[GitHub] jeremiedb commented on issue #9803: R Metrics

jeremiedb commented on issue #9803: R Metrics URL: https://github.com/apache/incubator-mxnet/pull/9803#issuecomment-366425707 Benchmark performed on AWS K80. Going to NDArray has no significant speed modification on MNIST: Original: ``` >

[GitHub] marcoabreu commented on issue #9791: CI test randomness 3

marcoabreu commented on issue #9791: CI test randomness 3 URL: https://github.com/apache/incubator-mxnet/pull/9791#issuecomment-366429224 Approved for merge. Please resolve the test failure if possible. This is an automated

[GitHub] marcoabreu commented on issue #9815: Equivalent of per_process_gpu_memory_fraction(TF) in MXNet

marcoabreu commented on issue #9815: Equivalent of per_process_gpu_memory_fraction(TF) in MXNet URL: https://github.com/apache/incubator-mxnet/issues/9815#issuecomment-366429325 @KellenSunderland @larroy This is an

[GitHub] marcoabreu commented on a change in pull request #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN

marcoabreu commented on a change in pull request #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN URL: https://github.com/apache/incubator-mxnet/pull/9810#discussion_r168917395 ## File path: MKL_README.md ## @@ -17,46 +17,3 @@ Installing and enabling the

[GitHub] stonedl3 commented on issue #9766: DeepLearning on Imagenet with mxnet issues translating .lst to .rec files

stonedl3 commented on issue #9766: DeepLearning on Imagenet with mxnet issues translating .lst to .rec files URL: https://github.com/apache/incubator-mxnet/issues/9766#issuecomment-366445508 Here is the script used to train alexnet # USAGE # python train_alexnet.py --checkpoints

[GitHub] tholcman opened a new issue #9817: Example SSD demo doesn't work

tholcman opened a new issue #9817: Example SSD demo doesn't work URL: https://github.com/apache/incubator-mxnet/issues/9817 I have trained network with command ``` $ python train.py --train-path data/train.rec --val-path data/val.rec --class-names 1,2 --num-class 2 --end-epoch

[GitHub] cjolivier01 commented on a change in pull request #9801: Set worker thread to use OMP when necessary (and not to when not nece?

cjolivier01 commented on a change in pull request #9801: Set worker thread to use OMP when necessary (and not to when not nece? URL: https://github.com/apache/incubator-mxnet/pull/9801#discussion_r168932848 ## File path: src/engine/threaded_engine_perdevice.cc ## @@

[GitHub] chsin commented on issue #9809: register optimizers only once in CPP-Package

chsin commented on issue #9809: register optimizers only once in CPP-Package URL: https://github.com/apache/incubator-mxnet/pull/9809#issuecomment-366472582 @marcoabreu, I don't understand how the Python test would fail when I never made an edit to the Python package? Could you explain?

[incubator-mxnet] branch master updated: Add Support for int64 (#9738)

This is an automated email from the ASF dual-hosted git repository. jxie pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/master by this push: new 44ead24 Add Support for int64 (#9738)

[GitHub] piiswrong commented on a change in pull request #9801: Set worker thread to use OMP when necessary (and not to when not nece?

piiswrong commented on a change in pull request #9801: Set worker thread to use OMP when necessary (and not to when not nece? URL: https://github.com/apache/incubator-mxnet/pull/9801#discussion_r168931945 ## File path: src/engine/threaded_engine_perdevice.cc ## @@ -264,9

[GitHub] ArmageddonKnight commented on issue #9811: How to reinitialize the begin_state of an RNN

ArmageddonKnight commented on issue #9811: How to reinitialize the begin_state of an RNN URL: https://github.com/apache/incubator-mxnet/issues/9811#issuecomment-366463858 @eric-haibin-lin Thanks. This is what I have been looking for.

