out of memory when using BatchNormalization on convolutional layers · Issue #6661 · keras-team/keras
Please make sure that the boxes below are checked before you submit your issue. If your issue is an implementation question, please ask your question on StackOverflow or join the Keras Slack channel and ask there instead of filing a GitHub issue. Thank you! [v ] Check that you are up-to-date with the master branch of Keras. You can update with: pip install git+git://github.com/fchollet/keras.git -
Refactor regularizers and add add_weight method. by fchollet · Pull Request #4703 · keras-team/keras
Keras 2.0 release notes
The nb_epoch argument has been renamed epochs everywhere. The methods fit_generator, evaluate_generator and predict_generator now work by drawing a number of batches from a generator (number of training steps), rather than a number of samples. samples_per_epoch was changed to steps_per_epoch in fit_generator. It now refers to the number of batches an epoch is considered as done. nb_val_samples was
fit_generator converges slower and tensorboard's OOM problem · Issue #4797 · keras-team/keras
problem 1: i changed my code from fit to fit_generator, however, i found it converged slower than using fit. Using fit it converges at 5th epoch, however, it doesn't converge even at 10th epoch when i change to fit_generator. Moreover, when validation_data also uses fit_generator, the performance becomes much worse. Is my generator function has errors? def batch_iter(x, y, batch_size): """ Generat
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
Amazon.co.jp: Pythonによるディープラーニング (Compass Booksシリーズ): François Chollet (著), 株式会社クイープ (翻訳), 巣籠悠輔 (その他): 本
Issues in Keras model loading in Tensorflow Serving · Issue #2310 · keras-team/keras
Batch Normalization Problem · Issue #1446 · keras-team/keras
changing loss weight during training · Issue #6446 · keras-team/keras
changeable loss weights for multiple output · Issue #2595 · keras-team/keras
LSTM Autoencoder · Issue #1401 · keras-team/keras
Does it make sense that the validation loss is lower than training loss? · Issue #472 · keras-team/keras
keras/topology.py at master · keras-team/keras · GitHub
Incorporating Word Vectors rather than using an Embedding class · Issue #853 · keras-team/keras
Keras use only single thread · Issue #1245 · keras-team/keras
None is not supported in input_shape [bug] · Issue #5900 · keras-team/keras![None is not supported in input_shape [bug] · Issue #5900 · keras-team/keras](https://cdn-ak-scissors.b.st-hatena.com/image/square/f3b23ea18582763f7fe3950c8b608e59cbecd218/height=288;version=1;width=512/https%3A%2F%2Fopengraph.githubassets.com%2Fbe4c96e0b64dc5304fb220281174e37548224728be1cfc311301231353dcb379%2Fkeras-team%2Fkeras%2Fissues%2F5900)
A concrete example for using data generator for large datasets such as ImageNet · Issue #1627 · keras-team/keras
keras/mnist_mlp.py at master · keras-team/keras · GitHub
Memory leak during model.fit() · Issue #5935 · keras-team/keras
How to obtain the gradient of each parameter in the last epoch of training · Issue #2226 · keras-team/keras
How to add Attention on top of a Recurrent Layer (Text Classification) · Issue #4962 · keras-team/keras
