[GitHub] shadogray commented on a change in pull request #14078: [MXNET-14075] repair example, enable fault tolerant build of R reference manual
shadogray commented on a change in pull request #14078: [MXNET-14075] repair example, enable fault tolerant build of R reference manual URL: https://github.com/apache/incubator-mxnet/pull/14078#discussion_r257450070 ## File path: Makefile ## @@ -619,8 +620,19 @@ rpkg: echo "import(Rcpp)" >> R-package/NAMESPACE R CMD INSTALL R-package Rscript -e "require(mxnet); mxnet:::mxnet.export('R-package'); warnings()" - rm R-package/NAMESPACE + $(RM) R-package/NAMESPACE Review comment: for this Makefile a global remove command is defined, and it is reasonable to make consistent use of it This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.
This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 922889b Bump the publish timestamp. 922889b is described below commit 922889b85e113b528452f98adf03fa85d25b66d2 Author: mxnet-ci AuthorDate: Sat Feb 16 07:07:23 2019 + Bump the publish timestamp. --- date.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/date.txt b/date.txt new file mode 100644 index 000..037a86e --- /dev/null +++ b/date.txt @@ -0,0 +1 @@ +Sat Feb 16 07:07:23 UTC 2019
[GitHub] eric-haibin-lin merged pull request #14082: Fix nd.pick large array issue
eric-haibin-lin merged pull request #14082: Fix nd.pick large array issue URL: https://github.com/apache/incubator-mxnet/pull/14082 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: Fix nd.pick large array issue (#14082)
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 ed0b791 Fix nd.pick large array issue (#14082) ed0b791 is described below commit ed0b791ed5c2c2dde8b7136302abbbd3a85e89da Author: Chaitanya Prakash Bapat AuthorDate: Sat Feb 16 00:01:30 2019 -0500 Fix nd.pick large array issue (#14082) * large op support * replaced size_t with index_t for M, added test case * changed shape --- src/operator/tensor/broadcast_reduce_op.h | 12 ++-- tests/nightly/test_large_array.py | 5 + 2 files changed, 11 insertions(+), 6 deletions(-) diff --git a/src/operator/tensor/broadcast_reduce_op.h b/src/operator/tensor/broadcast_reduce_op.h index 1edcb5a..6aeeadf 100644 --- a/src/operator/tensor/broadcast_reduce_op.h +++ b/src/operator/tensor/broadcast_reduce_op.h @@ -1172,12 +1172,12 @@ void L2NormComputeEx(const nnvm::NodeAttrs& attrs, template struct pick { template - MSHADOW_XINLINE static void Map(int i, DType* out, const DType* a, - const IType *idx, int M, int stride, + MSHADOW_XINLINE static void Map(index_t i, DType* out, const DType* a, + const IType *idx, index_t M, int stride, mshadow::Shape bshape, mshadow::Shape sshape) { using namespace broadcast; -int j = static_cast(idx[i]); +index_t j = static_cast(idx[i]); if (clip) { if (j <= 0) j = 0; else if (j >= M) j = M - 1; @@ -1194,12 +1194,12 @@ struct pick { template struct pick_grad { template - MSHADOW_XINLINE static void Map(int i, DType* igrad, const DType* ograd, - const IType *idx, int M, int stride, + MSHADOW_XINLINE static void Map(index_t i, DType* igrad, const DType* ograd, + const IType *idx, index_t M, int stride, mshadow::Shape bshape, mshadow::Shape sshape) { using namespace broadcast; -int j = static_cast(idx[i]); +index_t j = static_cast(idx[i]); if (clip) { if (j <= 0) j = 0; else if (j >= M) j = M - 1; diff --git a/tests/nightly/test_large_array.py b/tests/nightly/test_large_array.py index 696fdb1..0249f44 100644 --- a/tests/nightly/test_large_array.py +++ b/tests/nightly/test_large_array.py @@ -145,6 +145,11 @@ def test_where(): res = nd.sparse.where(csr_cond, a, b) assert np.sum(res[0].asnumpy() == 1) == b.shape[1] +def test_pick(): +a = mx.nd.ones(shape=(256*35, 1024*1024)) +b = mx.nd.ones(shape=(256*35,)) +res = mx.nd.pick(a,b) +assert res.shape == b.shape if __name__ == '__main__': import nose
[GitHub] eric-haibin-lin closed issue #9314: nd.pick core dumps on large array
eric-haibin-lin closed issue #9314: nd.pick core dumps on large array URL: https://github.com/apache/incubator-mxnet/issues/9314 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ufoym edited a comment on issue #12792: Fix broken amalgamation
ufoym edited a comment on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-464287027 @ankkhedia @vandanavk @zheng-da I merely changed the code (just simply resolved the merge conflicts). All checks have passed for several times. Did I miss something? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] eric-haibin-lin commented on a change in pull request #14082: Fix nd.pick large array issue
eric-haibin-lin commented on a change in pull request #14082: Fix nd.pick large array issue URL: https://github.com/apache/incubator-mxnet/pull/14082#discussion_r257346109 ## File path: tests/nightly/test_large_array.py ## @@ -145,6 +145,11 @@ def test_where(): res = nd.sparse.where(csr_cond, a, b) assert np.sum(res[0].asnumpy() == 1) == b.shape[1] +def test_pick(): Review comment: The new test would fail on master without the fix. Looks good now This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: Fix broken amalgamation (#12792)
This is an automated email from the ASF dual-hosted git repository. zhasheng 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 29fa446 Fix broken amalgamation (#12792) 29fa446 is described below commit 29fa446915a0dedfedf03226a5795f9a903d2b4f Author: Ming AuthorDate: Sat Feb 16 12:34:26 2019 +0800 Fix broken amalgamation (#12792) * Fix broken amalgamation * Fixes #8850 * Update CONTRIBUTORS.md * Update CONTRIBUTORS.md --- CONTRIBUTORS.md| 2 ++ amalgamation/amalgamation.py | 7 ++- amalgamation/dmlc-minimum0.cc | 1 + amalgamation/mxnet_predict0.cc | 7 +++ 4 files changed, 16 insertions(+), 1 deletion(-) diff --git a/CONTRIBUTORS.md b/CONTRIBUTORS.md index ffe0da6..caec6cf 100644 --- a/CONTRIBUTORS.md +++ b/CONTRIBUTORS.md @@ -213,6 +213,8 @@ List of Contributors * [Piyush Ghai](https://github.com/piyushghai) * [Zach Boldyga](https://github.com/zboldyga) +* [Ming Yang](http://ufoym.com) + Label Bot - * [mxnet-label-bot](https://github.com/mxnet-label-bot) diff --git a/amalgamation/amalgamation.py b/amalgamation/amalgamation.py index a3c28f7..0a4be02 100644 --- a/amalgamation/amalgamation.py +++ b/amalgamation/amalgamation.py @@ -143,7 +143,12 @@ def expand(x, pending, stage): continue path = m.groups()[0] h = path.strip('./') if "../3rdparty/" not in path else path -source = find_source(h, x, stage) +if h.endswith('complex.h') and x.endswith('openblas_config.h'): +source = '' +elif h.startswith('ps/'): +source = '../3rdparty/ps-lite/include/' + h +else: +source = find_source(h, x, stage) if not source: if (h not in blacklist and h not in sysheaders and diff --git a/amalgamation/dmlc-minimum0.cc b/amalgamation/dmlc-minimum0.cc index 87e08d3..2fe629b 100644 --- a/amalgamation/dmlc-minimum0.cc +++ b/amalgamation/dmlc-minimum0.cc @@ -29,6 +29,7 @@ #include "../3rdparty/dmlc-core/src/io/local_filesys.cc" #include "../3rdparty/dmlc-core/src/data.cc" #include "../3rdparty/dmlc-core/src/io.cc" +#include "../3rdparty/dmlc-core/src/io/filesys.cc" #include "../3rdparty/dmlc-core/src/recordio.cc" diff --git a/amalgamation/mxnet_predict0.cc b/amalgamation/mxnet_predict0.cc index c4653d3..a18e28f 100644 --- a/amalgamation/mxnet_predict0.cc +++ b/amalgamation/mxnet_predict0.cc @@ -43,25 +43,31 @@ #define DISABLE_OPENMP 1 #define DMLC_LOG_STACK_TRACE 0 +#include "src/common/utils.cc" #include "src/ndarray/ndarray_function.cc" #include "src/ndarray/ndarray.cc" #include "src/imperative/imperative.cc" +#include "src/imperative/imperative_utils.cc" #include "src/imperative/cached_op.cc" #include "src/engine/engine.cc" #include "src/engine/naive_engine.cc" +#include "src/engine/openmp.cc" #include "src/profiler/profiler.cc" +#include "src/profiler/aggregate_stats.cc" #include "src/executor/graph_executor.cc" #include "src/executor/attach_op_execs_pass.cc" #include "src/executor/attach_op_resource_pass.cc" #include "src/executor/inplace_addto_detect_pass.cc" +#include "src/executor/infer_graph_attr_pass.cc" #include "src/nnvm/legacy_json_util.cc" #include "src/nnvm/legacy_op_util.cc" +#include "src/nnvm/graph_editor.cc" #include "src/operator/operator.cc" #include "src/operator/operator_util.cc" @@ -93,4 +99,5 @@ #include "src/c_api/c_api_symbolic.cc" #include "src/c_api/c_api_ndarray.cc" #include "src/c_api/c_api_error.cc" +#include "src/c_api/c_api_profile.cc"
[GitHub] szha closed issue #8850: iOS compilation with amalgamate bash, undefined symbols for architecture arm64
szha closed issue #8850: iOS compilation with amalgamate bash, undefined symbols for architecture arm64 URL: https://github.com/apache/incubator-mxnet/issues/8850 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha merged pull request #12792: Fix broken amalgamation
szha merged pull request #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha commented on issue #12792: Fix broken amalgamation
szha commented on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-46429 Verification happened in http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Fmiscellaneous/detail/PR-12792/8/pipeline/ LGTM This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ufoym edited a comment on issue #12792: Fix broken amalgamation
ufoym edited a comment on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-464287027 @ankkhedia @vandanavk @zheng-da I merely changed the code (Every time I've been asked for fixing the existing issues, I just simply resolved the merge conflicts). All checks have passed for several times. Did I miss something? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ufoym edited a comment on issue #12792: Fix broken amalgamation
ufoym edited a comment on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-464287027 @ankkhedia @vandanavk @zheng-da I merely changed the code (Every time I've been asked for fixing the existing issues, I just resolved the merge conflicts). All checks have passed for several times. Did I miss something? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ufoym commented on issue #12792: Fix broken amalgamation
ufoym commented on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-464287027 @ankkhedia @vandanavk @zheng-da I merely changed the code (just resolved the merge conflicts), all checks have passed for several times. Did I miss something? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: [v1.4.x] Update MKL-DNN to fix the OSX build issue (#14141) (#14182)
This is an automated email from the ASF dual-hosted git repository. zhasheng 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 d2cc396 [v1.4.x] Update MKL-DNN to fix the OSX build issue (#14141) (#14182) d2cc396 is described below commit d2cc39639d8f4da9c2e7a7d7cb0dd0b2c43599aa Author: Tao Lv AuthorDate: Sat Feb 16 12:23:30 2019 +0800 [v1.4.x] Update MKL-DNN to fix the OSX build issue (#14141) (#14182) * update mkldnn to 0.17.x * update mkldnn to 0.17.4 * empty commit --- 3rdparty/mkldnn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/3rdparty/mkldnn b/3rdparty/mkldnn index a7c5f53..722901c 16 --- a/3rdparty/mkldnn +++ b/3rdparty/mkldnn @@ -1 +1 @@ -Subproject commit a7c5f53832acabade6e5086e72c960adedb3c38a +Subproject commit 722901c9aaefa579698df778d061d4848ab8c3e3
[GitHub] szha merged pull request #14182: Update MKL-DNN to fix the OSX build issue on master
szha merged pull request #14182: Update MKL-DNN to fix the OSX build issue on master URL: https://github.com/apache/incubator-mxnet/pull/14182 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ChaiBapchya commented on issue #14157: Inconsistent handling for nan
ChaiBapchya commented on issue #14157: Inconsistent handling for nan URL: https://github.com/apache/incubator-mxnet/issues/14157#issuecomment-464277857 Similarly found it for `mx.nd.minimum` ``` >>> mx.nd.minimum(b,a) [nan] >>> mx.nd.minimum(a,b) [3.] ``` This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] DickJC123 edited a comment on issue #14006: Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2.
