pydl
is built for understanding dl. Most of the calculations make
use of numpy. It may not be a good idea to use this toy framework
for solving your problem in real environment.
- PickleData
- ReLU
- Sigmoid
- TanH
- Convolution
- Pooling
- NaiveRNN
- LSTM
- SoftMaxLoss
- CrossEntropyLoss
- SigmoidCrossEntropyLoss
- FullyConnect
- Accuracy
- SoftMax
- Dropout
- Normalization
- NaiveSGD
- Adam