[GitHub] zhenglaizhang commented on issue #9859: How to run model trained with mxnet 1.0 on android

zhenglaizhang commented on issue #9859: How to run model trained with mxnet 1.0 on android URL: https://github.com/apache/incubator-mxnet/issues/9859#issuecomment-368206145 @larroy thanks for the reply, I have succeeded in running the trained model with amalgamation build after the model

[GitHub] zhenglaizhang closed issue #9859: How to run model trained with mxnet 1.0 on android

zhenglaizhang closed issue #9859: How to run model trained with mxnet 1.0 on android URL: https://github.com/apache/incubator-mxnet/issues/9859 This is an automated message from the Apache Git Service. To respond to the

[GitHub] anirudh2290 commented on issue #9813: Unable to save gluon model to symbolic network : neural style

anirudh2290 commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368204565 @samhodge the model makes calls to operators implemented in the C++ backend and leverages the backend

[GitHub] sxjscience commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly?

sxjscience commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly? URL: https://github.com/apache/incubator-mxnet/issues/9819#issuecomment-368203841 @piiswrong What do you think? I feel this type of problems should not be raised as an issue.

[GitHub] JulianSlzr opened a new pull request #9877: Better even_split=False support in gluon.split_data()

JulianSlzr opened a new pull request #9877: Better even_split=False support in gluon.split_data() URL: https://github.com/apache/incubator-mxnet/pull/9877 ## Description ## When someone uses `even_split=False`, it means they want the data to be processed irrespective of how divisible

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368200989 If I want to run this model in C++ it might make more sense to implement the network parts in C++ also so I

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368200910 OK a minor update to generalise the model to the width and height of the input

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368200661 OK now I have discovered the limit of this network it can only be saved as a symbolic network if the

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368199853 The hack ``` def print_summary(symbol, shape=None, line_length=120,

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368199481 Now I have a new error ```

[GitHub] juliusshufan commented on issue #9866: The default weight initialization strategy makes the VGG network difficult to converge when utilizing examples under 'example/image-classification'

juliusshufan commented on issue #9866: The default weight initialization strategy makes the VGG network difficult to converge when utilizing examples under 'example/image-classification' URL: https://github.com/apache/incubator-mxnet/issues/9866#issuecomment-368199338 @pengzhao-intel ,

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368198997 OK I have now added arguments to infer_shape_partial because without it I wasnt getting any values for the

[GitHub] szha closed issue #9871: cannot import name 'registry'

szha closed issue #9871: cannot import name 'registry' URL: https://github.com/apache/incubator-mxnet/issues/9871 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and

[GitHub] szha commented on issue #9871: cannot import name 'registry'

szha commented on issue #9871: cannot import name 'registry' URL: https://github.com/apache/incubator-mxnet/issues/9871#issuecomment-368198240 No problem, don't worry. Let us know what else we may help with. Closing the issue now.

[GitHub] samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style

samhodge commented on issue #9813: Unable to save gluon model to symbolic network : neural style URL: https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-368197281 I found this a little bit of help: https://github.com/apache/incubator-mxnet/issues/1337

[GitHub] eric-haibin-lin commented on a change in pull request #9625: sparse regression operators

eric-haibin-lin commented on a change in pull request #9625: sparse regression operators URL: https://github.com/apache/incubator-mxnet/pull/9625#discussion_r170409181 ## File path: src/operator/regression_output-inl.h ## @@ -121,6 +175,67 @@ void RegressionBackward(const

[GitHub] bitdata commented on issue #9871: cannot import name 'registry'

bitdata commented on issue #9871: cannot import name 'registry' URL: https://github.com/apache/incubator-mxnet/issues/9871#issuecomment-368194071 Finding that a lot of files are lost in site-packages\mxnet, I uninstall and reinstall mxnet. Then error message goes away. Thanks a lot,

[GitHub] moveforever commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly?

moveforever commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly? URL: https://github.com/apache/incubator-mxnet/issues/9819#issuecomment-368192685 @sxjscience ? i implement dataiter supporting multi input inheritated MXDataIter. The people of

[GitHub] solin319 commented on issue #8373: distribute training in fp16

solin319 commented on issue #8373: distribute training in fp16 URL: https://github.com/apache/incubator-mxnet/pull/8373#issuecomment-368188760 @rahul003 For alexnet, try to use fp16 with GPU in kvstore_dist_server. For resnet, try to use dist_sync.