DickJC123 edited a comment on issue #14006: Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2. URL: https://github.com/apache/incubator-mxnet/pull/14006#issuecomment-464276100 I've rerun some perf analysis of this PR, which I'll remind everyone changes nothing in the default. However, when I set MXNET_GPU_WORKER_NSTREAMS=2, I see higher performance for all batchsizes. The perf gains I measured on a run across 8 Volta GPUs of Resnet50 v1b (also with horovod and DALI in NVIDIA's MXNet container) were: ``` batchsize 32: 6.0% speedup batchsize 64: 0.8% speedup batchsize 128: 1.6% speedup batchsize 256: 0.4% speedup ``` The primary application area of the PR is one of scale-out training across multiple nodes, where a too-large global batchsize can impact final accuracy (thus driving per-GPU batchsize down). The RN50 global memory increase was from 1.4% (bs 32) to 2.6% (bs 256). This work is no longer "in progress." Requesting final review thanks. @szha @marcoabreu This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] DickJC123 commented on issue #14006: Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2.
DickJC123 commented on issue #14006: Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2. URL: https://github.com/apache/incubator-mxnet/pull/14006#issuecomment-464276100 I've rerun some perf analysis of this PR, which I'll remind everyone changes nothing in the default. However, when I set MXNET_GPU_WORKER_NSTREAMS=2, I see higher performance for all batchsizes. The perf gains I measured on a run across 8 Volta GPUs of Resnet50 v1b (also with horovod and DALI in NVIDIA's MXNet container) were: ``` batchsize 32: 6.0% speedup batchsize 64: 0.8% speedup batchsize 128: 1.6% speedup batchsize 256: 0.4% speedup ``` The primary application area of the PR is one of scale-out training across multiple nodes, where a too-large global batchsize can impact final accuracy (thus driving per-GPU batchsize down). The RN50 global memory increase was from 1.4% (bs 32) to 2.6% (bs 256). This work is no longer "in progress." Requesting final review thanks. @szha @marcoabreu This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: [MXNET-766] add dynamic_unroll RNN for HybridBlock (#11948)
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 4b527b6 [MXNET-766] add dynamic_unroll RNN for HybridBlock (#11948) 4b527b6 is described below commit 4b527b681f01bff02a770c1b49a3b5c9278cf4a0 Author: Da Zheng AuthorDate: Fri Feb 15 18:23:01 2019 -0800 [MXNET-766] add dynamic_unroll RNN for HybridBlock (#11948) * add contrib unroll. * reenable some tests. * fix a bug. * fix lint. * fix a bug. * support diff layouts. * update doc. * use a diff default layout. * remove _contrib_format_sequence. * fix lint. * rename. --- python/mxnet/gluon/contrib/rnn/rnn_cell.py | 115 tests/python/unittest/test_gluon_contrib.py | 95 ++- 2 files changed, 209 insertions(+), 1 deletion(-) diff --git a/python/mxnet/gluon/contrib/rnn/rnn_cell.py b/python/mxnet/gluon/contrib/rnn/rnn_cell.py index 0cbc9ea..3bd8e78 100644 --- a/python/mxnet/gluon/contrib/rnn/rnn_cell.py +++ b/python/mxnet/gluon/contrib/rnn/rnn_cell.py @@ -22,6 +22,7 @@ __all__ = ['VariationalDropoutCell', 'LSTMPCell'] from ...rnn import BidirectionalCell, SequentialRNNCell, ModifierCell, HybridRecurrentCell from ...rnn.rnn_cell import _format_sequence, _get_begin_state, _mask_sequence_variable_length from ... import tensor_types +from base import _as_list class VariationalDropoutCell(ModifierCell): """ @@ -320,3 +321,117 @@ class LSTMPCell(HybridRecurrentCell): return next_r, [next_r, next_c] # pylint: enable= arguments-differ + + +def dynamic_unroll(cell, inputs, begin_state, drop_inputs=0, drop_outputs=0, + layout='TNC', valid_length=None): +"""Unrolls an RNN cell across time steps. + +Currently, 'TNC' is a preferred layout. unroll on the input of this layout +runs much faster. + +Parameters +-- +cell : an object whose base class is RNNCell. +The RNN cell to run on the input sequence. +inputs : Symbol +It should have shape (batch_size, length, ...) if `layout` is 'NTC', +or (length, batch_size, ...) if `layout` is 'TNC'. +begin_state : nested list of Symbol +The initial states of the RNN sequence. +drop_inputs : float, default 0. +The dropout rate for inputs. Won't apply dropout if it equals 0. +drop_outputs : float, default 0. +The dropout rate for outputs. Won't apply dropout if it equals 0. +layout : str, optional +`layout` of input symbol. Only used if inputs +is a single Symbol. +valid_length : Symbol, NDArray or None +`valid_length` specifies the length of the sequences in the batch without padding. +This option is especially useful for building sequence-to-sequence models where +the input and output sequences would potentially be padded. +If `valid_length` is None, all sequences are assumed to have the same length. +If `valid_length` is a Symbol or NDArray, it should have shape (batch_size,). +The ith element will be the length of the ith sequence in the batch. +The last valid state will be return and the padded outputs will be masked with 0. +Note that `valid_length` must be smaller or equal to `length`. + +Returns +--- +outputs : Symbol +the output of the RNN from this unrolling. + +states : list of Symbol +The new state of this RNN after this unrolling. +The type of this symbol is same as the output of `begin_state`. + +Examples + +>>> seq_len = 3 +>>> batch_size = 2 +>>> input_size = 5 +>>> cell = mx.gluon.rnn.LSTMCell(input_size, prefix='rnn_') +>>> cell.initialize(ctx=mx.cpu()) +>>> rnn_data = mx.nd.normal(loc=0, scale=1, shape=(seq_len, batch_size, input_size)) +>>> state_shape = (batch_size, input_size) +>>> states = [mx.nd.normal(loc=0, scale=1, shape=state_shape) for i in range(2)] +>>> valid_length = mx.nd.array([2, 3]) +>>> output, states = mx.gluon.contrib.rnn.rnn_cell.dynamic_unroll(cell, rnn_data, states, + valid_length=valid_length, + layout='TNC') +>>> print(output) +[[[ 0.00767238 0.00023103 0.03973929 -0.00925503 -0.05660512] + [ 0.00881535 0.05428379 -0.02493718 -0.01834097 0.02189514]] + [[-0.00676967 0.01447039 0.01287002 -0.00574152 -0.05734247] + [ 0.01568508 0.02650866 -0.04270559 -0.04328435 0.00904011]] + [[ 0. 0. 0. 0. 0.] + [ 0.01055336 0.02734251 -0.03153727 -0.03742751 -0.01378113]]] + +""" + +
[GitHub] eric-haibin-lin merged pull request #11948: [MXNET-766] add unroll RNN for HybridBlock
eric-haibin-lin merged pull request #11948: [MXNET-766] add unroll RNN for HybridBlock URL: https://github.com/apache/incubator-mxnet/pull/11948 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14182: Update MKL-DNN to fix the OSX build issue on master
ankkhedia commented on issue #14182: Update MKL-DNN to fix the OSX build issue on master URL: https://github.com/apache/incubator-mxnet/pull/14182#issuecomment-464268878 @TaoLv Thanks for your contribution! @mxnet-label-bot add [pr-awaiting-review, MKLDNN, build] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] tag 1.4.0.rc3 created (now a03d59e)
This is an automated email from the ASF dual-hosted git repository. lanking pushed a change to tag 1.4.0.rc3 in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. at a03d59e (commit) No new revisions were added by this update.
[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.
This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 1afe4a1 Bump the publish timestamp. 1afe4a1 is described below commit 1afe4a1774cf5728e65eeae408cd52492251bf43 Author: mxnet-ci AuthorDate: Sat Feb 16 01:06:17 2019 + Bump the publish timestamp. --- date.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/date.txt b/date.txt new file mode 100644 index 000..24b0909 --- /dev/null +++ b/date.txt @@ -0,0 +1 @@ +Sat Feb 16 01:06:17 UTC 2019
[GitHub] TaoLv commented on issue #14182: Update MKL-DNN to fix the OSX build issue on master
TaoLv commented on issue #14182: Update MKL-DNN to fix the OSX build issue on master URL: https://github.com/apache/incubator-mxnet/pull/14182#issuecomment-464266450 @szha @lanking520 @pengzhao-intel Picked #14141 from v1.4.x to master. MKL-DNN 0.18 update will be done via another PR after it passed all validation. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] TaoLv opened a new pull request #14182: Update MKL-DNN to fix the OSX build issue on master
TaoLv opened a new pull request #14182: Update MKL-DNN to fix the OSX build issue on master URL: https://github.com/apache/incubator-mxnet/pull/14182 Cherry picked https://github.com/apache/incubator-mxnet/pull/14141 from v1.4.x to master. ## Description ## (Brief description on what this PR is about) ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [ ] The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant [JIRA issue](https://issues.apache.org/jira/projects/MXNET/issues) created (except PRs with tiny changes) - [ ] Changes are complete (i.e. I finished coding on this PR) - [ ] All changes have test coverage: - Unit tests are added for small changes to verify correctness (e.g. adding a new operator) - Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore) - Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL) - [ ] Code is well-documented: - For user-facing API changes, API doc string has been updated. - For new C++ functions in header files, their functionalities and arguments are documented. - For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable - Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html - [ ] To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change ### Changes ### - [ ] Feature1, tests, (and when applicable, API doc) - [ ] Feature2, tests, (and when applicable, API doc) ## Comments ## - If this change is a backward incompatible change, why must this change be made. - Interesting edge cases to note here This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ThomasDelteil commented on issue #8861: The first GPU cost more memory than the others in version 0.12.1
ThomasDelteil commented on issue #8861: The first GPU cost more memory than the others in version 0.12.1 URL: https://github.com/apache/incubator-mxnet/issues/8861#issuecomment-464264786 This bug with extra memory allocation on GPU 0 has been resolved now This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ThomasDelteil closed issue #8861: The first GPU cost more memory than the others in version 0.12.1
ThomasDelteil closed issue #8861: The first GPU cost more memory than the others in version 0.12.1 URL: https://github.com/apache/incubator-mxnet/issues/8861 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ThomasDelteil commented on issue #6418: Performance of deep speech is bad
ThomasDelteil commented on issue #6418: Performance of deep speech is bad URL: https://github.com/apache/incubator-mxnet/issues/6418#issuecomment-464263603 @megaSpoon have a look at mxboard https://github.com/awslabs/mxboard This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] NRauschmayr commented on a change in pull request #14179: Fix READMEs for examples
