Convolutional Neural Network

To run the program

pip install -r requirements.txt

python cnnTF.py

3 different convolutional models have been implemented.

• 7x7 Conv with stride 2
• Relu activation
• Fully connected that map to 10 output classes

Results:

• 7x7 Conv with stride 2
• Relu activation
• 2x2 Max Pooling
• Fully connected with 1024 hidden neurons § Relu activation
• Fully connected that map to 10 output classes

Results:

• 7x7 Conv with stride 2
• Relu activation
• 2x2 Max Pooling
• 5x5 Conv with stride 1
• Relu activation
• 2x2 Max Pooling
• Fully connected with 1024 hidden neurons
• Relu activation
• Fully connected with 128 hidden neurons
• Relu activation
• Fully connected that map to 10 output classes

Results:

Since I have a MAC book without GPU this entire program takes a long time to run (about 8-9 hours).