awslabs/keras-apache-mxnet issues and pull requests

#272 - Keras-MXNet directly crash when the hidden layer size of LSTM is 1

Issue - State: open - Opened by maybeLee about 3 years ago

#271 - [Bug] Keras misconsiders the output of bias initialization method in dense layer as a scalar instead of a vector

Issue - State: open - Opened by maybeLee about 3 years ago

#270 - Corrected grammatical error

Pull Request - State: closed - Opened by neel2299 about 3 years ago

#269 - NotImplementedError: Separable Conv1D is not supported in current MXNet

Issue - State: open - Opened by GundamBen over 3 years ago

#268 - Fixed test_sequential_lstm_mxnet_model_saving_with_sparse_embedding f…

Pull Request - State: closed - Opened by yashsaboo almost 4 years ago - 1 comment

#267 - Weighted Loss Function feature requirement

Issue - State: open - Opened by kdgutier almost 4 years ago

#266 - fix ci

Pull Request - State: closed - Opened by roywei over 4 years ago

#265 - Enabled the DeprecationWarning to show always

Pull Request - State: closed - Opened by karan6181 over 4 years ago

#264 - Added a package deprecation message

Pull Request - State: closed - Opened by karan6181 over 4 years ago

#263 - Remove redundant code and update the unit test for stack function

Pull Request - State: closed - Opened by pjelement over 4 years ago

#262 - Stack

Pull Request - State: closed - Opened by pjelement over 4 years ago - 2 comments

#261 - Adding stack support for mxnet backend

Pull Request - State: closed - Opened by pjelement over 4 years ago - 1 comment

#260 - use keras pretrained model, save_mxnet_model error

Issue - State: open - Opened by Juldeng over 4 years ago

#259 - disable theano test

Pull Request - State: closed - Opened by roywei over 4 years ago

#258 - Fix axis error in normalization layer when loading model from tf backend saved h5

Pull Request - State: closed - Opened by leondgarse over 4 years ago - 7 comments

#257 - Inconsistency between the result of model.predict() and the output value of the last layer of the model

Issue - State: open - Opened by Justobe over 4 years ago

#256 - mxnet.base.MXNetError: Error in operator transpose176: [02:14:06] src/operator/tensor/./matrix_op-inl.h:354: Check failed: shp.ndim() == param.axes.ndim() (-1 vs. 4)

Issue - State: open - Opened by Justobe over 4 years ago

#255 - fix gpu detection

Pull Request - State: closed - Opened by roywei over 4 years ago - 2 comments

#254 - fix kwargs and add comments

Pull Request - State: closed - Opened by roywei almost 5 years ago

#253 - ConvTranspose3D with 2 dim

Issue - State: open - Opened by JCuomo almost 5 years ago

#252 - fix batchnorm freeze mean and var

Pull Request - State: closed - Opened by roywei almost 5 years ago - 1 comment

#251 - Issues with setting MXNet context

Issue - State: closed - Opened by roywei almost 5 years ago - 1 comment
Labels: bug

#250 - fix set MXNet context

Pull Request - State: closed - Opened by roywei almost 5 years ago - 1 comment

#249 - update to python 3.7

Pull Request - State: closed - Opened by roywei almost 5 years ago

#248 - ping tf version in ci

Pull Request - State: closed - Opened by roywei almost 5 years ago

#247 - Keras switch (K.switch) with mxnet backend

Issue - State: closed - Opened by kirimaru-jp almost 5 years ago - 1 comment

#246 - Future of MXNet backend?

Issue - State: open - Opened by ssbusc1 about 5 years ago

#245 - new release

Pull Request - State: closed - Opened by roywei about 5 years ago - 1 comment

#244 - fix test dataset

Pull Request - State: closed - Opened by roywei about 5 years ago - 1 comment

#243 - Regression with MxNet 1.5.0: Cannot apply softmax to a tensor that is 1D. Received input: Symbol: broadcast_add30

Issue - State: closed - Opened by iamthebot about 5 years ago - 1 comment

#242 - Accuracy not improving for MNIST example with MXnet backend

Issue - State: closed - Opened by jackgle about 5 years ago - 1 comment

#241 - MXBoard with keras-mxnet

Issue - State: closed - Opened by sdbonte about 5 years ago - 1 comment

#240 - Conv3D in MXNet: error with Tensor shape

Issue - State: open - Opened by sdbonte about 5 years ago
Labels: bug

#239 - ping pytest version for pr build

Pull Request - State: closed - Opened by roywei about 5 years ago

#238 - [Test WebHook][Do not merge]Revert "ping pytest version in ci"

Pull Request - State: closed - Opened by roywei about 5 years ago - 1 comment

#237 - ping pytest version in ci

Pull Request - State: closed - Opened by roywei about 5 years ago - 1 comment

#236 - Support for Amazon Elastic Inference for prediction - Keras and MXNet backend

Pull Request - State: closed - Opened by sandeep-krishnamurthy over 5 years ago - 5 comments

#235 - InferShape cannot decide shapes when adding a Dropout layers

Issue - State: open - Opened by gregoirerllr over 5 years ago - 1 comment

#234 - update keras-mxnet pip packages

Issue - State: closed - Opened by roywei over 5 years ago - 1 comment

#233 - Softmax operator shape inference

Issue - State: open - Opened by roywei over 5 years ago
Labels: training

#232 - fix softmax shape

Pull Request - State: closed - Opened by roywei over 5 years ago - 3 comments

#231 - Different layer input/output shapes between tensorflow and mxnet backend

Issue - State: open - Opened by KocsisV over 5 years ago

#230 - change how mxnet backend detect gpus

Issue - State: closed - Opened by roywei over 5 years ago - 1 comment

#229 - Ignoring masking layer test for RNN with MXNet backend

Pull Request - State: closed - Opened by karan6181 over 5 years ago

#228 - Masking Layer doesn't work after adding a NaiveRunGraph feature in MXNet

Issue - State: open - Opened by karan6181 over 5 years ago - 2 comments

#227 - Unable to export to mxnet (unless immediately after training)

Issue - State: closed - Opened by 4sfaloth over 5 years ago - 2 comments
Labels: bug, good first issue

#226 - Old versions of mxnet and keras-mxnet not compatible

Issue - State: open - Opened by roywei over 5 years ago
Labels: bug

#225 - fix nightly on sgd

Pull Request - State: closed - Opened by roywei over 5 years ago - 3 comments

#224 - Use of constant tensor in custom loss fails

Issue - State: open - Opened by efournie over 5 years ago - 4 comments
Labels: feature request

#223 - Composing Models does not work with the MXNet backend

Issue - State: open - Opened by ssbusc1 over 5 years ago - 2 comments
Labels: feature request, training

#218 - no longer require --pre for 1.3.1

Pull Request - State: closed - Opened by roywei almost 6 years ago

#210 - Optimizer initialization should set rescale_grad appropriately

Issue - State: open - Opened by kalyc almost 6 years ago
Labels: enhancement, feature request

#209 - Add rerun for failed flaky test

Issue - State: open - Opened by roywei almost 6 years ago
Labels: CI

#208 - Coding Language Detection Example

Pull Request - State: closed - Opened by harshp8l almost 6 years ago - 2 comments