NRauschmayr commented on a change in pull request #14179: Fix READMEs for examples URL: https://github.com/apache/incubator-mxnet/pull/14179#discussion_r257432986 ## File path: example/gluon/audio/README.md ## @@ -0,0 +1,100 @@ +# Urban Sounds Classification in MXNet Gluon + +This example provides an end-to-end pipeline for a common datahack competition - Urban Sounds Classification Example. +Below is the link to the competition: +https://datahack.analyticsvidhya.com/contest/practice-problem-urban-sound-classification/ + +After logging in, the data set can be downloaded. +The details of the dataset and the link to download it are given below: + + +## Urban Sounds Dataset: +### Description + The dataset contains 8732 wav files which are audio samples(<= 4s)) of street sounds like engine_idling, car_horn, children_playing, dog_barking and so on. + The task is to classify these audio samples into one of the following 10 labels: + ``` + siren, + street_music, + drilling, + dog_bark, + children_playing, + gun_shot, + engine_idling, + air_conditioner, + jackhammer, + car_horn + ``` + +To be able to run this example: + +1. `pip install -r requirements.txt` + +If you are in the directory where the requirements.txt file lies, +this step installs the required libraries to run the example. +The main dependency that is required is: Librosa. +The version used to test the example is: `0.6.2` +For more details, refer here: +https://librosa.github.io/librosa/install.html + +2. Download the dataset(train.zip, test.zip) required for this example from the location: +https://drive.google.com/drive/folders/0By0bAi7hOBAFUHVXd1JCN3MwTEU + Review comment: I agree with you. We need to figure out what the license of the dataset is. I assume the original files are from here https://datahack.analyticsvidhya.com/contest/practice-problem-urban-sound-classification/#data_dictionary And the Google drive link belongs to someone from analyticsvidhya, so maybe this link is the official source? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] drivanov commented on issue #14154: Improving the run time of the tests which use assert_almost_equal OR check_numeric_gradient functions
drivanov commented on issue #14154: Improving the run time of the tests which use assert_almost_equal OR check_numeric_gradient functions URL: https://github.com/apache/incubator-mxnet/pull/14154#issuecomment-464255784 I think that fixed previously found problems. The problem in `test_operator_gpu.py:test_laop` was NOT reproducible due to nondeterministic behavior we see in that test. One of the cause of the non-determinism was eliminated. I am not sure that it was the last one, but at least, this test should become more predictable. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] lanking520 commented on issue #14104: Error in mx.nd.internal.empty.array(shape, ctx) : [14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enab
lanking520 commented on issue #14104: Error in mx.nd.internal.empty.array(shape, ctx) :[14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enable GPU usage in R URL: https://github.com/apache/incubator-mxnet/issues/14104#issuecomment-464249398 @Ishitori Thanks for your input. Should we make it cpu as a default? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] Ishitori commented on issue #14089: Load dataset directly to vRAM
Ishitori commented on issue #14089: Load dataset directly to vRAM URL: https://github.com/apache/incubator-mxnet/issues/14089#issuecomment-464247795 I don't think it is possible at the moment. The data always loaded first in RAM, and then you need to manually move it to the proper GPU memory with either `split_and_load()` or `as_in_context()`. If you can load the batch of 16 items in RAM, but would like to load more into GPUs before doing a forward pass, then I would: 1. Write a custom loop which would go through ImageRecordIter as many times as you need (let's say you want each GPU to work with 32 items: for 8 GPUs you need to do 8 iterations) 2. On each iteration do `split_and_load()` and concatenate the result for each GPU with the data that is already there if any. Something like this (I haven't run the code myself, so it might need adjustments): ``` import mxnet as mx from mxnet import gluon num_gpus = 8 max_items_for_ram = 16 max_items_per_gpu = 32 ctx = [mx.gpu(i) for i in range(num_gpus)] data_train = mx.io.ImageRecordIter( path_imgrec='dataset_train.rec', data_shape=(3, 256, 256), batch_size=max_items_for_ram, label_width = 22 ) gpu_items = [] iters_to_load_max_items_per_gpu = max_items_per_gpu / (max_items_for_ram / num_gpus) for iteration in range(int(iters_to_load_max_items_per_gpu)): for batch in data_train: items = gluon.utils.split_and_load(batch, ctx) if not gpu_items: gpu_items = items else: for i in range(len(ctx)): gpu_items[i] = mx.nd.concat(gpu_items[i], items[i]) ``` This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] andrewfayres commented on issue #14181: Scala Infer ObjectDector crashes on models which don't return
andrewfayres commented on issue #14181: Scala Infer ObjectDector crashes on models which don't return URL: https://github.com/apache/incubator-mxnet/issues/14181#issuecomment-464243736 @mxnet-label-bot add [Bug, Scala] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] mxnet-label-bot commented on issue #14181: Scala Infer ObjectDector crashes on models which don't return
mxnet-label-bot commented on issue #14181: Scala Infer ObjectDector crashes on models which don't return URL: https://github.com/apache/incubator-mxnet/issues/14181#issuecomment-464243381 Hey, this is the MXNet Label Bot. Thank you for submitting the issue! I will try and suggest some labels so that the appropriate MXNet community members can help resolve it. Here are my recommended labels: Scala, Bug This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] andrewfayres opened a new issue #14181: Scala Infer ObjectDector crashes on models which don't return
andrewfayres opened a new issue #14181: Scala Infer ObjectDector crashes on models which don't return URL: https://github.com/apache/incubator-mxnet/issues/14181 ## Description The Scala Infer API's ObjectDetector will crash when provided with a model which doesn't return coordinates for bounding boxes. We should handle this situation more gracefully. Either return an error message stating that the model doesn't have the required outputs or return the results without the bounding boxes. ## Error Message: ``` $ ./bin/run_det.sh SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder". SLF4J: Defaulting to no-operation (NOP) logger implementation SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details. [22:18:13] src/nnvm/legacy_json_util.cc:209: Loading symbol saved by previous version v1.3.1. Attempting to upgrade... [22:18:13] src/nnvm/legacy_json_util.cc:217: Symbol successfully upgraded! [22:18:17] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (setting env variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable) Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 1 at org.apache.mxnet.infer.ObjectDetector$$anonfun$2.apply(ObjectDetector.scala:132) at org.apache.mxnet.infer.ObjectDetector$$anonfun$2.apply(ObjectDetector.scala:132) at scala.math.Ordering$$anon$5.compare(Ordering.scala:122) at java.util.TimSort.countRunAndMakeAscending(TimSort.java:355) at java.util.TimSort.sort(TimSort.java:234) at java.util.Arrays.sort(Arrays.java:1438) at scala.collection.SeqLike$class.sorted(SeqLike.scala:648) at scala.collection.mutable.ArrayOps$ofRef.sorted(ArrayOps.scala:186) at scala.collection.SeqLike$class.sortBy(SeqLike.scala:623) at scala.collection.mutable.ArrayOps$ofRef.sortBy(ArrayOps.scala:186) at org.apache.mxnet.infer.ObjectDetector.sortAndReformat(ObjectDetector.scala:132) at org.apache.mxnet.infer.ObjectDetector$$anonfun$objectDetectWithNDArray$1.apply$mcVI$sp(ObjectDetector.scala:106) at org.apache.mxnet.infer.ObjectDetector$$anonfun$objectDetectWithNDArray$1.apply(ObjectDetector.scala:104) at org.apache.mxnet.infer.ObjectDetector$$anonfun$objectDetectWithNDArray$1.apply(ObjectDetector.scala:104) at scala.collection.parallel.mutable.ParArray$ParArrayIterator.foreach_quick(ParArray.scala:143) at scala.collection.parallel.mutable.ParArray$ParArrayIterator.foreach(ParArray.scala:136) at scala.collection.parallel.ParIterableLike$Foreach.leaf(ParIterableLike.scala:972) at scala.collection.parallel.Task$$anonfun$tryLeaf$1.apply$mcV$sp(Tasks.scala:49) at scala.collection.parallel.Task$$anonfun$tryLeaf$1.apply(Tasks.scala:48) at scala.collection.parallel.Task$$anonfun$tryLeaf$1.apply(Tasks.scala:48) at scala.collection.parallel.Task$class.tryLeaf(Tasks.scala:51) at scala.collection.parallel.ParIterableLike$Foreach.tryLeaf(ParIterableLike.scala:969) at scala.collection.parallel.AdaptiveWorkStealingTasks$WrappedTask$class.compute(Tasks.scala:152) at scala.collection.parallel.AdaptiveWorkStealingForkJoinTasks$WrappedTask.compute(Tasks.scala:443) at scala.concurrent.forkjoin.RecursiveAction.exec(RecursiveAction.java:160) at scala.concurrent.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260) at scala.concurrent.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339) at scala.concurrent.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979) at scala.concurrent.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107) ``` ## Minimum reproducible example ``` public static void main(String[] args) { String modelPathPrefix = "./resnet-152"; String inputImagePath = "./cat.jpg"; List context = new ArrayList(); context.add(Context.gpu()); // Prepare the model List inputDesc = new ArrayList<>(); Shape inputShape = new Shape(new int[]{1, 3, 224, 224}); inputDesc.add(new DataDesc("data", inputShape, DType.Float32(), "NCHW")); ObjectDetector objDet = new ObjectDetector(modelPathPrefix, inputDesc, context, 0); // predict int iter = 100; ArrayList data = new ArrayList(iter); for(int i=0; i> result = objDet.imageObjectDetect(img,3); long stop = System.nanoTime(); double elapsed = (stop-start)/100.0; data.add(elapsed);
[GitHub] vandanavk commented on a change in pull request #13735: update wavenet codes
vandanavk commented on a change in pull request #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#discussion_r257421984 ## File path: example/gluon/wavenet/main.py ## @@ -0,0 +1,60 @@ +""" +Descrition : main module to run code +""" +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +import argparse +from trainer import Train + +def main(): +""" +Description : run wavenet code using argument info +- batch_size : Define batch size (default=64) +- epochs : Define the total number of epochs (default=1000) +- mu : Define mu value for mu-law algorithm (default=128) +- n_residue : Define number of residue (default=24) +- n_skip : Define number of skip (default=128) +- dilation_depth : Define dilation depth (default=10) +- use_gpu : use gpu for trainin (default=True) +- generation : whether or not to generate a wave file for model (default=True) +- load_file : file name in loading wave file Review comment: could you add the default value in this comment as well This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] vandanavk commented on a change in pull request #13735: update wavenet codes
