 869fa54095e0316b3e32c93f87beaa728294c973
    "outputs": [
     {
      "name": "stdout",
@@ -155,12 +294,111 @@
       "Done 700 Docs\n",
       "Done 800 Docs\n",
       "Done 900 Docs\n",
+<<<<<<< HEAD
       "0.024929452872102225\n",
       "(1000, 10) (10, 500)\n",
       "\n",
       "\n",
       "DocTopicForWord (1000, 15) [[0, 7, 9, 9, 7, 7, 7, 7, 0, 7, 9, 9, 9, 7, 9]]\n",
       "DocWordFromTopic (1000, 15) [[315, 33, 125, 144, 408, 35, 245, 164, 338, 418, 292, 125, 278, 448, 125]]\n"
+=======
+      "Done 1000 Docs\n",
+      "Done 1100 Docs\n",
+      "Done 1200 Docs\n",
+      "Done 1300 Docs\n",
+      "Done 1400 Docs\n",
+      "Done 1500 Docs\n",
+      "Done 1600 Docs\n",
+      "Done 1700 Docs\n",
+      "Done 1800 Docs\n",
+      "Done 1900 Docs\n",
+      "Done 2000 Docs\n",
+      "Done 2100 Docs\n",
+      "Done 2200 Docs\n",
+      "Done 2300 Docs\n",
+      "Done 2400 Docs\n",
+      "Done 2500 Docs\n",
+      "Done 2600 Docs\n",
+      "Done 2700 Docs\n",
+      "Done 2800 Docs\n",
+      "Done 2900 Docs\n",
+      "Done 3000 Docs\n",
+      "Done 3100 Docs\n",
+      "Done 3200 Docs\n",
+      "Done 3300 Docs\n",
+      "Done 3400 Docs\n",
+      "Done 3500 Docs\n",
+      "Done 3600 Docs\n",
+      "Done 3700 Docs\n",
+      "Done 3800 Docs\n",
+      "Done 3900 Docs\n",
+      "Done 4000 Docs\n",
+      "Done 4100 Docs\n",
+      "Done 4200 Docs\n",
+      "Done 4300 Docs\n",
+      "Done 4400 Docs\n",
+      "Done 4500 Docs\n",
+      "Done 4600 Docs\n",
+      "Done 4700 Docs\n",
+      "Done 4800 Docs\n",
+      "Done 4900 Docs\n",
+      "Done 5000 Docs\n",
+      "Done 5100 Docs\n",
+      "Done 5200 Docs\n",
+      "Done 5300 Docs\n",
+      "Done 5400 Docs\n",
+      "Done 5500 Docs\n",
+      "Done 5600 Docs\n",
+      "Done 5700 Docs\n",
+      "Done 5800 Docs\n",
+      "Done 5900 Docs\n",
+      "Done 6000 Docs\n",
+      "Done 6100 Docs\n",
+      "Done 6200 Docs\n",
+      "Done 6300 Docs\n",
+      "Done 6400 Docs\n",
+      "Done 6500 Docs\n",
+      "Done 6600 Docs\n",
+      "Done 6700 Docs\n",
+      "Done 6800 Docs\n",
+      "Done 6900 Docs\n",
+      "Done 7000 Docs\n",
+      "Done 7100 Docs\n",
+      "Done 7200 Docs\n",
+      "Done 7300 Docs\n",
+      "Done 7400 Docs\n",
+      "Done 7500 Docs\n",
+      "Done 7600 Docs\n",
+      "Done 7700 Docs\n",
+      "Done 7800 Docs\n",
+      "Done 7900 Docs\n",
+      "Done 8000 Docs\n",
+      "Done 8100 Docs\n",
+      "Done 8200 Docs\n",
+      "Done 8300 Docs\n",
+      "Done 8400 Docs\n",
+      "Done 8500 Docs\n",
+      "Done 8600 Docs\n",
