This is my code for classification with MNIST data
- dataset(mnist) is available in link: https://drive.google.com/drive/u/0/folders/113e-7OW1r2rCnDZx-ma23yR13K3spamT - Please download folder MNIST into project
- file train_eval.py for training loop and evaluate accuracy
- file predict.py for predicting your image
- folder 'utils' contain some neccessary function like loss function, optimizer, load_data function, class dataset, plot_dataset_function, open_function...
How to predict your dataset?
- easily, first you can put your test_set in folder 'data'
- second, please change the index in the line 20 (getitem function)
- third, you can change your model if you want
How to train your custom data?
- first, you can put your dataset in folder 'data'
- second, change path to dataset in line 9,10,23,24 folder dataset.py
- may be you should modified the class Dataset
The accuracy: accuracy.png
HOW YOU ENJOY IT!