Update mkldnn window build instructions in MKLDNN_README.md #2

yinghu5 · 2019-05-14T07:13:20Z

Description

Updates mkldnn & MKL window cmake build instructions in MKLDNN_README.md, if the PR apache#14877 is merged

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
Changes are complete (i.e. I finished coding on this PR)
All changes have test coverage:
Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
Code is well-documented:
For user-facing API changes, API doc string has been updated.
For new C++ functions in header files, their functionalities and arguments are documented.
For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$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

The PR apache#14570 has introduced a bug where the "stash name" is not matching: - It is **packed** as `ubuntu_cpu_int64` - It is **unpacked** as `cpu_int64`

* [clojure][generator] add random namespace generation * `ndarray_random_api` * `symbol_random_api` * fix tests

* Fix cached_op * try to fix ci * Fix CI * Fix ci

* moved file * Empty Commit * Retrigger

* add fix in the code * add unit test * update comments

* Refactor requantize * fix ci * Fix CI * Fix ci

* Enable zero-copy from numpy to MXNet NDArray * should work * make lint happy * fix stupid typos * wip to address comments * Address comments * Address comments * Remove redundant code

…#14599) * Use USE_SIGNAL_HANDLER by default set to ON in CMakeLists.txt * Fix lint error with NULL vs nullptr

* Add unpooled gpu memory type * Adding missing header * undo bad rebase change

…pache#14806) This reverts commit c437d5b.

…nalg (apache#14501)

…pache#14597)

* Clojure BERT example: minor code cleanup * Remove unused requires * Remove unused vars & function * Use `io` alias * Clojure BERT example: whitespace fix * Clojure BERT example: allow running with GPU The `infer` function accepts a CPU/GPU context, which the command line version of this example exposes as a `:cpu`/`:gpu` keyword. Previously, these options were ignored and the context was overridden to the default context (CPU). This commit allows users (both REPL and shell) to pass in a GPU context.

* add quantized sum * fix gpu compiler error and cpu testcase fail * add default forward function for quantized_sum * skip quantized_sum for gpu ctx * fix comments * fix indetation and comments * retrigger CI * alloc memeory through TmpMemMgr * fix comments Apr.12 * change sum to elemwise_add * change Sum to ElemwiseAdd * fix indents * retrigger CI * trigger CI * fix indentation and typo * trigger CI * fix typo * fix typo * remove USE_MKLDNN macro for requantize params * rename param same as its op * trigger CI * trigger CI * trigger CI

* add resnet18 and mobilenetv2 models * add readme * support mkldnn s8s8 goihw16g weight format * fix_readme_typo

* upgrade pylint to latest version & change the Makefile to use python3 pylint * fix pylint when upgrade * specify the pylint version * suppress the pylint not-callable bug * remove pylint from python2 * fix the typo * upgrade cpp-package pylint * move pylint disable to each block

…the code (apache#14840) * doc fix * trigger ci

* Updates Ubuntu GPU CI image base image to cuda10-devel and manually installs cuDNN version 7.3.1.20 * Updates CentOS 7 GPU CI image base image to cuda10-devel and manually installs cuDNN version 7.3.1.20

* remove unneeded test files * add test files to gitignore

…apache#14868) This reverts commit 369b66d.

Fix the typo.

* Fix sample_multinomial number of outputs bug * Fix lint

* modified: docs/api/python/gluon/contrib.md modified: python/mxnet/gluon/contrib/__init__.py new file: python/mxnet/gluon/contrib/cnn/__init__.py new file: python/mxnet/gluon/contrib/cnn/conv_layers.py new file: tests/python/gpu/test_gluon_contrib_gpu.py * modified: python/mxnet/gluon/contrib/cnn/conv_layers.py * modified: python/mxnet/gluon/contrib/cnn/conv_layers.py * Update conv_layers.py * Update conv_layers.py

* cpu optimized data loader * Fix CI * Fix CI * Fix ci * Fix doc

update MKL_ROOT

fix nomkldnn issue.

change MKL_ROOT

windows md5

export MKLROOT

powershell 'cd $Env:MKL_ROOT'

fix windows build instruction

edisongustavo and others added 30 commits April 26, 2019 17:18

Use correct stash name when running nightly tests (apache#14809)

680bade

The PR apache#14570 has introduced a bug where the "stash name" is not matching: - It is **packed** as `ubuntu_cpu_int64` - It is **unpacked** as `cpu_int64`

[clojure][generator] ndarray/symbol api random merged (apache#14800)

5dd9fa2

* [clojure][generator] add random namespace generation * `ndarray_random_api` * `symbol_random_api` * fix tests

Improve cached_op performance for static mode (apache#14785)

