Code modification for testcases of various network models in directory example #12498

luobao-intel · 2018-09-10T12:22:17Z

Description

In the example directory, there are some test cases that exits some small bugs and missed options running on CPU. Thie pull request is to add the CPU options and fix some small bugs.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

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

add CPU option

small bugs fix

rcnn
cnn_visualization

file location change

module

Comments

If this change is a backward incompatible change, why must this change be made.
Interesting edge cases to note here

@pengzhao-intel @TaoLv @juliusshufan

kalyc · 2018-09-14T17:41:15Z

@luobao-intel thanks for your contribution. Could you update the PR title to be inline with what exactly this is fixing?

@mxnet-label-bot[pr-awaiting-testing]

vandanavk · 2018-09-25T17:25:52Z

@luobao-intel Could you check the build failure so that we can proceed with the review

@ThomasDelteil @safrooze for review

vrakesh · 2018-10-09T06:49:24Z

@luobao-intel Requesting an update on the PR, any progress on the build failures?

sandeep-krishnamurthy

It is more confusing and not very intuitive for the user to say, gpus= any -ve number for CPU and any +ve number for GPU.

Maybe you can consider having a parameter like "gpus" where gpus=None by default which means run on CPU. gpus=0 to use gpu(0), gpus=0,1,2,3 to use gpu(0), gpu(1), gpu(2), gpu(3).
What do you think?

luobao-intel · 2018-10-18T08:14:19Z

@sandeep-krishnamurthy The code has been modified according to your requirements. Can you review it again? Thank you!

sandeep-krishnamurthy

Thanks for the changes.

sandeep-krishnamurthy · 2018-10-21T19:12:05Z

example/bayesian-methods/bdk_demo.py

 X_test=X_test, Y_test=Y_test,
 total_iter_num=1000000,
 initializer=initializer,
 learning_rate=4E-6, prior_precision=1.0, minibatch_size=100,
 thin_interval=100, burn_in_iter_num=1000)


-def run_mnist_DistilledSGLD(training_num=50000):
+def run_mnist_DistilledSGLD(training_num=50000, xpu=0):


This means, by default, mx.gpu(0). Should it default to CPU instead by making xpu=None?

ankkhedia · 2018-10-30T18:47:13Z

@luobao-intel
Thanks for updating the PR. There are few minor review comments. Could you please take a look?

kalyc · 2018-11-12T23:13:29Z

ping! @luobao-intel could you please take a look?

pengzhao-intel · 2018-11-13T00:24:43Z

@kalyc @ankkhedia Sorry for the delay. @luobao-intel is our summer intern and he is back to school now.
@huangzhiyuan will take over this PR :)

juliusshufan · 2018-11-13T05:09:11Z

@pengxin99

juliusshufan · 2018-11-15T14:04:44Z

@sandeep-krishnamurthy Could you take a look if your comments have been addressed with latest changes? Thanks.

pengxin99 · 2018-12-03T03:39:05Z

@vandanavk @sandeep-krishnamurthy I have change the code reference to the review comments, could you please make a review again? Thanks

vandanavk · 2018-12-03T03:55:25Z

LGTM

roywei · 2018-12-14T17:50:09Z

@sandeep-krishnamurthy could you take a look again? Thanks!

roywei

LGTM

pengxin99 · 2018-12-16T03:16:58Z

@vandanavk @roywei @sandeep-krishnamurthy Thanks for your reviews, and could this be merged?

pengzhao-intel · 2018-12-19T01:39:39Z

@sandeep-krishnamurthy could you help take a look again? Thanks.

TaoLv · 2018-12-24T15:09:42Z

example/bayesian-methods/bdk_demo.py

@@ -156,42 +156,42 @@ def get_toy_sym(teacher=True, teacher_noise_precision=None):
 return net


-def dev():
- return mx.gpu()
+def dev(xpu = None):


No need to have white spaces before and after =. Also, xpu is a little confusing for me as its value should be either gpu or cpu according to the name. But here in this case it equals to gpu numbers.