[GitHub] BiranLi commented on issue #8373: distribute training in fp16

BiranLi commented on issue #8373: distribute training in fp16 URL: https://github.com/apache/incubator-mxnet/pull/8373#issuecomment-368188462 @rahul003 Because of the limited range of expression of FP16, gradient diffusion occurs in BP. The easiest way to handle this is to scale the

[GitHub] rahul003 commented on issue #8373: distribute training in fp16

rahul003 commented on issue #8373: distribute training in fp16 URL: https://github.com/apache/incubator-mxnet/pull/8373#issuecomment-368187293 @solin319 It was being trained in fp16 too. I also tried implementing a system which just converts data to fp32 in kv_dist and sends weights as

[GitHub] BiranLi commented on issue #8373: distribute training in fp16

BiranLi commented on issue #8373: distribute training in fp16 URL: https://github.com/apache/incubator-mxnet/pull/8373#issuecomment-368186682 @solin319 Is it possible to consider gradient diffusion in computational calculations, such as adding a grad_scale processing interface?

[GitHub] solin319 commented on issue #8373: distribute training in fp16

solin319 commented on issue #8373: distribute training in fp16 URL: https://github.com/apache/incubator-mxnet/pull/8373#issuecomment-368185953 Witch data type were used in training? We ues fp16 in training computation. @rahul003

[GitHub] feevos commented on issue #9822: gluon HybridBlock wrapper of constant nd.array, is it possible?

feevos commented on issue #9822: gluon HybridBlock wrapper of constant nd.array, is it possible? URL: https://github.com/apache/incubator-mxnet/issues/9822#issuecomment-367901846 Thanks @jmacglashan , based on your suggestions I tried the following solution: I am trying to

[GitHub] pengzhao-intel commented on issue #9866: The default weight initialization strategy makes the VGG network difficult to converge when utilizing examples under 'example/image-classification'

pengzhao-intel commented on issue #9866: The default weight initialization strategy makes the VGG network difficult to converge when utilizing examples under 'example/image-classification' URL: https://github.com/apache/incubator-mxnet/issues/9866#issuecomment-368178001 @juliusshufan

[GitHub] aaronmarkham commented on a change in pull request #9876: temporary fix to update the verionsioning of 1.1.0 that is skipped du?

aaronmarkham commented on a change in pull request #9876: temporary fix to update the verionsioning of 1.1.0 that is skipped du? URL: https://github.com/apache/incubator-mxnet/pull/9876#discussion_r170399617 ## File path: docs/build_version_doc/build_doc.sh ## @@ -49,12

[GitHub] anirudh2290 commented on issue #9868: MKL and CMake

anirudh2290 commented on issue #9868: MKL and CMake URL: https://github.com/apache/incubator-mxnet/issues/9868#issuecomment-368172335 I think it is safe to set USE_MKLML_MKL to be same as USE_MKLDNN. @cjolivier01 can you please confirm ?

[GitHub] cjolivier01 closed pull request #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 closed pull request #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875 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

[incubator-mxnet] branch master updated: Fixed indentation and added test outputs to gitignore (#9875)

This is an automated email from the ASF dual-hosted git repository. cjolivier01 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 fbbc080 Fixed indentation and

[GitHub] thinksanky opened a new pull request #9876: temporary fix to update the verionsioning of 1.1.0 that is skipped du?

thinksanky opened a new pull request #9876: temporary fix to update the verionsioning of 1.1.0 that is skipped du? URL: https://github.com/apache/incubator-mxnet/pull/9876 ## Description ## Because of missed process in building new versioned mxnet we are in a situation that the

[GitHub] cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368157229 I would prefer we didn't encourage building from source tree for just that reason, but okay. After all, it is legal.

[GitHub] cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368157229 I would prefer we didn't encourage building from source tree for just that reason, but okay.

[GitHub] cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368157229 I would prefer we didn't encourage building from source for just that reason, but okay.

[GitHub] dabraude commented on issue #9875: Fixed indentation and added test outputs to gitignore

dabraude commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368155354 On my laptop I build out of source (../build/Release|Debug etc), but our test server builds in source, that's how I

[GitHub] cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368152900 are you build cmake Makefiles in the source directory? Generally you do it in a subdirectory. i.e.: ```bash

[GitHub] cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore

cjolivier01 commented on issue #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875#issuecomment-368152900 are you build cmake Makefiles in the source directory? Generally you do it in a subdirectory. i.e.: ```bash

[GitHub] dabraude opened a new pull request #9875: Fixed indentation and added test outputs to gitignore

dabraude opened a new pull request #9875: Fixed indentation and added test outputs to gitignore URL: https://github.com/apache/incubator-mxnet/pull/9875 Fixed some indentation in the CMakeLists.txt Added the outputs of the tests to the .gitignore for in-source builds.