+      "Done 8700 Docs\n",
+      "Done 8800 Docs\n",
+      "Done 8900 Docs\n",
+      "Done 9000 Docs\n",
+      "Done 9100 Docs\n",
+      "Done 9200 Docs\n",
+      "Done 9300 Docs\n",
+      "Done 9400 Docs\n",
+      "Done 9500 Docs\n",
+      "Done 9600 Docs\n",
+      "Done 9700 Docs\n",
+      "Done 9800 Docs\n",
+      "Done 9900 Docs\n",
+      "3.6766032786614873e-06\n",
+      "(10000, 10) (10, 50000)\n",
+      "\n",
+      "\n",
+      "DocTopicForWord (10000, 100) [[2, 0, 2, 2, 0, 7, 0, 2, 2, 8, 8, 8, 7, 8, 0, 0, 8, 7, 2, 7, 7, 8, 8, 8, 7, 2, 2, 2, 2, 0, 2, 0, 0, 8, 8, 7, 8, 7, 2, 2, 8, 2, 7, 8, 2, 0, 2, 8, 2, 2, 8, 2, 2, 2, 2, 8, 8, 2, 2, 2, 2, 2, 8, 2, 8, 7, 8, 7, 2, 7, 2, 2, 2, 0, 7, 8, 7, 2, 2, 2, 8, 2, 7, 8, 2, 8, 2, 2, 7, 0, 2, 7, 2, 7, 7, 8, 7, 2, 2, 7]]\n",
+      "DocWordFromTopic (10000, 100) [[29136, 20033, 42183, 46277, 20099, 25341, 36541, 9807, 3479, 15265, 10842, 29587, 32940, 4434, 44855, 23885, 13655, 45328, 30450, 26022, 29542, 10794, 40479, 22292, 25043, 43065, 25881, 17959, 2432, 20099, 13611, 19339, 27548, 48971, 4756, 44667, 9041, 9363, 38213, 44382, 28267, 26735, 15955, 49974, 6398, 30288, 12756, 33243, 35385, 5945, 24119, 2721, 47372, 3022, 44806, 20623, 46065, 16426, 24785, 8472, 10359, 6093, 5937, 1199, 46626, 40893, 13164, 13601, 14113, 34026, 10252, 19997, 38973, 41449, 36231, 39460, 31191, 14950, 4004, 19324, 35385, 17314, 45509, 31636, 554, 22896, 16639, 37692, 22934, 4573, 39683, 16262, 15743, 35130, 38153, 23864, 15143, 24056, 39253, 38354]]\n"
+>>>>>>> 869fa54095e0316b3e32c93f87beaa728294c973
      ]
     }
    ],
@@ -254,9 +492,10 @@
     "    return DocTopicForWord, DocWordFromTopic, theta, phi\n",
     "    \n",
     "\n",
-    "collection0 = create_collection(1000, 15, 10, 500, 0.1, 0.01)\n",
-    "# collection1 = create_collection(10_000,100,10,50_000,0.1,0.01)\n",
-    "\n"
+    "# collection0 = create_collection(1000, 15, 10, 500, 0.1, 0.01)\n",
+    "\n",
+    "\n",
+    "# collection1v1 = create_collection(10_000,100,10,50_000,0.1,0.01) --- DONE --- DONE --- DONE --- DONE --- DONE"
    ]
   },
   {
diff --git a/Uni/ML Coursework.ipynb b/Uni/ML Coursework.ipynb
index a8d6d0b..1e995fd 100644
--- a/Uni/ML Coursework.ipynb	
+++ b/Uni/ML Coursework.ipynb	
@@ -48,7 +48,7 @@
     "C2 = (100_000,100,100,50_000,0.1,0.01) \n",
     "C3 = (100_000,10,1000,50_000,0.1,0.01)\n",
     "\n",
-    "def create_collection(M, N, T, V, alpha, beta):\n",
+    "def create_collectionVxT(M, N, T, V, alpha, beta):\n",
     "    # M = Number of Docs \n",
     "    # N = Max words in doc \n",
