Deep learning project. During the project we worked with the cifar 10 data set and with the help of the pytorch libraries we built a model that classifies and learns the data in the best way and predicts correct predictions for new data. During the project we examined several hyperparameters: 1. Regularization 2. dropout 3. Learning step 4. Batch normalization 5. Human optimizer 6. SGD optimizer Each experiment has a cell in the code, and the results are shown in a graph. To run, the 3 folders must be uploaded to the personal drive, then the drive address must be changed in order for the code to recognize the folders. Using Google colab (using gpu A100).