TaoLv · 2018-12-24T15:11:29Z

example/rcnn/symdata/loader.py

@@ -129,7 +129,7 @@ def getpad(self):

 class AnchorLoader(mx.io.DataIter):
 def __init__(self, roidb, batch_size, short, max_size, mean, std,
- feat_sym, anchor_generator: AnchorGenerator, anchor_sampler: AnchorSampler,
+ feat_sym, anchor_generator=AnchorGenerator, anchor_sampler=AnchorSampler,


Can you explain?

The type of parameter anchor_generator must by Class AnchorGenerator, so i think it should be :, not = , i will do changes corresponding

TaoLv · 2018-12-24T15:12:40Z

example/rcnn/test.py

@@ -94,7 +94,7 @@ def parse_args():
 parser.add_argument('--params', type=str, default='', help='path to trained model')
 parser.add_argument('--dataset', type=str, default='voc', help='training dataset')
 parser.add_argument('--imageset', type=str, default='', help='imageset splits')
- parser.add_argument('--gpu', type=int, default=0, help='gpu device eg. 0')
+ parser.add_argument('--gpu', type=int, default=0, help='gpu device eg. 0 if None then use cpu')


if not provide, then use cpu.

TaoLv · 2018-12-24T15:13:24Z

example/rcnn/train.py

@@ -127,7 +127,7 @@ def parse_args():
 parser.add_argument('--pretrained', type=str, default='', help='path to pretrained model')
 parser.add_argument('--dataset', type=str, default='voc', help='training dataset')
 parser.add_argument('--imageset', type=str, default='', help='imageset splits')
- parser.add_argument('--gpus', type=str, default='0', help='gpu devices eg. 0,1')
+ parser.add_argument('--gpus', type=str, help='gpu devices eg. 0,1 if null then use cpu')


if not provide, then use cpu.

TaoLv · 2018-12-24T15:14:43Z

example/fcn-xs/fcn_xs.py

@@ -85,6 +85,7 @@ def main():
 help='the init type of fcn-xs model, e.g. vgg16, fcnxs')
 parser.add_argument('--retrain', action='store_true', default=False,
 help='true means continue training.')
+ parser.add_argument("--gpu", type=int, help="if None then use cpu else use gpu")


TaoLv · 2018-12-24T15:16:35Z

example/fcn-xs/fcn_xs.py


 def main():
+ ctx = mx.cpu() if not args.gpu else mx.gpu(0)


mx.gpu(args.gpu)? What's the value of args.gpu?

TaoLv

Minor comments. The rest looks good to me.

example/bayesian-methods/bdk_demo.py

TaoLv

Seems CI status is not reflected on the page. @pengxin99 Please re-trigger with an empty commit.

TaoLv · 2019-01-08T12:33:03Z

@sandeep-krishnamurthy Please check if your comments are addressed.

TaoLv · 2019-01-11T14:05:30Z

Thank you for the contribution @pengxin99. I think it's good to merge now.

* upstream/master: (109 commits) Code modification for testcases of various network models in directory example (apache#12498) [CI] Prevent timeouts when rebuilding containers with docker. (apache#13818) fix Makefile for rpkg (apache#13590) change to compile time (apache#13835) Disabled flaky test (apache#13758) Improve license_header tool by only traversing files under revision c… (apache#13803) Removes unneeded nvidia driver ppa installation (apache#13814) Add Local test stage and option to jump directly to menu item from commandline (apache#13809) Remove MXNET_STORAGE_FALLBACK_LOG_VERBOSE from test_autograd.py (apache#13830) Fix scala doc build break for v1.3.1 (apache#13820) [MXNET-1263] Unit Tests for Java Predictor and Object Detector APIs (apache#13794) [MXNET-1260] Float64 DType computation support in Scala/Java (apache#13678) onnx export ops (apache#13821) [MXNET-880] ONNX export: Random uniform, Random normal, MaxRoiPool (apache#13676) fix minor indentation (apache#13827) Fixing a symlink issue with R install (apache#13708) remove useless code (apache#13777) ONNX ops: norm exported and lpnormalization imported (apache#13806) Add new Maven build for Scala package (apache#13819) Dockerfiles for Publish Testing (apache#13707) ...

