in_channels must equal to previous n_filter

tensorlayer / TensorLayer

Deep Learning and Reinforcement Learning Library for Scientists and Engineers

http://tensorlayerx.com

Other

7.33k stars 1.61k forks source link

in_channels must equal to previous n_filter #1078

Closed tiancheng2000 closed 4 years ago

tiancheng2000 commented 4 years ago

Checklist

[x] I've read the Contribution Guidelines
[x] I've updated the documentation if necessary.
[x] I've tested that my changes are compatible with the latest version of Tensorflow.

Motivation and Context

Simple modification for rubustness of code. Otherwise all n_filters configured in a same layer_group must keep same, which is not neccessary. Tested.

Description

Simple modification. Please refer to the diff.

zsdonghao commented 4 years ago

please update the changelog and make the travis pass

tiancheng2000 commented 4 years ago

Travis-CI is complaining about RNN test failure, after I modified changelog?...

=========================== short test summary info ============================
FAILED tests/layers/test_layers_recurrent.py::Layer_RNN_Test::test_basic_simplernn_dropout_1
============ 1 failed, 229 passed, 10 warnings in 347.14s (0:05:47) ============