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170375469 ## File path: src/kvstore/comm.h ## @@ -406,12 +422,12 @@ class CommCPU : public Comm { });

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170375348 ## File path: src/kvstore/comm.h ## @@ -523,95 +537,236 @@ class CommDevice : public Comm { }

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170374876 ## File path: src/kvstore/comm.h ## @@ -178,85 +179,94 @@ class CommCPU : public Comm {

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170377029 ## File path: src/kvstore/comm.h ## @@ -523,95 +537,236 @@ class CommDevice : public Comm { }

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170374531 ## File path: src/kvstore/comm.h ## @@ -22,30 +22,29 @@ */ #ifndef MXNET_KVSTORE_COMM_H_ #define

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170374385 ## File path: src/kvstore/comm.h ## @@ -22,30 +22,29 @@ */ #ifndef MXNET_KVSTORE_COMM_H_ #define

[GitHub] piiswrong commented on a change in pull request #8915: NVLink communication pattern updated

piiswrong commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r170374385 ## File path: src/kvstore/comm.h ## @@ -22,30 +22,29 @@ */ #ifndef MXNET_KVSTORE_COMM_H_ #define

[GitHub] Caenorst opened a new issue #9874: ResNet-50 is slower on Volta since #8302

Caenorst opened a new issue #9874: ResNet-50 is slower on Volta since #8302 URL: https://github.com/apache/incubator-mxnet/issues/9874 ## Description I ran the Minimum reproducible example with the setup below at two different version (before and after #8302): Here are the results:

[GitHub] larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170369334 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[GitHub] anirudh2290 commented on a change in pull request #9869: Exception handling documentation

anirudh2290 commented on a change in pull request #9869: Exception handling documentation URL: https://github.com/apache/incubator-mxnet/pull/9869#discussion_r170366189 ## File path: docs/tutorials/basic/exception_handling.md ## @@ -0,0 +1,122 @@ +# Exception Handling in

[GitHub] szha commented on a change in pull request #9869: Exception handling documentation

szha commented on a change in pull request #9869: Exception handling documentation URL: https://github.com/apache/incubator-mxnet/pull/9869#discussion_r170358618 ## File path: docs/tutorials/basic/exception_handling.md ## @@ -0,0 +1,122 @@ +# Exception Handling in MXNet +

[GitHub] szha commented on a change in pull request #9869: Exception handling documentation

szha commented on a change in pull request #9869: Exception handling documentation URL: https://github.com/apache/incubator-mxnet/pull/9869#discussion_r170358355 ## File path: docs/tutorials/basic/exception_handling.md ## @@ -0,0 +1,122 @@ +# Exception Handling in MXNet +

[GitHub] szha commented on a change in pull request #9869: Exception handling documentation

szha commented on a change in pull request #9869: Exception handling documentation URL: https://github.com/apache/incubator-mxnet/pull/9869#discussion_r170358313 ## File path: docs/tutorials/basic/exception_handling.md ## @@ -0,0 +1,122 @@ +# Exception Handling in MXNet +

[GitHub] szha opened a new pull request #9873: add doc for gluon contrib

szha opened a new pull request #9873: add doc for gluon contrib URL: https://github.com/apache/incubator-mxnet/pull/9873 ## Description ## add api doc that was missed in previous PRs. ## Checklist ## ### Essentials ### - [x] Changes are complete (i.e. I finished coding on

[GitHub] sxjscience commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly?

sxjscience commented on issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly? URL: https://github.com/apache/incubator-mxnet/issues/9819#issuecomment-368122186 According the description, the question is not related to MXNet and should not be raised as a

[GitHub] sxjscience closed issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly?

sxjscience closed issue #9819: Sometime MXDataIter load data quickly, sometime it load data slowly? URL: https://github.com/apache/incubator-mxnet/issues/9819 This is an automated message from the Apache Git Service. To

[GitHub] sxjscience commented on issue #9842: Custom Function Shape Inference

sxjscience commented on issue #9842: Custom Function Shape Inference URL: https://github.com/apache/incubator-mxnet/issues/9842#issuecomment-368120785 I think it's not supported. Need to double check with @piiswrong . This

[GitHub] sxjscience commented on issue #9865: Confusing behavior of some evaluation metrics

sxjscience commented on issue #9865: Confusing behavior of some evaluation metrics URL: https://github.com/apache/incubator-mxnet/issues/9865#issuecomment-368120147 I find this is actually a bug. The `pred` will be reshaped if it has ndim=1, see

[GitHub] szha commented on issue #9871: cannot import name 'registry'

szha commented on issue #9871: cannot import name 'registry' URL: https://github.com/apache/incubator-mxnet/issues/9871#issuecomment-368119573 I can't reproduce this on mac using `mxnet==1.1.0b20180213` and don't have a windows machine to reproduce. I tested the same for master branch and

[GitHub] sxjscience commented on issue #9865: Confusing behavior of some evaluation metrics

sxjscience commented on issue #9865: Confusing behavior of some evaluation metrics URL: https://github.com/apache/incubator-mxnet/issues/9865#issuecomment-368117301 Thanks for reporting this, we will improve our documentation and support NDArray as the input.

