danijar layered issues - Githubissues

danijar / layered

Clean implementation of feed forward neural networks

MIT License

240 stars 33 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Question about Softmax delta function

#30 howardyclo opened 8 years ago
0
Fix spelling of gradient descent

#29 danijar opened 8 years ago
0
is there an example use simple data like iris to run this code

#28 nlp-study closed 8 years ago
4
Color training costs based on label value

#27 danijar opened 8 years ago
4
syntax error in tutorial

#26 ebin123456 closed 8 years ago
0
Keep window open after training

#25 danijar closed 8 years ago
0
Matplotlib is super slow on Mac

#24 danijar opened 8 years ago
2
Rename squared cost to SquaredError

#23 danijar closed 8 years ago
0
Rename linear activation function to identity

#22 danijar closed 8 years ago
0
Add descriptions to activation and cost function tables in readme

#21 danijar closed 8 years ago
0
Plot not showing up inside virtualenv on Linux

#20 danijar closed 8 years ago
1
Plot window is not responding on Mac

#19 danijar closed 8 years ago
0
Test activation and cost function deltas against their numeric gradients

#18 danijar closed 8 years ago
1
Update readme to include quickstart explaining features

#17 danijar closed 8 years ago
0
Extract logic from main file into a class

#16 danijar closed 8 years ago
0
Dump evaluation results

#15 danijar opened 8 years ago
0
Update readme since downloading datasets is no longer needed

#14 danijar closed 8 years ago
0
Set up Sphinx documentation

#13 danijar closed 8 years ago
0
Test installation from PyPi

#12 danijar closed 8 years ago
0
Store and load models and learned weights

#11 danijar closed 8 years ago
1
Set up Pylint on Travis

#10 danijar closed 8 years ago
0
Plot crashes on exit

#9 danijar closed 8 years ago
1
Rename rounds on training set to epochs

#8 danijar closed 8 years ago
0
Plot evaluation results

#7 danijar closed 8 years ago
1
Implement batch backpropagation as matrix times matrix multiplication

#6 danijar closed 8 years ago
1
Split user and testing dependencies for faster Travis builds

#5 danijar closed 8 years ago
0
Add tests to validate activation and cost functions against their numerical gradient

#4 danijar opened 8 years ago
0
Set up Travis CI integration

#3 danijar closed 8 years ago
0
Automatically download and cache datasets

#2 danijar closed 8 years ago
0
Average or just add up gradients in batch backpropagation?

#1 danijar closed 8 years ago
1