     "    # T = Number of Topics\n",
@@ -137,7 +137,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 232,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
@@ -155,12 +155,102 @@
       "Done 700 Docs\n",
       "Done 800 Docs\n",
       "Done 900 Docs\n",
-      "0.024929452872102225\n",
-      "(1000, 10) (10, 500)\n",
+      "Done 1000 Docs\n",
+      "Done 1100 Docs\n",
+      "Done 1200 Docs\n",
+      "Done 1300 Docs\n",
+      "Done 1400 Docs\n",
+      "Done 1500 Docs\n",
+      "Done 1600 Docs\n",
+      "Done 1700 Docs\n",
+      "Done 1800 Docs\n",
+      "Done 1900 Docs\n",
+      "Done 2000 Docs\n",
+      "Done 2100 Docs\n",
+      "Done 2200 Docs\n",
+      "Done 2300 Docs\n",
+      "Done 2400 Docs\n",
+      "Done 2500 Docs\n",
+      "Done 2600 Docs\n",
+      "Done 2700 Docs\n",
+      "Done 2800 Docs\n",
+      "Done 2900 Docs\n",
+      "Done 3000 Docs\n",
+      "Done 3100 Docs\n",
+      "Done 3200 Docs\n",
+      "Done 3300 Docs\n",
+      "Done 3400 Docs\n",
+      "Done 3500 Docs\n",
+      "Done 3600 Docs\n",
+      "Done 3700 Docs\n",
+      "Done 3800 Docs\n",
+      "Done 3900 Docs\n",
+      "Done 4000 Docs\n",
+      "Done 4100 Docs\n",
+      "Done 4200 Docs\n",
+      "Done 4300 Docs\n",
+      "Done 4400 Docs\n",
+      "Done 4500 Docs\n",
+      "Done 4600 Docs\n",
+      "Done 4700 Docs\n",
+      "Done 4800 Docs\n",
+      "Done 4900 Docs\n",
+      "Done 5000 Docs\n",
+      "Done 5100 Docs\n",
+      "Done 5200 Docs\n",
+      "Done 5300 Docs\n",
+      "Done 5400 Docs\n",
+      "Done 5500 Docs\n",
+      "Done 5600 Docs\n",
+      "Done 5700 Docs\n",
+      "Done 5800 Docs\n",
+      "Done 5900 Docs\n",
+      "Done 6000 Docs\n",
+      "Done 6100 Docs\n",
+      "Done 6200 Docs\n",
+      "Done 6300 Docs\n",
+      "Done 6400 Docs\n",
+      "Done 6500 Docs\n",
+      "Done 6600 Docs\n",
+      "Done 6700 Docs\n",
+      "Done 6800 Docs\n",
+      "Done 6900 Docs\n",
+      "Done 7000 Docs\n",
+      "Done 7100 Docs\n",
+      "Done 7200 Docs\n",
+      "Done 7300 Docs\n",
+      "Done 7400 Docs\n",
+      "Done 7500 Docs\n",
+      "Done 7600 Docs\n",
+      "Done 7700 Docs\n",
+      "Done 7800 Docs\n",
+      "Done 7900 Docs\n",
+      "Done 8000 Docs\n",
+      "Done 8100 Docs\n",
+      "Done 8200 Docs\n",
+      "Done 8300 Docs\n",
+      "Done 8400 Docs\n",
+      "Done 8500 Docs\n",
+      "Done 8600 Docs\n",
+      "Done 8700 Docs\n",
+      "Done 8800 Docs\n",
+      "Done 8900 Docs\n",
+      "Done 9000 Docs\n",
+      "Done 9100 Docs\n",
+      "Done 9200 Docs\n",
+      "Done 9300 Docs\n",