vandanavk commented on a change in pull request #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#discussion_r257421943 ## File path: example/gluon/wavenet/models.py ## @@ -0,0 +1,119 @@ + +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +""" +Module: WaveNet network block +""" +from mxnet import nd +from mxnet.gluon import nn +# pylint: disable=invalid-name, too-many-arguments, arguments-differ, attribute-defined-outside-init, too-many-instance-attributes, invalid-sequence-index, no-self-use + +class One_Hot(nn.Block): +""" +Description : generate one hot result +""" +def __init__(self, depth): +super(One_Hot, self).__init__() +self.depth = depth + +def forward(self, X_in): +with X_in.context: +X_in = X_in Review comment: you mean this line will be deleted? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] vandanavk commented on a change in pull request #13735: update wavenet codes
vandanavk commented on a change in pull request #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#discussion_r257421402 ## File path: example/gluon/wavenet/models.py ## @@ -0,0 +1,119 @@ + +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +""" +Module: WaveNet network block +""" +from mxnet import nd +from mxnet.gluon import nn +# pylint: disable=invalid-name, too-many-arguments, arguments-differ, attribute-defined-outside-init, too-many-instance-attributes, invalid-sequence-index, no-self-use + +class One_Hot(nn.Block): +""" +Description : generate one hot result +""" +def __init__(self, depth): +super(One_Hot, self).__init__() +self.depth = depth + +def forward(self, X_in): +with X_in.context: +X_in = X_in Review comment: you mean this line will be deleted? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] vandanavk commented on a change in pull request #13735: update wavenet codes
vandanavk commented on a change in pull request #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#discussion_r257421840 ## File path: example/gluon/wavenet/data_loader.py ## @@ -0,0 +1,61 @@ +""" +Description : Set Dataset module for Wavenet +""" +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License +import os +import numpy as np +from mxnet import nd +from scipy.io import wavfile +from utils import encode_mu_law +# pylint: disable=invalid-name, too-many-arguments + +def load_wav(file_nm): +""" +Description : load wav file +""" +fs, data = wavfile.read(os.getcwd()+'/dataset/'+file_nm) +return fs, data + +def data_generation(data, framerate, seq_size, mu, ctx, gen_mode=None): +""" +Description : data generation to loading data +""" +if gen_mode == 'sin': +t = np.linspace(0, 5, framerate*5) +data = np.sin(2*np.pi*220*t) + np.sin(2*np.pi*224*t) +div = max(data.max(), abs(data.min())) +data = data/div Review comment: will this line be modified? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] vandanavk commented on a change in pull request #13735: update wavenet codes
vandanavk commented on a change in pull request #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#discussion_r257421254 ## File path: example/gluon/wavenet/main.py ## @@ -0,0 +1,60 @@ +""" +Descrition : main module to run code +""" +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +import argparse +from trainer import Train + +def main(): +""" +Description : run wavenet code using argument info +- batch_size : Define batch size (default=64) +- epochs : Define the total number of epochs (default=1000) +- mu : Define mu value for mu-law algorithm (default=128) +- n_residue : Define number of residue (default=24) +- n_skip : Define number of skip (default=128) +- dilation_depth : Define dilation depth (default=10) +- use_gpu : use gpu for trainin (default=True) +- generation : whether or not to generate a wave file for model (default=True) +- load_file : file name in loading wave file Review comment: please mention the default value here as well This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] Ishitori commented on issue #14104: Error in mx.nd.internal.empty.array(shape, ctx) : [14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enable
Ishitori commented on issue #14104: Error in mx.nd.internal.empty.array(shape, ctx) :[14:58:38] C:\Jenkins\workspace\mxnet\mxnet\src\storage\storage.cc:137: Compile with USE_CUDA=1 to enable GPU usage in R URL: https://github.com/apache/incubator-mxnet/issues/14104#issuecomment-464241939 Yes, @lanking520, it is using GPU by default. See it here: https://github.com/apache/incubator-mxnet/blob/master/example/captcha/mxnet_captcha.R This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] Ishitori commented on issue #14110: how to get picture for capture recognition task
Ishitori commented on issue #14110: how to get picture for capture recognition task URL: https://github.com/apache/incubator-mxnet/issues/14110#issuecomment-464241496 `Rec` file is a file that contains pictures. It is a special RecordIO format which allows to store many small binary files in one bigger file. It is done to increase speed of reading data from the disk. Use [ImageRecordDataset](https://mxnet.incubator.apache.org/versions/master/api/python/gluon/data.html#mxnet.gluon.data.vision.datasets.ImageRecordDataset) to open and use it. If you want to know more about RecordIO format, take a look here:https://mxnet.incubator.apache.org/versions/master/architecture/note_data_loading.html This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] vandanavk commented on a change in pull request #14042: [MXNET-891] Support tuple of scales in upsample operator
vandanavk commented on a change in pull request #14042: [MXNET-891] Support tuple of scales in upsample operator URL: https://github.com/apache/incubator-mxnet/pull/14042#discussion_r257419777 ## File path: src/operator/nn/upsampling-inl.h ## @@ -82,6 +84,57 @@ struct UpSamplingParam : public dmlc::Parameter { } }; // struct UpSamplingParam +template +void SpatialUpSamplingNearestUpdateOutput(mshadow::Stream *s, + const std::vector _data, + std::vector *out_data) { + Tensor itensor = in_data[0].get(s); + Tensor otensor = (*out_data)[0].get(s); + + int outputHeight = otensor.size(2); + int outputWidth = otensor.size(3); + int inputHeight = itensor.size(2); + int inputWidth = itensor.size(3); + + int dW = outputWidth / inputWidth; + int dH = outputHeight / inputHeight; + int idim = itensor.shape_.kDimension; + + // dims + int osz0 = otensor.size(0); + int osz1 = otensor.size(1); + int osz2 = otensor.size(2); + int osz3 = 1; + if (idim > 3) { +osz3 = otensor.size(3); + } + + // perform the upsampling + int i0, i1, i2, i3; Review comment: Sure, I'll make this change This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] Ishitori commented on issue #14164: How can i create .rec file for R
Ishitori commented on issue #14164: How can i create .rec file for R URL: https://github.com/apache/incubator-mxnet/issues/14164#issuecomment-464239711 Unfortunately, im2rec doesn’t work from R. There is an open issue about that, but so far it hasn’t been done: https://github.com/apache/incubator-mxnet/issues/7273 Please, use python or C++ to generate both .lst and .rec files. Take a look at this tutorial for more information: https://github.com/leocvml/mxnet-im2rec_tutorial This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] foamliu commented on issue #10393: ImageRecordIOParser2 Assertion failed
foamliu commented on issue #10393: ImageRecordIOParser2 Assertion failed URL: https://github.com/apache/incubator-mxnet/issues/10393#issuecomment-464237936 same issue. Gerkam, have you solved this? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #13735: update wavenet codes
ankkhedia commented on issue #13735: update wavenet codes URL: https://github.com/apache/incubator-mxnet/pull/13735#issuecomment-464236149 @vandanavk @ThomasDelteil Could you please check if your comments have been addressed? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #13757: Set ImageNet data augmentation by default
ankkhedia commented on issue #13757: Set ImageNet data augmentation by default URL: https://github.com/apache/incubator-mxnet/pull/13757#issuecomment-464235359 @vishaalkapoor Could you suggest a way forward on this PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #13895: Add an inference script providing both accuracy and benchmark result for original wide_n_deep example
ankkhedia commented on issue #13895: Add an inference script providing both accuracy and benchmark result for original wide_n_deep example URL: https://github.com/apache/incubator-mxnet/pull/13895#issuecomment-464234602 @vandanavk @larroy Could you please check if your comments are resolved. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #13886: Fixes for wine detection tutorial
ankkhedia commented on issue #13886: Fixes for wine detection tutorial URL: https://github.com/apache/incubator-mxnet/pull/13886#issuecomment-464231336 @larroy Could you look into comments by @aaronmarkham This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] NRauschmayr commented on a change in pull request #14179: Fix READMEs for examples
NRauschmayr commented on a change in pull request #14179: Fix READMEs for examples URL: https://github.com/apache/incubator-mxnet/pull/14179#discussion_r257406974 ## File path: example/gluon/mnist/README.md ## @@ -0,0 +1,37 @@ +# MNIST classification example + +This script shows a simple example how to do image classification with Gluon. +The model is trained on MNIST digits image dataset and the goal is to classify the digits ```0-9```. The model has the following layout: +``` +net = nn.Sequential() +with net.name_scope(): +net.add(nn.Dense(128, activation='relu')) +net.add(nn.Dense(64, activation='relu')) +net.add(nn.Dense(10)) +``` + +The script provides the following commandline arguments: + + +``` +MXNet Gluon MNIST Example + +optional arguments: + -h, --helpshow this help message and exit + --batch-size BATCH_SIZE +batch size for training and testing (default: 100) + --epochs EPOCHS number of epochs to train (default: 10) + --lr LR learning rate (default: 0.1) + --momentum MOMENTUM SGD momentum (default: 0.9) + --cudaTrain on GPU with CUDA + --log-interval N how many batches to wait before logging training +status +``` + +After one epoch we get the following output vector for the given test image: + + Review comment: I moved all images to dmlc/web-data This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] aaronmarkham commented on issue #14172: update raspberry pi install instructions
aaronmarkham commented on issue #14172: update raspberry pi install instructions URL: https://github.com/apache/incubator-mxnet/pull/14172#issuecomment-464161923 @larroy @IvyBazan - can you review? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14178: [WIP] Testing the commit status setter
ankkhedia commented on issue #14178: [WIP] Testing the commit status setter URL: https://github.com/apache/incubator-mxnet/pull/14178#issuecomment-464159250 @perdasilva Thanks for the contribution! @mxnet-label-bot add [pr-work-in-progress] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: Fixes spelling (#14168)