369b66d

* Fix cached_op * try to fix ci * Fix CI * Fix ci

data preparation file moved in example (apache#14781)

40e3d7c

* moved file * Empty Commit * Retrigger

Scala/Java Predict API fix apache#14756 (apache#14804)

6cbc273

* add fix in the code * add unit test * update comments

clean up submodule (apache#14645)

003800c

[MKLDNN]Refactor requantize to speed up execution (apache#14608)

6c60025

* Refactor requantize * fix ci * Fix CI * Fix ci

[MXNET-1398] Enable zero-copy from numpy to MXNet NDArray (apache#14733)

c18381d

* Enable zero-copy from numpy to MXNet NDArray * should work * make lint happy * fix stupid typos * wip to address comments * Address comments * Address comments * Remove redundant code

Speed up SequenceReverse (apache#14627)

64287dd

Fixes for wine detection tutorial (apache#13886)

1af29e9

Use USE_SIGNAL_HANDLER by default set to ON in CMakeLists.txt (apache…

3a46980

…#14599) * Use USE_SIGNAL_HANDLER by default set to ON in CMakeLists.txt * Fix lint error with NULL vs nullptr

Add unpooled gpu memory type (apache#14716)

07aef13

* Add unpooled gpu memory type * Adding missing header * undo bad rebase change

Revert "use mkl sparse matrix to improve performance (apache#14492)" (a…

1238aa0

…pache#14806) This reverts commit c437d5b.

added extraction/generation of diagonal and triangonal matrices to li…

c7577e5

…nalg (apache#14501)

add clojure tutorials to index (apache#14814)

5fda0a5

Fix iterator over symbol when multiple children have the same name (a…

488fad2

…pache#14597)

[int8] Add MobileNetV2_1.0 & ResNet18 Quantization (apache#14823)

bde1b84

* add resnet18 and mobilenetv2 models * add readme * support mkldnn s8s8 goihw16g weight format * fix_readme_typo

reformat trt to use subgraph API, add fp16 support (apache#14040)

1c874cf

Make docblocks for Gluon BatchNorm and SyncBatchNorm consistent with …

977e558

…the code (apache#14840) * doc fix * trigger ci

Print reproduction command on CI failure (apache#14815)

381a9da

[MXNET-13578] Fix cmake installation failed (apache#14692)

e17b7e2

Update base CUDA image for CI to v10.0 cuDNN 7.3.1 (apache#14513)

36c3306

* Updates Ubuntu GPU CI image base image to cuda10-devel and manually installs cuDNN version 7.3.1.20 * Updates CentOS 7 GPU CI image base image to cuda10-devel and manually installs cuDNN version 7.3.1.20

[Clojure] Remove unneeded test files (apache#14813)

1540a84

* remove unneeded test files * add test files to gitignore

[DEV] update code owner (apache#14862)

372f531

Revert "Improve cached_op performance for static mode (apache#14785)" (…

204f3f2

…apache#14868) This reverts commit 369b66d.

Update lstm_crf.py (apache#14865)

d09f68a

Fix the typo.

Fix sample_multinomial number of outputs bug (apache#14873)

5ba285b

* Fix sample_multinomial number of outputs bug * Fix lint

suyz526 and others added 17 commits May 5, 2019 09:51

Refactor ImageRecordIter (apache#14824)

621b391

* cpu optimized data loader * Fix CI * Fix CI * Fix ci * Fix doc

fix cmake for win and mkl

634bfea

Update build_windows.py

2f8adc0

update MKL_ROOT

Update Jenkins_steps.groovy

90eb723

fix nomkldnn issue.

Update build_windows.py

44ef104

change MKL_ROOT

msshadow_blas

b4c8ddd

update mshasow

b22ce7d

Update DownloadMKLML.cmake

5b2a7cb

windows md5

Update DownloadMKLML.cmake

d9bd487

windows md5

Update Jenkins_steps.groovy

8e6b4ad

export MKLROOT

Update Jenkins_steps.groovy

4e54ca8

powershell 'cd $Env:MKL_ROOT'

Update MKLDNN_README.md

46bf7e4

fix windows build instruction

Update MKLDNN_README.md

7695c45

Update MKLDNN_README.md

f14b582

Update MKLDNN_README.md

0300b5d

Update MKLDNN_README.md

ea307b2

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update mkldnn window build instructions in MKLDNN_README.md #2

Update mkldnn window build instructions in MKLDNN_README.md #2

yinghu5 commented May 14, 2019

Update mkldnn window build instructions in MKLDNN_README.md #2

Are you sure you want to change the base?

Update mkldnn window build instructions in MKLDNN_README.md #2

Conversation

yinghu5 commented May 14, 2019

Description

Checklist

Essentials

Changes

Comments