[GitHub] ArmageddonKnight closed issue #9811: How to reinitialize the begin_state of an RNN

ArmageddonKnight closed issue #9811: How to reinitialize the begin_state of an RNN URL: https://github.com/apache/incubator-mxnet/issues/9811 This is an automated message from the Apache Git Service. To respond to the

[GitHub] szha commented on a change in pull request #9492: fix print_summary bug and add groups of convolution

szha commented on a change in pull request #9492: fix print_summary bug and add groups of convolution URL: https://github.com/apache/incubator-mxnet/pull/9492#discussion_r168931667 ## File path: python/mxnet/visualization.py ## @@ -134,20 +134,26 @@ def

[GitHub] piiswrong closed pull request #9738: Add Support for int64

piiswrong closed pull request #9738: Add Support for int64 URL: https://github.com/apache/incubator-mxnet/pull/9738 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance: As this is a foreign pull

[GitHub] piiswrong commented on a change in pull request #9807: temporary solution for instancenorm, will refactor using backend

piiswrong commented on a change in pull request #9807: temporary solution for instancenorm, will refactor using backend URL: https://github.com/apache/incubator-mxnet/pull/9807#discussion_r168931896 ## File path: python/mxnet/gluon/nn/basic_layers.py ## @@ -463,11 +468,12

[GitHub] chsin commented on issue #9809: register optimizers only once in CPP-Package

chsin commented on issue #9809: register optimizers only once in CPP-Package URL: https://github.com/apache/incubator-mxnet/pull/9809#issuecomment-366472791 @marcoabreu, I never touched the Python and Gluon package, why do I get this: ```

[GitHub] szha opened a new pull request #9818: add MUSE embeddings

szha opened a new pull request #9818: add MUSE embeddings URL: https://github.com/apache/incubator-mxnet/pull/9818 ## Description ## Add Multilingual Unsupervised and Supervised Embeddings (MUSE) embeddings. ## Checklist ## ### Essentials ### - [x] Passed code style checking

[GitHub] szha commented on a change in pull request #9747: Add contrib.rand_log_uniform

szha commented on a change in pull request #9747: Add contrib.rand_log_uniform URL: https://github.com/apache/incubator-mxnet/pull/9747#discussion_r168930990 ## File path: python/mxnet/ndarray/contrib.py ## @@ -18,9 +18,81 @@ # coding: utf-8 # pylint:

[GitHub] piiswrong commented on a change in pull request #9790: make array.reshape compatible with numpy

piiswrong commented on a change in pull request #9790: make array.reshape compatible with numpy URL: https://github.com/apache/incubator-mxnet/pull/9790#discussion_r168931845 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -968,6 +973,17 @@ def reshape(self, shape):

[GitHub] cjolivier01 commented on issue #9801: Set worker thread to use OMP when necessary (and not to when not nece?

cjolivier01 commented on issue #9801: Set worker thread to use OMP when necessary (and not to when not nece? URL: https://github.com/apache/incubator-mxnet/pull/9801#issuecomment-366483979 don?t merge this yet. i want to make a unit test.

[GitHub] moveforever opened a new issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly?

moveforever opened a new issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly? URL: https://github.com/apache/incubator-mxnet/issues/9819 I have implenmented a DataIter of C++ supporting multi data input. Sometime, usage of cpu loading data is low, that is less

[GitHub] dwSun commented on issue #9614: MobileNetV2

dwSun commented on issue #9614: MobileNetV2 URL: https://github.com/apache/incubator-mxnet/pull/9614#issuecomment-366483368 Add relu in first conv2d as requested. This is an automated message from the Apache Git Service. To

[GitHub] chinakook commented on a change in pull request #9492: fix print_summary bug and add groups of convolution

chinakook commented on a change in pull request #9492: fix print_summary bug and add groups of convolution URL: https://github.com/apache/incubator-mxnet/pull/9492#discussion_r168936777 ## File path: python/mxnet/visualization.py ## @@ -134,20 +134,26 @@ def

[GitHub] ascust commented on issue #9688: Adapt operators from PyTorch, will keep adding

ascust commented on issue #9688: Adapt operators from PyTorch, will keep adding URL: https://github.com/apache/incubator-mxnet/pull/9688#issuecomment-366481984 I think it would be good that we can pass a reference symbol as a target shape. For example like mx.sym.Crop, where you can crop