This is an automated email from the ASF dual-hosted git repository. zhasheng 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 b0ab8d5 Fixes spelling (#14168) b0ab8d5 is described below commit b0ab8d54c1ea32ff7d3808d88fd288ee10824951 Author: perdasilva AuthorDate: Fri Feb 15 23:01:33 2019 +0100 Fixes spelling (#14168) --- tests/jenkins/run_test_installation_docs.sh | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tests/jenkins/run_test_installation_docs.sh b/tests/jenkins/run_test_installation_docs.sh index 4f62865..954b370 100755 --- a/tests/jenkins/run_test_installation_docs.sh +++ b/tests/jenkins/run_test_installation_docs.sh @@ -255,12 +255,11 @@ function set_instruction_set() { # $ git clone --recursive https://github.com/apache/incubator-mxnet # $ cd incubator-mxnet # if these commands get executed in the jenkins job, we will be testing the build from source instructions -# against the master branch and not against the version of the repository that Jenkins checksout for testing. +# against the master branch and not against the version of the repository that Jenkins checks out for testing. # This presents a particularly big problem for the version branches and their nightly builds. Because, # we would, in effect, be testing the build from source instructions for one version of MXNet against # the master branch. -# in this function we target the commands cited in the example above, but leave it open for expantion -# in the future. +# in this function we target the commands cited in the example above. # See also gh issue: https://github.com/apache/incubator-mxnet/issues/13800 function filter_build_commands() { filtered_build_commands="${1}"
[GitHub] szha merged pull request #14168: Fixes spelling
szha merged pull request #14168: Fixes spelling URL: https://github.com/apache/incubator-mxnet/pull/14168 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: add Qing's Key to master (#14180)
This is an automated email from the ASF dual-hosted git repository. zhasheng 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 da9afe1 add Qing's Key to master (#14180) da9afe1 is described below commit da9afe1e69b10af983fd2a700f9dc7e3cf76fc89 Author: Lanking AuthorDate: Fri Feb 15 14:00:37 2019 -0800 add Qing's Key to master (#14180) --- KEYS | 90 1 file changed, 90 insertions(+) diff --git a/KEYS b/KEYS index 03b3ed4..7b78ea9 100644 --- a/KEYS +++ b/KEYS @@ -599,3 +599,93 @@ a4LYL628Ksuv1Yxn/Uhb5nDPxU5RKRDeogn07wtaqSG12T9pcIwmExW1RN5kvXc2 3Os= =XL0V -END PGP PUBLIC KEY BLOCK- + +pub rsa2048 2019-02-15 [SC] [expires: 2021-02-14] + 0812952358B12DC30536E7E0C06916C3AB88ABFE +uid [ultimate] Qing Lan +sig 3C06916C3AB88ABFE 2019-02-15 Qing Lan +sub rsa2048 2019-02-15 [E] [expires: 2021-02-14] +sig C06916C3AB88ABFE 2019-02-15 Qing Lan + +-BEGIN PGP PUBLIC KEY BLOCK- + +mQENBFwb8GEBCADjmi5ZXihPfPbLxhNpbD4HVdT7xGEQgfgkeTA4TdFyBP81zM0F +dTHPQOhHzPGkHrVwUPt3ir2hS46q3L8wni3VRkUU8KPbqbS/a7Wl7LHFFS0lU36J +3uQLElZOFITlaL1dl7cIv+c8xCfmOPlmEtNAPIB26sIM5qzc5l4xvNf1H0Oq0wo6 +VKCsYb4el4nys2U3UBYVQjGyBEwwemHQmFPKg6a2bc/2UhWn4Z+//g0hzIpYtT/S +jua16r5SHy6BUtFGWfU6LQIwmxqc0TNqRdkDU0QY0A+nT6cgx6ghp/qxoLOhgken +Uw8rutC/oFg2VzS9yVsJNrkbQq5Fl/Mz/wqNABEBAAG0HWxhbmtpbmcgPGxhbmtp +bmdAbGFua2luZy5uZXQ+iQFUBBMBCAA+FiEECukpOOn0VXTzKWFpZ3gsZdQlyvsF +Alwb8GECGwMFCQPCZwAFCwkIBwIGFQoJCAsCBBYCAwECHgECF4AACgkQZ3gsZdQl +yvtmiAgAnFxzWwPSoHZW9U8ejNtLXxqKmfrUVHDJKVVE0zZPEs9kEk95YZ99jqUS +LagTZwdQOGCJe6Fc9XESW5KdzvnwQyJKwGPCIYDVA/CRuGDbs+tY6HCZ7V1DMGYm +EJsQEzdqWKaNLEVNk1LB3uizEjTS3BIVaVZ0+uovv2joM1mgHX9obGZAZRpBm7zG +TsDwQMuAaWpgnh+BQ9b8O0m6Mdlqh4lAHJmqLdwIz3rD1SLKst0WenjVtyBE1k5t +gBSVfQeMThhcbPITUTLdhAmc27QrsJdDdsH42CVhIJMW6aQnBnEKE3QB+C6DzpTi +UIKqPg99PXmBc9BvlugrZRM4KYiqOLkBDQRcG/BhAQgAzS6PeCCIwImwqSRRnwta +R/7ZbjhMDbnbiHhBKjy08vt2tR8cRK8X7utVwd96u0b3a6J4k0VzMcDP7wBYiY0x +GDbEVsmeXxiqh3+Dv2R+/CBebbswlMJqG+imEaGAAA1bk8U8RrjxuKAUOlyVjEaG +m9flPKOo33SK7UTujfnpSZhA6s0wzRSMn3/26c6+/9vpo8pnqooiD+K7eMhIUfjB ++hFRDJxrEPg9pj3hCJhXA1aCIMrq8jjkm+8PyDQWdt9TdnHLD+b5X+DKNNBpZsIk +IxNTqxNNLan4JZ+70z5m68Lqy4EEHFoHwhn8IFjxOMiJ+6MuP32UdYmvUD3owc8y +nwARAQABiQE8BBgBCAAmFiEECukpOOn0VXTzKWFpZ3gsZdQlyvsFAlwb8GECGwwF +CQPCZwAACgkQZ3gsZdQlyvvYwQf/ctlGApkvSxmhKgCTPBCrudGDpO3QsEF+bR6y +w9WvyvniCt6t5M2A+QPOCqdBvNaP/4wP1H2XghKr9XZcTBlxAJc6pdwxQ6IH6SD8 +syEDVTp0oiQytuyWeNoTM9bOFsliCkjiAVphTDTuVrNEgqdnTMo2cDZgry6gwD0Q +1ZOejcb7kIjiThzaKmgpPnnDNkiK2j25fiNhpiIzEVryLKoXvyoj//C/p8lm9KRk +/2bzRm84uoRsUlrr6I5qizakoODkh5+DlzAaes1fv8ED11zzrlIZjh1vBBgDd3QT +jurgAMn1+WYXhnTVUw8+JjWbmZX4NqJJGZyozkF7rgzI3A9CPJkBDQRcLmPJAQgA +6P5AdfBp4QJKfJC96BjDIrXAFXTytD5wxBsyl1WUcq9vnkX6NZGEtTgsuR/F+FBt +Za7fuDRf+ZT5J7sFQbKFJKRvR8shg1dqnIsoztBD0ESC46trD/YZrisDb09JUXN+ +zvha7jUisA7/pVwfenQM8CpnZJiuKIDU5qCTUr7qOjds6g9aCtUSOK41xLAn1y9b +hKJko4xtjUQu6hXjRXXKltl1N0X5TQCw3sR105xTTWx9nAgFu7cc1uGGuOPCpSms +FPkXFMz+75ff9eXtnmpM+9ZxE8c0jTN7p3U16nd2spsQOqQd83nKb3AbfmWrUYC6 +lmiXGlyOqf8fXSSj/d1ZWwARAQABtB5RaW5nIExhbiA8bGFua2luZzUyMEBsaXZl +LmNvbT6JAVQEEwEIAD4WIQSEDPQiKrv4QoSvRxUvo3RD/I+Y+gUCXC5jyQIbAwUJ +A8JnAAULCQgHAgYVCgkICwIEFgIDAQIeAQIXgAAKCRAvo3RD/I+Y+vjlB/4h3trK +2wGFn396m/jqLqBGObCrOTQYJVycfKdndTEvzIm5/U1gU0Q2BzP25k3gYQjjYKDB +ZN5ETrvRa+NUHzK9p4L4mEQpyWI1WR7jJ19Qg3wfjXW2lOJ8hVvGDILKm7JHEOFb +xf8bUsIftAJ+eDs0dFkDEZT+VLGIXWghd236oUy/uYKL0nLofyFZL3IHiBRZkZQf +6nW+wuYlNXJ+Gy8ATUOKmqbhKScOYLT88nvd6XK8ht3agvATTiEZNa7uWeSOZqp/ +Sx10jUKvKif1L3KiRADb461zQdijx1VOpj3OZPKwCViag14TGXot/YsjAHLCROPg +RH7Ey+WHqMzrkI/EuQENBFwuY8kBCADG74gBceYTyUZbwP6y1xKElXacZ4r2aovT +xoyEXGGovkVjR30GT4XWpKBbqFk5S+QtRnJJGCMeVeKbruZc2Jv5cNe5WL8t20SU +shkBi89h2ZWOgw5c8h+5axjyJfPzWuq0PoXZZ0poFVuSojEFuQKotBWqalri3SKu +E7dUlPQ1JzCj60dX03Lqb2fZZEuwNTf3VwXZZHCgRKw0jhvlr0yfZVcIx4C1Vizo +rkmgZoKbTTkcjBwG4xU+QAwtZqtqGUcZsGwGRD/VPYoDwVknTWJN9JwGasM2gdtG +y0IvhjHfhahZXPyiDsUpkvHyrjaKUwF9qkNLlnmV0LxRJYfNYkk1ABEBAAGJATwE +GAEIACYWIQSEDPQiKrv4QoSvRxUvo3RD/I+Y+gUCXC5jyQIbDAUJA8JnAAAKCRAv +o3RD/I+Y+ql8CADZJxKXAEhBN6ao0+OE2RM/+IPUgIuW/B33KxYN/SewVaUeZvIe +Nu3TWLiFa4eTOjc2hNZxRmQRtpmX7faJaynuz/bsfdtdsscuWxNUUiF3Tdd9/Eon +eFaywH8as9uFwYLRmfdb5cvst97XofQOJBEfolOgyGOEEuEVxs6bFa43/SSYCliE +MjhxubT8NqQmIsHGQ4nltxCbmVZNKi7eAjms+QkWX5bqWMqXF5hJisagXvO1Jr75 +ncoq3Wp+X1b+9x8KDGCVd1F5+d0OksPrnNIQVOS8TMw6bJzXLw4yQni+1ZtYTw8p +9eG+gakLwBnBvqs+7HmuvPNELZX5SsKfwRYzmQENBFxnEhcBCADFLKeuIwb8XnYd +MsAEbQIyueHsON04/XJVvULVHlbpF7X4XyUU2aHxkydFN5yGtJZjE7/BXrPSeoK8 +0qtTblb7ZugUgnyhXaN97RqdD6Lv6kha6EFSylUG9Wyfi1Onr0zlgqQ5TwbaTbLJ +UuzN5d7UIPNFZBrWgvH6AC982M3kMGBWEPRiJgCs/k601oX84g6HbynuxaySts1W +N8mC1TjzVuTRq6Cbfr1RAkNcj3mz66DpZLnHsBUPuQ5u0gJTKZyEGRxtT02eXpMI +61eP+79hnAwqZds9BXwO+3cCx/x3qJRLNDpR4mPq9QqETkqTIQd6JPjWPRh0OQDK +300hDmizABEBAAG0HVFpbmcgTGFuIDxsYW5raW5nQGFwYWNoZS5vcmc+iQFUBBMB
[GitHub] szha merged pull request #14180: add Qing's Key to master
szha merged pull request #14180: add Qing's Key to master URL: https://github.com/apache/incubator-mxnet/pull/14180 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] aaronmarkham commented on issue #14170: add preprocessed data and pretrained model info
aaronmarkham commented on issue #14170: add preprocessed data and pretrained model info URL: https://github.com/apache/incubator-mxnet/pull/14170#issuecomment-464161728 @IvyBazan - can you review? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14171: multi-precision AdamW update op
ankkhedia commented on issue #14171: multi-precision AdamW update op URL: https://github.com/apache/incubator-mxnet/pull/14171#issuecomment-464161325 @eric-haibin-lin Thanks for the contribution! @mxnet-label-bot add [pr-awaiting-review, operator] @apeforest Could you please help review this PR This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14173: [WIP] MXNet AMP (automatic mixed precision)
ankkhedia commented on issue #14173: [WIP] MXNet AMP (automatic mixed precision) URL: https://github.com/apache/incubator-mxnet/pull/14173#issuecomment-464157769 @ptrendx Thanks for the contribution! @mxnet-label-bot add [pr-work-in-progress ] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.
This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 1a483a6 Bump the publish timestamp. 1a483a6 is described below commit 1a483a6b390a3e3086828df282972261376fd493 Author: mxnet-ci AuthorDate: Fri Feb 15 19:06:21 2019 + Bump the publish timestamp. --- date.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/date.txt b/date.txt new file mode 100644 index 000..e641cc7 --- /dev/null +++ b/date.txt @@ -0,0 +1 @@ +Fri Feb 15 19:06:21 UTC 2019
[GitHub] ankkhedia commented on issue #14172: update raspberry pi install instructions
ankkhedia commented on issue #14172: update raspberry pi install instructions URL: https://github.com/apache/incubator-mxnet/pull/14172#issuecomment-464162031 @aaronmarkham Thanks for the contribution @mxnet-label-bot add [pr-awaiting-review] @marcoabreu @larroy @lebeg Could you help review this PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia edited a comment on issue #14169: [MXNET-1305] Scala interpreter instructions
ankkhedia edited a comment on issue #14169: [MXNET-1305] Scala interpreter instructions URL: https://github.com/apache/incubator-mxnet/pull/14169#issuecomment-464163054 @zachgk Thanks for the contribution! @mxnet-label-bot add [pr-awaiting-review, Scala] @lanking520 @piyushghai @andrewfayres Could you please help review this PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14168: Fixes spelling
ankkhedia commented on issue #14168: Fixes spelling URL: https://github.com/apache/incubator-mxnet/pull/14168#issuecomment-464160687 @perdasilva Thanks for your contribution! @mxnet-label-bot add [pr-awaiting-review] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #12792: Fix broken amalgamation
ankkhedia commented on issue #12792: Fix broken amalgamation URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-464156641 @ufoym Could you please fix build errors? @zheng-da @reminisce Could you please help review the PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14094: Enhance gpu quantization
ankkhedia commented on issue #14094: Enhance gpu quantization URL: https://github.com/apache/incubator-mxnet/pull/14094#issuecomment-464154983 @rajeshii Thanks for the quick turnaround. Could you please look into comments by @anirudh2290 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14179: Fix READMEs for examples
ankkhedia commented on issue #14179: Fix READMEs for examples URL: https://github.com/apache/incubator-mxnet/pull/14179#issuecomment-464157367 @NRauschmayr Thanks for the contribution! @mxnet-label-bot add [pr-awaiting-review, Example] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: Parallelize CPU version and add GPU version of boolean_mask op (#14090)