[GitHub] analog-cbarber commented on issue #8428: Missing constant symbol (equivalent of tf.constant)?

analog-cbarber commented on issue #8428: Missing constant symbol (equivalent of tf.constant)? URL: https://github.com/apache/incubator-mxnet/issues/8428#issuecomment-368111418 As a workaround, you can implement a constant CustomOp. It requires that you serialize the data through the

[GitHub] cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368110885 That's not to say the same trick works for any race condition (fewer cores). But varying the processor affinity

[GitHub] analog-cbarber commented on issue #9822: gluon HybridBlock wrapper of constant nd.array, is it possible?

analog-cbarber commented on issue #9822: gluon HybridBlock wrapper of constant nd.array, is it possible? URL: https://github.com/apache/incubator-mxnet/issues/9822#issuecomment-368109593 It is surprising here isn't already support for constants (see #8428). It is possible to implement a

[GitHub] cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368109559 @zheng-da Well, you can vary the cores up or down just to change the timing and try to make the race condition

[GitHub] sxjscience commented on issue #9872: A bug in an example in the python API document

sxjscience commented on issue #9872: A bug in an example in the python API document URL: https://github.com/apache/incubator-mxnet/issues/9872#issuecomment-368107374 The gradient should be ograd * layer_grad and should be correct. `dy` means the gradient passed by the next layer and

[GitHub] sxjscience closed issue #9872: A bug in an example in the python API document

sxjscience closed issue #9872: A bug in an example in the python API document URL: https://github.com/apache/incubator-mxnet/issues/9872 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368098057 @larroy this is the design doc of mkldnn:

[GitHub] zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368098057 @larroy not yet. this is the design doc of mkldnn:

[GitHub] zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368095680 @cjolivier01 why race condition happens more frequently when threads run in a smaller number of CPU cores? It seems

[GitHub] zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

zheng-da commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368095680 @cjolivier01 why race condition happens more frequently when threads run in a smaller number of CPU cores? It seems

[GitHub] zheng-da commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

zheng-da commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170326873 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[incubator-mxnet] branch master updated: New example of custom operator using RTC (#9870)

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 b23e0a9 New example of custom operator

[GitHub] piiswrong closed pull request #9870: New example of custom operator using RTC

piiswrong closed pull request #9870: New example of custom operator using RTC URL: https://github.com/apache/incubator-mxnet/pull/9870 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

[GitHub] piiswrong commented on issue #9870: New example of custom operator using RTC

piiswrong commented on issue #9870: New example of custom operator using RTC URL: https://github.com/apache/incubator-mxnet/pull/9870#issuecomment-368093220 great example. Thanks! This is an automated message from the Apache

[GitHub] chsin commented on issue #9809: fix optimizer bug in CPP-Package

chsin commented on issue #9809: fix optimizer bug in CPP-Package URL: https://github.com/apache/incubator-mxnet/pull/9809#issuecomment-368090106 But we'll still have the weirdness of initializing with Find that @szha is unhappy about..

[GitHub] ashokei commented on issue #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN

ashokei commented on issue #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN URL: https://github.com/apache/incubator-mxnet/pull/9810#issuecomment-368081631 @marcoabreu jenkins issue is fixed, it passes the build.

[GitHub] marcoabreu commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

marcoabreu commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170297168 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[GitHub] larroy commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

larroy commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368051882 Is this need to reorder data documented somewhere?

[GitHub] cjolivier01 commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

cjolivier01 commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170288951 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[GitHub] larroy commented on issue #9859: How to run model trained with mxnet 1.0 on android

larroy commented on issue #9859: How to run model trained with mxnet 1.0 on android URL: https://github.com/apache/incubator-mxnet/issues/9859#issuecomment-368036876 We haven't had time to invest on figuring out why the amalgamation build is smaller than the arm64 docker build. I think

[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_r170274912 ## File path: Jenkinsfile ## @@ -24,6 +24,7 @@ mx_lib = 'lib/libmxnet.so,

[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_r170274815 ## File path: Jenkinsfile ## @@ -24,6 +24,7 @@ mx_lib = 'lib/libmxnet.so,