+      "Done 9400 Docs\n",
+      "Done 9500 Docs\n",
+      "Done 9600 Docs\n",
+      "Done 9700 Docs\n",
+      "Done 9800 Docs\n",
+      "Done 9900 Docs\n",
+      "3.6766032786614873e-06\n",
+      "(10000, 10) (10, 50000)\n",
       "\n",
       "\n",
-      "DocTopicForWord (1000, 15) [[0, 7, 9, 9, 7, 7, 7, 7, 0, 7, 9, 9, 9, 7, 9]]\n",
-      "DocWordFromTopic (1000, 15) [[315, 33, 125, 144, 408, 35, 245, 164, 338, 418, 292, 125, 278, 448, 125]]\n"
+      "DocTopicForWord (10000, 100) [[2, 0, 2, 2, 0, 7, 0, 2, 2, 8, 8, 8, 7, 8, 0, 0, 8, 7, 2, 7, 7, 8, 8, 8, 7, 2, 2, 2, 2, 0, 2, 0, 0, 8, 8, 7, 8, 7, 2, 2, 8, 2, 7, 8, 2, 0, 2, 8, 2, 2, 8, 2, 2, 2, 2, 8, 8, 2, 2, 2, 2, 2, 8, 2, 8, 7, 8, 7, 2, 7, 2, 2, 2, 0, 7, 8, 7, 2, 2, 2, 8, 2, 7, 8, 2, 8, 2, 2, 7, 0, 2, 7, 2, 7, 7, 8, 7, 2, 2, 7]]\n",
+      "DocWordFromTopic (10000, 100) [[29136, 20033, 42183, 46277, 20099, 25341, 36541, 9807, 3479, 15265, 10842, 29587, 32940, 4434, 44855, 23885, 13655, 45328, 30450, 26022, 29542, 10794, 40479, 22292, 25043, 43065, 25881, 17959, 2432, 20099, 13611, 19339, 27548, 48971, 4756, 44667, 9041, 9363, 38213, 44382, 28267, 26735, 15955, 49974, 6398, 30288, 12756, 33243, 35385, 5945, 24119, 2721, 47372, 3022, 44806, 20623, 46065, 16426, 24785, 8472, 10359, 6093, 5937, 1199, 46626, 40893, 13164, 13601, 14113, 34026, 10252, 19997, 38973, 41449, 36231, 39460, 31191, 14950, 4004, 19324, 35385, 17314, 45509, 31636, 554, 22896, 16639, 37692, 22934, 4573, 39683, 16262, 15743, 35130, 38153, 23864, 15143, 24056, 39253, 38354]]\n"
      ]
     }
    ],
@@ -256,11 +346,18 @@
     "    return DocTopicForWord, DocWordFromTopic, theta, phi\n",
     "    \n",
     "\n",
+<<<<<<< HEAD
     "collection0 = create_collection(1000, 15, 10, 500, 0.1, 0.01)\n",
     "# collection1 = create_collection(10_000,100,10,50_000,0.1,0.01)\n",
     "\n",
     "\n",
     "\n"
+=======
+    "# collection0 = create_collection(1000, 15, 10, 500, 0.1, 0.01)\n",
+    "\n",
+    "\n",
+    "# collection1v1 = create_collection(10_000,100,10,50_000,0.1,0.01) --- DONE --- DONE --- DONE --- DONE --- DONE"
+>>>>>>> 869fa54095e0316b3e32c93f87beaa728294c973
    ]
   },
   {
@@ -1579,7 +1676,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -1593,12 +1690,16 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
+<<<<<<< HEAD
    "version": "3.7.3"
   },
   "vscode": {
    "interpreter": {
     "hash": "6195794cbcdce20f0e270d374b1f7d69304b590c495656fadb23ef4a134eb729"
    }
+=======
+   "version": "3.9.12"
+>>>>>>> 869fa54095e0316b3e32c93f87beaa728294c973
   }
  },
  "nbformat": 4,