This is an automated email from the ASF dual-hosted git repository. zhengda 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 ba97fb6 Parallelize CPU version and add GPU version of boolean_mask op (#14090) ba97fb6 is described below commit ba97fb6663bf1116c8db492668fbc984e3541612 Author: HyperZealot <40650949+hyperzea...@users.noreply.github.com> AuthorDate: Fri Feb 15 13:28:24 2019 -0800 Parallelize CPU version and add GPU version of boolean_mask op (#14090) --- src/operator/contrib/boolean_mask-inl.h | 102 +++- src/operator/contrib/boolean_mask.cc| 116 +- src/operator/contrib/boolean_mask.cu| 165 tests/python/unittest/test_operator.py | 2 - 4 files changed, 313 insertions(+), 72 deletions(-) diff --git a/src/operator/contrib/boolean_mask-inl.h b/src/operator/contrib/boolean_mask-inl.h index ac0681b..775981f 100644 --- a/src/operator/contrib/boolean_mask-inl.h +++ b/src/operator/contrib/boolean_mask-inl.h @@ -50,83 +50,53 @@ struct BooleanMaskParam : public dmlc::Parameter { } }; +struct BooleanMaskForwardKernel { + template + static void MSHADOW_XINLINE Map(int i, + DType* out, + const DType* data, + const int32_t* idx, + const size_t col_size) { +int row_id = i / col_size; +int col_id = i % col_size; +int32_t prev = (row_id == 0) ? 0 : idx[row_id - 1]; +int32_t curr = idx[row_id]; +if (prev != curr) { + out[prev * col_size + col_id] = data[i]; +} + } +}; + +struct BooleanMaskBackwardKernel { + template + static void MSHADOW_XINLINE Map(int i, + DType* igrad, + const DType* ograd, + const int32_t* idx, + const size_t col_size) { +int row_id = i / col_size; +int col_id = i % col_size; +int32_t prev = (row_id == 0) ? 0 : idx[row_id - 1]; +int32_t curr = idx[row_id]; +if (prev != curr) { + igrad[i] = ograd[prev * col_size + col_id]; +} + } +}; + template inline void BooleanMaskForward(const nnvm::NodeAttrs& attrs, const OpContext , const std::vector , const std::vector , - const std::vector ) { - // TODO(@junrushao1994): This implementation is a proof-of-concept, - // hence very slow actually. Performance should be improved in the future. - CHECK_EQ(inputs.size(), 2U); - CHECK_EQ(outputs.size(), 1U); - const BooleanMaskParam& param = nnvm::get(attrs.parsed); - const int axis = param.axis; - const NDArray = inputs[0]; - const NDArray = inputs[1]; - const NDArray = outputs[0]; - CHECK_EQ(axis, 0) << "Not supported yet"; - CHECK_EQ(data.shape()[axis], idx.shape()[0]); - CHECK_EQ(idx.shape().ndim(), 1U); - // count the number of 1s in `idx`, so that we could know the output dimension - size_t valid_num = 0; - MSHADOW_TYPE_SWITCH(idx.dtype(), DType, { -DType* idx_dptr = idx.data().dptr(); -int length = idx.shape()[0]; -for (int i = 0; i < length; i++) { - if (idx_dptr[i]) { -++valid_num; - } -} - }); - // set the output shape forcefully - TShape s = data.shape(); - s[axis] = valid_num; - const_cast(out).Init(s); - // do the copy - MSHADOW_TYPE_SWITCH(idx.dtype(), DType, { -DType* idx_dptr = idx.data().dptr(); -int length = idx.shape()[0]; -mshadow::Stream *stream = ctx.get_stream(); -for (int i = 0, j = 0; i < length; ++i) { - if (idx_dptr[i]) { -NDArray src = data.At(i); -NDArray dst = out.At(j++); -CHECK(src.shape() == dst.shape()); -mxnet_op::copy(stream, dst.data(), src.data()); - } -} - }); -} + const std::vector ); template inline void BooleanMaskBackward(const nnvm::NodeAttrs& attrs, const OpContext , const std::vector , const std::vector , -const std::vector ) { - CHECK_EQ(inputs.size(), 3U); - CHECK_EQ(outputs.size(), 2U); - // inputs: {ograd, data, idx} - // outputs: {igrad_data, igrad_idx} - const NDArray& ograd = inputs[0]; - const NDArray& idx = inputs[2]; - const NDArray& igrad_data = outputs[0]; - MSHADOW_TYPE_SWITCH(idx.dtype(), DType, { -DType* idx_dptr = idx.data().dptr(); -int length = idx.shape()[0]; -mshadow::Stream *stream = ctx.get_stream(); -Fill(stream, igrad_data.data(), req[0], 0); -for (int i = 0, j = 0; i < length; ++i) { - if (idx_dptr[i]) { -NDArray src = ograd.At(j++); -
[GitHub] zheng-da merged pull request #14090: Add GPU version of contrib.boolean_mask
zheng-da merged pull request #14090: Add GPU version of contrib.boolean_mask URL: https://github.com/apache/incubator-mxnet/pull/14090 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
svn commit: r32522 - in /dev/incubator/mxnet/1.4.0.rc3: ./ apache-mxnet-src-1.4.0.rc3-incubating.tar.gz apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.asc apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.sh
Author: lanking Date: Fri Feb 15 21:08:57 2019 New Revision: 32522 Log: Add mxnet-1.4.0.rc3 Added: dev/incubator/mxnet/1.4.0.rc3/ dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz (with props) dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.asc dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.sha512 Added: dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz == Binary file - no diff available. Propchange: dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz -- svn:mime-type = application/octet-stream Added: dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.asc == --- dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.asc (added) +++ dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.asc Fri Feb 15 21:08:57 2019 @@ -0,0 +1,11 @@ +-BEGIN PGP SIGNATURE- + +iQEzBAABCAAdFiEECBKVI1ixLcMFNufgwGkWw6uIq/4FAlxnHbsACgkQwGkWw6uI +q/5f3wf/fM/42LnOMBfBIBA2u6/xrjzOB4r8YGYPZsPnaq1nQsLZLmLHXyCfsbQP +cQe2g5/AkouWbLkiDIrFh2p3KBYwXFT9DTF6cx4SbyV7CFj0Lnk3iG8Cwm0nlM2s +fTvOpXqyQp9PAdITd9ByTnkU+TKQQogCa2hxOzPeWgTW7K7HNv7CsGjRVZ3FChXO +heLuwxSSSv6XWtdV47jmocoiwagnY+6edRfMw52A4+pcc77epP8T9a4zUo6vIAGG +flFf1u0xQKJdKdCC4/UcGpKs4S94T+5RppAcV1+fCY28KTzTIFO29ns1P/dP1MJE +q6+pysOWskJdqgQ0IPN6ualLvHcdmA== +=XD2b +-END PGP SIGNATURE- Added: dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.sha512 == --- dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.sha512 (added) +++ dev/incubator/mxnet/1.4.0.rc3/apache-mxnet-src-1.4.0.rc3-incubating.tar.gz.sha512 Fri Feb 15 21:08:57 2019 @@ -0,0 +1 @@ +977b4048541375c8c70865464927d83b9f8824cc58bfbfc52bcbfb6eab13f5fae79a491812b6c2c080357bb93312b31941f4fac8fa386fda18bc5d9f558930f8 apache-mxnet-src-1.4.0.rc3-incubating.tar.gz
[GitHub] piyushghai commented on issue #14141: [v1.4.x] Update MKL-DNN to fix the OSX build issue
piyushghai commented on issue #14141: [v1.4.x] Update MKL-DNN to fix the OSX build issue URL: https://github.com/apache/incubator-mxnet/pull/14141#issuecomment-464198742 Update --> Ignore my previous comment. I resolved the issue thanks to @lanking520 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14180: add Qing's Key to master
ankkhedia commented on issue #14180: add Qing's Key to master URL: https://github.com/apache/incubator-mxnet/pull/14180#issuecomment-464192122 @mxnet-label-bot add [pr-awaiting-review] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.
This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 2c84639 Bump the publish timestamp. 2c84639 is described below commit 2c84639ed5226e479d39462e210b749cbcc78c4a Author: mxnet-ci AuthorDate: Fri Feb 15 20:41:33 2019 + Bump the publish timestamp. --- date.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/date.txt b/date.txt new file mode 100644 index 000..f33b630 --- /dev/null +++ b/date.txt @@ -0,0 +1 @@ +Fri Feb 15 20:41:33 UTC 2019
[GitHub] piyushghai commented on issue #14141: [v1.4.x] Update MKL-DNN to fix the OSX build issue
piyushghai commented on issue #14141: [v1.4.x] Update MKL-DNN to fix the OSX build issue URL: https://github.com/apache/incubator-mxnet/pull/14141#issuecomment-464189583 @TaoLv The OSX build from source is not working still. It's broken. I tried a simple command : ``` make -j4 ``` This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: R-Package Makefile (#14068)
This is an automated email from the ASF dual-hosted git repository. the 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 3d0ffd2 R-Package Makefile (#14068) 3d0ffd2 is described below commit 3d0ffd23382ca85b24357c5c03d169855a502672 Author: jeremiedb AuthorDate: Fri Feb 15 15:23:01 2019 -0500 R-Package Makefile (#14068) * rpkg Makefile * rpkg makefile fix --- Makefile | 8 ++-- 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/Makefile b/Makefile index cc7aebc..ed45019 100644 --- a/Makefile +++ b/Makefile @@ -607,11 +607,7 @@ rpkg: fi mkdir -p R-package/inst/include - cp -rf include/* R-package/inst/include - rm R-package/inst/include/dmlc - rm R-package/inst/include/nnvm - cp -rf 3rdparty/dmlc-core/include/* R-package/inst/include/ - cp -rf 3rdparty/tvm/nnvm/include/* R-package/inst/include + cp -rl include/* R-package/inst/include Rscript -e "if(!require(devtools)){install.packages('devtools', repo = 'https://cloud.r-project.org/')}" Rscript -e "if(!require(roxygen2)||packageVersion('roxygen2') < '6.1.1'){install.packages('roxygen2', repo = 'https://cloud.r-project.org/')}" Rscript -e "library(devtools); library(methods); options(repos=c(CRAN='https://cloud.r-project.org/')); install_deps(pkg='R-package', dependencies = TRUE)" @@ -620,7 +616,7 @@ rpkg: R CMD INSTALL R-package Rscript -e "require(mxnet); mxnet:::mxnet.export('R-package'); warnings()" rm R-package/NAMESPACE - Rscript -e "devtools::document('R-package'); warnings()" + Rscript -e "devtools::document('R-package');warnings()" R CMD INSTALL R-package rpkgtest:
[GitHub] hetong007 merged pull request #14068: R-Package Makefile
hetong007 merged pull request #14068: R-Package Makefile URL: https://github.com/apache/incubator-mxnet/pull/14068 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: support dot(vector, vector) for fp16 inputs on GPU (#14102)
This is an automated email from the ASF dual-hosted git repository. zhasheng 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 549ecde support dot(vector, vector) for fp16 inputs on GPU (#14102) 549ecde is described below commit 549ecde472f208f1c458bfdf585318340a586639 Author: Haibin Lin AuthorDate: Fri Feb 15 12:19:05 2019 -0800 support dot(vector, vector) for fp16 inputs on GPU (#14102) * support dot(vector, vector) for fp16 inputs on GPU * add comments * update comments --- src/operator/tensor/dot-inl.h | 3 +- tests/python/unittest/test_operator.py | 73 ++ 2 files changed, 41 insertions(+), 35 deletions(-) diff --git a/src/operator/tensor/dot-inl.h b/src/operator/tensor/dot-inl.h index 69c35f8..69f87ae 100644 --- a/src/operator/tensor/dot-inl.h +++ b/src/operator/tensor/dot-inl.h @@ -83,7 +83,8 @@ void DotForward_(const nnvm::NodeAttrs& attrs, (outputs[0].type_flag_ == kFloat16 && ctx.run_ctx.ctx.dev_mask() == mshadow::gpu::kDevMask)) << "dot only supports float32/float64 for CPU, and float16/float32/float64 for GPU"; MSHADOW_REAL_TYPE_SWITCH(outputs[0].type_flag_, DType, { -if (inputs[0].ndim() == 1 && inputs[1].ndim() == 1) { +// VectorDot() with fp16 is not supported in mshadow. Dispatch to dot() instead. +if (inputs[0].ndim() == 1 && inputs[1].ndim() == 1 && inputs[0].type_flag_ != kFloat16) { CHECK_NE(req[0], kAddTo) << "AddTo not yet supported"; Tensor out = outputs[0].get(s); VectorDot(out, diff --git a/tests/python/unittest/test_operator.py b/tests/python/unittest/test_operator.py index fc003b2..b0c640b 100644 --- a/tests/python/unittest/test_operator.py +++ b/tests/python/unittest/test_operator.py @@ -2631,47 +2631,52 @@ def test_stn_valid_sampling(): ) + target_shape)) -# @haojin2: Getting rid of fixed seed as flakiness could not be reproduced, -# tracked at