[GitHub] dotelos opened a new issue #9872: A bug in an example in the python API document

dotelos opened a new issue #9872: A bug in an example in the python API document URL: https://github.com/apache/incubator-mxnet/issues/9872 This is an example found in the doc for

[GitHub] cjolivier01 commented on a change in pull request #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN

cjolivier01 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_r170273454 ## File path: Jenkinsfile ## @@ -24,6 +24,7 @@ mx_lib = 'lib/libmxnet.so,

[GitHub] larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170271059 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[GitHub] larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170141588 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[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_r170268616 ## File path: Jenkinsfile ## @@ -24,6 +24,7 @@ mx_lib = 'lib/libmxnet.so,

[GitHub] asitstands commented on issue #9847: Imperative regression output layers are broken

asitstands commented on issue #9847: Imperative regression output layers are broken URL: https://github.com/apache/incubator-mxnet/issues/9847#issuecomment-367594510 @ZiyueHuang Thanks, the fix works. For the `waitall` call, why do we need explicit synchronization here? In my machine the

[GitHub] eric-haibin-lin closed pull request #9747: Add contrib.rand_zipfian

eric-haibin-lin closed pull request #9747: Add contrib.rand_zipfian URL: https://github.com/apache/incubator-mxnet/pull/9747 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

[incubator-mxnet] branch master updated: Add contrib.rand_zipfian (#9747)

This is an automated email from the ASF dual-hosted git repository. haibin 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 9158352 Add contrib.rand_zipfian

[GitHub] cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN.

cjolivier01 commented on issue #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#issuecomment-368013885 one thing i?ve done in the past to try and make race conditions happen more often is start changing the process

[GitHub] cjolivier01 commented on issue #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN

cjolivier01 commented on issue #9810: remove MKL_EXPERIMENTAL and update make files for MKL-DNN URL: https://github.com/apache/incubator-mxnet/pull/9810#issuecomment-368012372 which files exactly? On Fri, Feb 23, 2018 at 1:19 AM Marco de Abreu wrote:

[GitHub] kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader

kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader URL: https://github.com/apache/incubator-mxnet/issues/9259#issuecomment-367998899 This problem disappeared on the latest version.

[GitHub] kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader

kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader URL: https://github.com/apache/incubator-mxnet/issues/9259#issuecomment-367998899 This problem disappeared in the latest version.

[GitHub] kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader

kohillyang commented on issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader URL: https://github.com/apache/incubator-mxnet/issues/9259#issuecomment-367998899 This problem disappeared in latest version.

[GitHub] kohillyang closed issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader

kohillyang closed issue #9259: Shared memory leak when using mxnet.gluon.data.DataLoader URL: https://github.com/apache/incubator-mxnet/issues/9259 This is an automated message from the Apache Git Service. To respond to the

[incubator-mxnet-site] branch test-revert created (now c989156)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch test-revert in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git. at c989156 Reverting state before execution of faulty website build This branch includes the

[GitHub] larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN.

larroy commented on a change in pull request #9862: Fix a race condition in converting data layouts in MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/9862#discussion_r170227496 ## File path: src/ndarray/ndarray.cc ## @@ -1017,6 +1017,7 @@ inline void

[GitHub] bitdata opened a new issue #9871: cannot import name 'registry'

bitdata opened a new issue #9871: cannot import name 'registry' URL: https://github.com/apache/incubator-mxnet/issues/9871 ## Description error while import mxnet.contrib.text ## Environment info mxnet-cu901.1.0b20180213 -Python

[GitHub] marcoabreu opened a new pull request #53: Revert "fixed the release number to 1.1.0"

marcoabreu opened a new pull request #53: Revert "fixed the release number to 1.1.0" URL: https://github.com/apache/incubator-mxnet-site/pull/53 Reverts apache/incubator-mxnet-site#52 This is an automated message from the

[incubator-mxnet-site] 01/01: Merge pull request #53 from apache/revert-52-release_news_1.1.0

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git commit 3b4ab4823f4689ec5e31124e806def9e999eb4d3 Merge: ca32008 4cdb455 Author: Marco de Abreu

[GitHub] marcoabreu closed pull request #53: Revert "fixed the release number to 1.1.0"

marcoabreu closed pull request #53: Revert "fixed the release number to 1.1.0" URL: https://github.com/apache/incubator-mxnet-site/pull/53 This is an automated message from the Apache Git Service. To respond to the message,

[incubator-mxnet-site] branch revert-52-release_news_1.1.0 created (now 4cdb455)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch revert-52-release_news_1.1.0 in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git. at 4cdb455 Revert "fixed the release number to 1.1.0" This branch includes

  1   2   >