Skip to content

Latest commit

 

History

History
31 lines (16 loc) · 1.27 KB

File metadata and controls

31 lines (16 loc) · 1.27 KB

TensorFlow_FlowersClassificator

This project is the beginning of my journey into the world of AI. And is an AI trained with TensorFlow in Google Colab to classify flowers from their images.


Results

10 epochs

After 10 epochs, I ran into the problem of retraining. This can be verified by looking at the graph of accuracy and loss on the training and test datasets.

1 training and validation accuracy and loss

The following approaches were used as a solution to the problem: augmentation and regularization (Dropout). Also increased the number of learning epochs (20 epochs).

20 epochs

already better

2 training and validation accuracy and loss


Inference on a new image

inference

Good luck!