https://github.com/apache/incubator-mxnet/issues/11714 @with_seed() def test_dot(): -ctx=default_context() +ctx = default_context() dtypes = ['float32', 'float64'] +ndims = [2] if ctx.device_type == 'gpu': dtypes += ['float16'] +ndims += [1] # Test normal dot. -for data_type in dtypes: -for m in range(1, 5): -for k in range(1, 5): -for n in range(1, 5): -a_npy = np.random.normal(0, 1, (m, k)) -a_npy = a_npy.astype(data_type) -b_npy = np.random.normal(0, 1, (k, n)) -b_npy = b_npy.astype(data_type) -c_npy = np.empty((m, n), dtype=data_type) -ograd_npy = np.random.normal(0, 1, (m, n)) -ograd_npy = ograd_npy.astype(data_type) -agrad_npy = np.empty((m, k), dtype=data_type) -bgrad_npy = np.empty((k, n), dtype=data_type) -c_npy[:, :] = np.dot(a_npy[:, :], b_npy[:, :]) -bgrad_npy[:, :] = np.dot(a_npy[:, :].T, ograd_npy[:, :]) -agrad_npy[:, :] = np.dot(ograd_npy[:, :], b_npy[:, :].T) -a = mx.sym.Variable('a', dtype=data_type) -b = mx.sym.Variable('b', dtype=data_type) -c = mx.sym.dot(a, b) -exe = c.simple_bind(ctx=ctx, a=a_npy.shape, b=b_npy.shape) -outputs = exe.forward(is_train=True, a=a_npy, b=b_npy) -assert_almost_equal(outputs[0].asnumpy(), c_npy, -rtol=1e-2 if data_type == 'float16' else 1e-3, -atol=1e-2 if data_type == 'float16' else 1e-3) -exe.backward(out_grads=[mx.nd.array(ograd_npy, mx.cpu()).astype(data_type)]) -assert_almost_equal(exe.grad_dict['a'].asnumpy(), agrad_npy, -rtol=1e-2 if data_type == 'float16' else 1e-3, -atol=1e-2 if data_type == 'float16' else 1e-3) -assert_almost_equal(exe.grad_dict['b'].asnumpy(), bgrad_npy, -rtol=1e-2 if data_type == 'float16' else 1e-3, -atol=1e-2 if data_type == 'float16' else 1e-3) +for ndim in ndims: +for data_type in dtypes: +for m in range(1, 5): +for k in range(1, 5): +if ndim == 1 and k != 1: +pass +for n in range(1, 5): +a_shape = (m, k) if ndim == 2 else (m,) +b_shape = (k, n) if ndim == 2 else (n,) +a_npy = np.random.normal(0, 1, (m, k)) +a_npy = a_npy.astype(data_type) +
[GitHub] szha merged pull request #14102: support dot(vector, vector) for fp16 inputs on GPU
szha merged pull request #14102: support dot(vector, vector) for fp16 inputs on GPU URL: https://github.com/apache/incubator-mxnet/pull/14102 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] leleamol commented on issue #14152: data layer grad problem
leleamol commented on issue #14152: data layer grad problem URL: https://github.com/apache/incubator-mxnet/issues/14152#issuecomment-464181528 @songziqin Can you please provide a mimim reproducible example? It isn't clear what are the expectations. @mxnet-label-bot add [Pending Requester Info] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] leleamol commented on issue #14159: how to infer using fp16 for c++ api
leleamol commented on issue #14159: how to infer using fp16 for c++ api URL: https://github.com/apache/incubator-mxnet/issues/14159#issuecomment-464180802 @IvyGongoogle Currently cpp-package APIs doesn't support fp16 datatype. We can mark this question as feature request. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
svn commit: r32521 - /dev/incubator/mxnet/KEYS
Author: lanking Date: Fri Feb 15 20:00:06 2019 New Revision: 32521 Log: update keys file for Qing Lan Modified: dev/incubator/mxnet/KEYS Modified: dev/incubator/mxnet/KEYS == --- dev/incubator/mxnet/KEYS (original) +++ dev/incubator/mxnet/KEYS Fri Feb 15 20:00:06 2019 @@ -717,3 +717,93 @@ S+td6k7QO2oVhdRDFWt8XesaIk9/2iEnZ+MAalCE hwYZBxZQLw== =yxZo -END PGP PUBLIC KEY BLOCK- + +pub rsa2048 2019-02-15 [SC] [expires: 2021-02-14] + 0812952358B12DC30536E7E0C06916C3AB88ABFE +uid [ultimate] Qing Lan +sig 3C06916C3AB88ABFE 2019-02-15 Qing Lan +sub rsa2048 2019-02-15 [E] [expires: 2021-02-14] +sig C06916C3AB88ABFE 2019-02-15 Qing Lan + +-BEGIN PGP PUBLIC KEY BLOCK- + +mQENBFwb8GEBCADjmi5ZXihPfPbLxhNpbD4HVdT7xGEQgfgkeTA4TdFyBP81zM0F +dTHPQOhHzPGkHrVwUPt3ir2hS46q3L8wni3VRkUU8KPbqbS/a7Wl7LHFFS0lU36J +3uQLElZOFITlaL1dl7cIv+c8xCfmOPlmEtNAPIB26sIM5qzc5l4xvNf1H0Oq0wo6 +VKCsYb4el4nys2U3UBYVQjGyBEwwemHQmFPKg6a2bc/2UhWn4Z+//g0hzIpYtT/S +jua16r5SHy6BUtFGWfU6LQIwmxqc0TNqRdkDU0QY0A+nT6cgx6ghp/qxoLOhgken +Uw8rutC/oFg2VzS9yVsJNrkbQq5Fl/Mz/wqNABEBAAG0HWxhbmtpbmcgPGxhbmtp +bmdAbGFua2luZy5uZXQ+iQFUBBMBCAA+FiEECukpOOn0VXTzKWFpZ3gsZdQlyvsF +Alwb8GECGwMFCQPCZwAFCwkIBwIGFQoJCAsCBBYCAwECHgECF4AACgkQZ3gsZdQl +yvtmiAgAnFxzWwPSoHZW9U8ejNtLXxqKmfrUVHDJKVVE0zZPEs9kEk95YZ99jqUS +LagTZwdQOGCJe6Fc9XESW5KdzvnwQyJKwGPCIYDVA/CRuGDbs+tY6HCZ7V1DMGYm +EJsQEzdqWKaNLEVNk1LB3uizEjTS3BIVaVZ0+uovv2joM1mgHX9obGZAZRpBm7zG +TsDwQMuAaWpgnh+BQ9b8O0m6Mdlqh4lAHJmqLdwIz3rD1SLKst0WenjVtyBE1k5t +gBSVfQeMThhcbPITUTLdhAmc27QrsJdDdsH42CVhIJMW6aQnBnEKE3QB+C6DzpTi +UIKqPg99PXmBc9BvlugrZRM4KYiqOLkBDQRcG/BhAQgAzS6PeCCIwImwqSRRnwta +R/7ZbjhMDbnbiHhBKjy08vt2tR8cRK8X7utVwd96u0b3a6J4k0VzMcDP7wBYiY0x +GDbEVsmeXxiqh3+Dv2R+/CBebbswlMJqG+imEaGAAA1bk8U8RrjxuKAUOlyVjEaG +m9flPKOo33SK7UTujfnpSZhA6s0wzRSMn3/26c6+/9vpo8pnqooiD+K7eMhIUfjB ++hFRDJxrEPg9pj3hCJhXA1aCIMrq8jjkm+8PyDQWdt9TdnHLD+b5X+DKNNBpZsIk +IxNTqxNNLan4JZ+70z5m68Lqy4EEHFoHwhn8IFjxOMiJ+6MuP32UdYmvUD3owc8y +nwARAQABiQE8BBgBCAAmFiEECukpOOn0VXTzKWFpZ3gsZdQlyvsFAlwb8GECGwwF +CQPCZwAACgkQZ3gsZdQlyvvYwQf/ctlGApkvSxmhKgCTPBCrudGDpO3QsEF+bR6y +w9WvyvniCt6t5M2A+QPOCqdBvNaP/4wP1H2XghKr9XZcTBlxAJc6pdwxQ6IH6SD8 +syEDVTp0oiQytuyWeNoTM9bOFsliCkjiAVphTDTuVrNEgqdnTMo2cDZgry6gwD0Q +1ZOejcb7kIjiThzaKmgpPnnDNkiK2j25fiNhpiIzEVryLKoXvyoj//C/p8lm9KRk +/2bzRm84uoRsUlrr6I5qizakoODkh5+DlzAaes1fv8ED11zzrlIZjh1vBBgDd3QT +jurgAMn1+WYXhnTVUw8+JjWbmZX4NqJJGZyozkF7rgzI3A9CPJkBDQRcLmPJAQgA +6P5AdfBp4QJKfJC96BjDIrXAFXTytD5wxBsyl1WUcq9vnkX6NZGEtTgsuR/F+FBt +Za7fuDRf+ZT5J7sFQbKFJKRvR8shg1dqnIsoztBD0ESC46trD/YZrisDb09JUXN+ +zvha7jUisA7/pVwfenQM8CpnZJiuKIDU5qCTUr7qOjds6g9aCtUSOK41xLAn1y9b +hKJko4xtjUQu6hXjRXXKltl1N0X5TQCw3sR105xTTWx9nAgFu7cc1uGGuOPCpSms +FPkXFMz+75ff9eXtnmpM+9ZxE8c0jTN7p3U16nd2spsQOqQd83nKb3AbfmWrUYC6 +lmiXGlyOqf8fXSSj/d1ZWwARAQABtB5RaW5nIExhbiA8bGFua2luZzUyMEBsaXZl +LmNvbT6JAVQEEwEIAD4WIQSEDPQiKrv4QoSvRxUvo3RD/I+Y+gUCXC5jyQIbAwUJ +A8JnAAULCQgHAgYVCgkICwIEFgIDAQIeAQIXgAAKCRAvo3RD/I+Y+vjlB/4h3trK +2wGFn396m/jqLqBGObCrOTQYJVycfKdndTEvzIm5/U1gU0Q2BzP25k3gYQjjYKDB +ZN5ETrvRa+NUHzK9p4L4mEQpyWI1WR7jJ19Qg3wfjXW2lOJ8hVvGDILKm7JHEOFb +xf8bUsIftAJ+eDs0dFkDEZT+VLGIXWghd236oUy/uYKL0nLofyFZL3IHiBRZkZQf +6nW+wuYlNXJ+Gy8ATUOKmqbhKScOYLT88nvd6XK8ht3agvATTiEZNa7uWeSOZqp/ +Sx10jUKvKif1L3KiRADb461zQdijx1VOpj3OZPKwCViag14TGXot/YsjAHLCROPg +RH7Ey+WHqMzrkI/EuQENBFwuY8kBCADG74gBceYTyUZbwP6y1xKElXacZ4r2aovT +xoyEXGGovkVjR30GT4XWpKBbqFk5S+QtRnJJGCMeVeKbruZc2Jv5cNe5WL8t20SU +shkBi89h2ZWOgw5c8h+5axjyJfPzWuq0PoXZZ0poFVuSojEFuQKotBWqalri3SKu +E7dUlPQ1JzCj60dX03Lqb2fZZEuwNTf3VwXZZHCgRKw0jhvlr0yfZVcIx4C1Vizo +rkmgZoKbTTkcjBwG4xU+QAwtZqtqGUcZsGwGRD/VPYoDwVknTWJN9JwGasM2gdtG +y0IvhjHfhahZXPyiDsUpkvHyrjaKUwF9qkNLlnmV0LxRJYfNYkk1ABEBAAGJATwE +GAEIACYWIQSEDPQiKrv4QoSvRxUvo3RD/I+Y+gUCXC5jyQIbDAUJA8JnAAAKCRAv +o3RD/I+Y+ql8CADZJxKXAEhBN6ao0+OE2RM/+IPUgIuW/B33KxYN/SewVaUeZvIe +Nu3TWLiFa4eTOjc2hNZxRmQRtpmX7faJaynuz/bsfdtdsscuWxNUUiF3Tdd9/Eon +eFaywH8as9uFwYLRmfdb5cvst97XofQOJBEfolOgyGOEEuEVxs6bFa43/SSYCliE +MjhxubT8NqQmIsHGQ4nltxCbmVZNKi7eAjms+QkWX5bqWMqXF5hJisagXvO1Jr75 +ncoq3Wp+X1b+9x8KDGCVd1F5+d0OksPrnNIQVOS8TMw6bJzXLw4yQni+1ZtYTw8p +9eG+gakLwBnBvqs+7HmuvPNELZX5SsKfwRYzmQENBFxnEhcBCADFLKeuIwb8XnYd +MsAEbQIyueHsON04/XJVvULVHlbpF7X4XyUU2aHxkydFN5yGtJZjE7/BXrPSeoK8 +0qtTblb7ZugUgnyhXaN97RqdD6Lv6kha6EFSylUG9Wyfi1Onr0zlgqQ5TwbaTbLJ +UuzN5d7UIPNFZBrWgvH6AC982M3kMGBWEPRiJgCs/k601oX84g6HbynuxaySts1W +N8mC1TjzVuTRq6Cbfr1RAkNcj3mz66DpZLnHsBUPuQ5u0gJTKZyEGRxtT02eXpMI +61eP+79hnAwqZds9BXwO+3cCx/x3qJRLNDpR4mPq9QqETkqTIQd6JPjWPRh0OQDK +300hDmizABEBAAG0HVFpbmcgTGFuIDxsYW5raW5nQGFwYWNoZS5vcmc+iQFUBBMB +CAA+FiEECBKVI1ixLcMFNufgwGkWw6uIq/4FAlxnEhcCGwMFCQPCZwAFCwkIBwIG +FQoJCAsCBBYCAwECHgECF4AACgkQwGkWw6uIq/5KDgf/S8XUCzm+aVJUAcx9BuQV +SzvG8C1Bf/XrYXcm3NxEOMejdkt8FAnaPt9dZM3U8CaMu1a90iiuLyvr3aJvb7/K +/WuaK79jFB+zz8bifQANXgwAQPiUjBU/8FNOjiEywaSYwIvy8SMo5lLs2edV4Y8g +J73BU/WID3aH/BK4sclAs3brNY5le3B2GiAXhqCWB567ZTlc6vpf1DBaivVX8J0H
[GitHub] lanking520 opened a new pull request #14180: add Qing's Key to master
lanking520 opened a new pull request #14180: add Qing's Key to master URL: https://github.com/apache/incubator-mxnet/pull/14180 ## Description ## Add Qing Lan's key to master for release process ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [ ] The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant [JIRA issue](https://issues.apache.org/jira/projects/MXNET/issues) created (except PRs with tiny changes) - [ ] Changes are complete (i.e. I finished coding on this PR) - [ ] All changes have test coverage: - Unit tests are added for small changes to verify correctness (e.g. adding a new operator) - Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore) - Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL) - [ ] Code is well-documented: - For user-facing API changes, API doc string has been updated. - For new C++ functions in header files, their functionalities and arguments are documented. - For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable - Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html - [ ] To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] IvyBazan commented on issue #14172: update raspberry pi install instructions
IvyBazan commented on issue #14172: update raspberry pi install instructions URL: https://github.com/apache/incubator-mxnet/pull/14172#issuecomment-464177352 LGTM This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] leleamol commented on issue #13996: Make error
leleamol commented on issue #13996: Make error URL: https://github.com/apache/incubator-mxnet/issues/13996#issuecomment-464176379 @yuyijie1995 Thanks for the update. If the issue is solved, I would recommend closing this issue. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: add preprocessed data and pretrained model info; minor format/spelling fixes (#14170)
This is an automated email from the ASF dual-hosted git repository. thomasdelteil 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 e2c69f3 add preprocessed data and pretrained model info; minor format/spelling fixes (#14170) e2c69f3 is described below commit e2c69f30ab1538b2cce746849b00e17fa7e7b573 Author: Aaron Markham AuthorDate: Fri Feb 15 11:43:58 2019 -0800 add preprocessed data and pretrained model info; minor format/spelling fixes (#14170) --- example/gluon/lipnet/README.md | 93 ++ 1 file changed, 67 insertions(+), 26 deletions(-) diff --git a/example/gluon/lipnet/README.md b/example/gluon/lipnet/README.md index 894384b..89c27a1 100644 --- a/example/gluon/lipnet/README.md +++ b/example/gluon/lipnet/README.md @@ -21,14 +21,16 @@ --- -Gluon implementation of [LipNet: End-to-End Sentence-level Lipreading](https://arxiv.org/abs/1611.01599) +This is a Gluon implementation of [LipNet: End-to-End Sentence-level Lipreading](https://arxiv.org/abs/1611.01599) ![net_structure](asset/network_structure.png) +![sample output](https://user-images.githubusercontent.com/11376047/52533982-d7227680-2d7e-11e9-9f18-c15b952faf0e.png) + ## Requirements - Python 3.6.4 -- MXnet 1.3.0 -- The Required Disk Space: 35Gb +- MXNet 1.3.0 +- Required disk space: 35 GB ``` pip install -r requirements.txt ``` @@ -37,18 +39,58 @@ pip install -r requirements.txt ## The Data - The GRID audiovisual sentence corpus (http://spandh.dcs.shef.ac.uk/gridcorpus/) - - GRID is a large multitalker audiovisual sentence corpus to support joint computational-behavioral studies in speech perception. In brief, the corpus consists of high-quality audio and video (facial) recordings of 1000 sentences spoken by each of 34 talkers (18 male, 16 female). Sentences are of the form "put red at G9 now". The corpus, together with transcriptions, is freely available for research use. + - GRID is a large multi-talker audiovisual sentence corpus to support joint computational-behavioral studies in speech perception. In brief, the corpus consists of high-quality audio and video (facial) recordings of 1000 sentences spoken by each of 34 talkers (18 male, 16 female). Sentences are of the form "put red at G9 now". The corpus, together with transcriptions, is freely available for research use. - Video: (normal)(480 M each) - Each movie has one sentence consist of 6 words. -- Align: word alignments(190 K each) +- Align: word alignments (190 K each) - One align has 6 words. Each word has start time and end time. But this tutorial needs just sentence because of using ctc-loss. - + --- +## Pretrained model +You can train the model yourself in the following sections, you can test a pretrained model's inference, or resume training from the model checkpoint. To work with the provided pretrained model, first download it, then run one of the provided Python scripts for inference (infer.py) or training (main.py). + +* Download the [pretrained model](https://github.com/soeque1/temp_files/files/2848870/epoches_81_loss_15.7157.zip) +* Try inference with the following: + +``` +python infer.py model_path='checkpoint/epoches_81_loss_15.7157' +``` + +* Resume training with the following: + +``` +python main.py model_path='checkpoint/epoches_81_loss_15.7157' +``` + ## Prepare the Data -### (1) Download the data + +You can prepare the data yourself, or you can download preprocessed data. + +### Option 1 - Download the preprocessed data + +There are two download routes provided for the preprocessed data. + + Download and untar the data +To download tar zipped files by link, download the following files and extract in a folder called `data` in the root of this example folder. You should have the following structure: +``` +/lipnet/data/align +/lipnet/data/datasets +``` + +* [align files](https://mxnet-public.s3.amazonaws.com/lipnet/data-archives/align.tgz) +* [datasets files](https://mxnet-public.s3.amazonaws.com/lipnet/data-archives/datasets.tgz) + + Use AWS CLI to sync the data +To get the folders and files all unzipped with AWS CLI, can use the following command. This will provide the folder structure for you. Run this command from `/lipnet/`: + +``` + aws s3 sync s3://mxnet-public/lipnet/data . +``` + +### Option 2 (part 1)- Download the raw dataset - Outputs - - The Total Moives(mp4): 16GB + - The Total Movies(mp4): 16GB - The Total Aligns(text): 134MB - Arguments - src_path : Path for videos (default='./data/mp4s/') @@ -59,17 +101,17 @@ pip install -r requirements.txt cd ./utils && python download_data.py --n_process=$(nproc) ``` -### (2) Preprocess the Data: Extracting the mouth images from a video and save it. +### Option 2 (part 2) Preprocess the raw dataset: Extracting the mouth images from a video and save it
[GitHub] ThomasDelteil merged pull request #14170: add preprocessed data and pretrained model info
ThomasDelteil merged pull request #14170: add preprocessed data and pretrained model info URL: https://github.com/apache/incubator-mxnet/pull/14170 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #13966: upload example code for snail with gluon
ankkhedia commented on issue #13966: upload example code for snail with gluon URL: https://github.com/apache/incubator-mxnet/pull/13966#issuecomment-464172985 @ThomasDelteil ping for review! This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14068: R-Package Makefile
ankkhedia commented on issue #14068: R-Package Makefile URL: https://github.com/apache/incubator-mxnet/pull/14068#issuecomment-464172784 @hetong007 @marcoabreu Could you please help review? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on a change in pull request #14170: add preprocessed data and pretrained model info
ankkhedia commented on a change in pull request #14170: add preprocessed data and pretrained model info URL: https://github.com/apache/incubator-mxnet/pull/14170#discussion_r257362887 ## File path: example/gluon/lipnet/README.md ## @@ -21,14 +21,16 @@ --- -Gluon implementation of [LipNet: End-to-End Sentence-level Lipreading](https://arxiv.org/abs/1611.01599) +This is a Gluon implementation of [LipNet: End-to-End Sentence-level Lipreading](https://arxiv.org/abs/1611.01599) ![net_structure](asset/network_structure.png) +![sample output](https://user-images.githubusercontent.com/11376047/52533982-d7227680-2d7e-11e9-9f18-c15b952faf0e.png) + ## Requirements - Python 3.6.4 -- MXnet 1.3.0 -- The Required Disk Space: 35Gb +- MXNet 1.3.0 Review comment: Thanks for fixing CamelCase This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on a change in pull request #14179: Fix READMEs for examples
ankkhedia commented on a change in pull request #14179: Fix READMEs for examples URL: https://github.com/apache/incubator-mxnet/pull/14179#discussion_r257361952 ## File path: example/gluon/audio/README.md ## @@ -0,0 +1,100 @@ +# Urban Sounds Classification in MXNet Gluon + +This example provides an end-to-end pipeline for a common datahack competition - Urban Sounds Classification Example. +Below is the link to the competition: +https://datahack.analyticsvidhya.com/contest/practice-problem-urban-sound-classification/ + +After logging in, the data set can be downloaded. +The details of the dataset and the link to download it are given below: + + +## Urban Sounds Dataset: +### Description + The dataset contains 8732 wav files which are audio samples(<= 4s)) of street sounds like engine_idling, car_horn, children_playing, dog_barking and so on. + The task is to classify these audio samples into one of the following 10 labels: + ``` + siren, + street_music, + drilling, + dog_bark, + children_playing, + gun_shot, + engine_idling, + air_conditioner, + jackhammer, + car_horn + ``` + +To be able to run this example: + +1. `pip install -r requirements.txt` + +If you are in the directory where the requirements.txt file lies, +this step installs the required libraries to run the example. +The main dependency that is required is: Librosa. +The version used to test the example is: `0.6.2` +For more details, refer here: +https://librosa.github.io/librosa/install.html + +2. Download the dataset(train.zip, test.zip) required for this example from the location: +https://drive.google.com/drive/folders/0By0bAi7hOBAFUHVXd1JCN3MwTEU + Review comment: Should this dataset be added to the some mxnet related S3 bucket rather than a private google drive? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zachgk commented on a change in pull request #14169: [MXNET-1305] Scala interpreter instructions
zachgk commented on a change in pull request #14169: [MXNET-1305] Scala interpreter instructions URL: https://github.com/apache/incubator-mxnet/pull/14169#discussion_r257360367 ## File path: docs/install/scala_setup.md ## @@ -116,6 +117,33 @@ mvn install +## Interpreter + +To run the scala interpreter, first download and install scala 2.11.x (run `scala -version` to make sure you have the right version installed.** + +### Installing the Interpreter + +**Ubuntu*** + +``` +sudo apt-get install scala Review comment: It should be fine. Scala uses a slightly different versioning scheme (https://docs.scala-lang.org/overviews/core/binary-compatibility-for-library-authors.html) so with version x.y.z, changes to either x or y can be breaking but changes to z are safe. Apt-get installs 2.11.x on both 16.04 and 18.04 so I think it is better to give the simpler command since it is valid. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14053: in-place reshape ops
ankkhedia commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-464166842 @mxnet-label-bot remove [pr-awaiting-review] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14053: in-place reshape ops
ankkhedia commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-464166671 @mxnet-label-bot add [pr-work-in-progress] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha commented on issue #14053: in-place reshape ops
szha commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-464166096 I haven't finished this PR yet as I haven't got around to add the in-place option. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14169: [MXNET-1305] Scala interpreter instructions
ankkhedia commented on issue #14169: [MXNET-1305] Scala interpreter instructions URL: https://github.com/apache/incubator-mxnet/pull/14169#issuecomment-464163054 @zachgk Thanks for the cotribution! @mxnet-label-bot add [pr-awaiting-review, Scala] @lanking520 @piyushghai @andrewfayres Could you please help review this PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14170: add preprocessed data and pretrained model info
ankkhedia commented on issue #14170: add preprocessed data and pretrained model info URL: https://github.com/apache/incubator-mxnet/pull/14170#issuecomment-464162527 @aaronmarkham Thanks for the contribution! @mxnet-label-bot add [pr-awaiting-review, Example] @ThomasDelteil Could you please help review this PR? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14176: uncomment memory pool free
ankkhedia commented on issue #14176: uncomment memory pool free URL: https://github.com/apache/incubator-mxnet/pull/14176#issuecomment-464160871 @leleamol Ping for review! This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] perdasilva closed pull request #14178: [WIP] Testing the commit status setter
perdasilva closed pull request #14178: [WIP] Testing the commit status setter URL: https://github.com/apache/incubator-mxnet/pull/14178 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ankkhedia commented on issue #14176: uncomment memory pool free
ankkhedia commented on issue #14176: uncomment memory pool free URL: https://github.com/apache/incubator-mxnet/pull/14176#issuecomment-464159789 @arcadiaphy Thanks for the contribution! @mxnet-label-bot add [pr-awaiting-review, C API] This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] yuxihu edited a comment on issue #14158: [Flaky Test] Master Python2: MKLDNN-GPU test_operator_gpu.test_multinomial_generator
yuxihu edited a comment on issue #14158: [Flaky Test] Master Python2: MKLDNN-GPU test_operator_gpu.test_multinomial_generator URL: https://github.com/apache/incubator-mxnet/issues/14158#issuecomment-464158455 The test I added (test_dataloader_context) in #14136 is not part of GPU CI. I do not see how it causes the flakiness for the test mentioned here. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services