…y example (apache#12498) * example testcase modified * rcnn file add * license add * license init * CI test trigger * rcnn modify give up * trigger * modify for better user experience * change the default parameter to xpu=None * Update bdk_demo.py * Update fcn_xs.py * Update test.py * Update train.py * Update bdk_demo.py * Update bdk_demo.py * modify review comments * refine * modify Readmes according to the changed code. * finetune READMEs * re-trigger ci * re-trigger ci twice

example testcase modified

b44acbf

luobao-intel requested a review from szha as a code owner September 10, 2018 12:22

luobao-intel added 2 commits September 11, 2018 10:48

Merge branch 'master' of https://github.com/apache/incubator-mxnet

61afc8f

rcnn file add

39f2304

marcoabreu added the pr-awaiting-testing PR is reviewed and waiting CI build and test label Sep 14, 2018

luobao-intel changed the title ~~example testcase modified~~ Code modification for testcases of various network model in directory example Sep 20, 2018

luobao-intel changed the title ~~Code modification for testcases of various network model in directory example~~ Code modification for testcases of various network models in directory example Sep 20, 2018

luobao-intel added 2 commits September 21, 2018 15:26

license add

f579e22

license init

4bd6259

CI test trigger

23ef75f

luobao-intel added 3 commits October 9, 2018 15:05

rcnn modify give up

717a943

trigger

6bc4a82

Merge branch 'master' of https://github.com/apache/incubator-mxnet

15c7907

sandeep-krishnamurthy suggested changes Oct 10, 2018

View reviewed changes

sandeep-krishnamurthy added Example pr-work-in-progress PR is still work in progress and removed pr-awaiting-testing PR is reviewed and waiting CI build and test labels Oct 10, 2018

modify for better user experience

4d48036

sandeep-krishnamurthy reviewed Oct 21, 2018

View reviewed changes

pengxin99 added 2 commits November 15, 2018 10:02

change the default parameter to xpu=None

a3a4a41

Merge remote-tracking branch 'upstream/master'

cac0676

pengxin99 added 5 commits November 29, 2018 13:27

Update fcn_xs.py

e12170f

Update test.py

eb63d58

Update train.py

eda77a6

Update bdk_demo.py

e4c4153

Update bdk_demo.py

8aeba26

roywei approved these changes Dec 14, 2018

View reviewed changes

vandanavk approved these changes Dec 16, 2018

View reviewed changes

TaoLv reviewed Dec 24, 2018

View reviewed changes

modify review comments

e55058f

TaoLv reviewed Dec 26, 2018

View reviewed changes

example/bayesian-methods/bdk_demo.py Outdated Show resolved Hide resolved

example/bayesian-methods/bdk_demo.py Outdated Show resolved Hide resolved

pengxin99 added 3 commits December 27, 2018 11:16

refine

0c8803f

modify Readmes according to the changed code.

3447dab

finetune READMEs

0fbca75

TaoLv approved these changes Jan 8, 2019

View reviewed changes

TaoLv added pr-awaiting-review PR is waiting for code review and removed pr-work-in-progress PR is still work in progress labels Jan 8, 2019

pengxin99 added 2 commits January 9, 2019 08:46

re-trigger ci

dd27ee3

re-trigger ci twice

485f647

pengxin99 mentioned this pull request Jan 9, 2019

CI error with "raise child_exception" sphinx-doc/sphinx#5914

Closed

Merge remote-tracking branch 'upstream/master'

e9b138a

TaoLv merged commit a6ed619 into apache:master Jan 11, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Code modification for testcases of various network models in directory example #12498

Code modification for testcases of various network models in directory example #12498

luobao-intel commented Sep 10, 2018 •

edited

Loading

kalyc commented Sep 14, 2018

vandanavk commented Sep 25, 2018

vrakesh commented Oct 9, 2018

sandeep-krishnamurthy left a comment

luobao-intel commented Oct 18, 2018

sandeep-krishnamurthy left a comment

sandeep-krishnamurthy Oct 21, 2018

ankkhedia commented Oct 30, 2018

kalyc commented Nov 12, 2018

pengzhao-intel commented Nov 13, 2018

juliusshufan commented Nov 13, 2018

juliusshufan commented Nov 15, 2018

pengxin99 commented Dec 3, 2018

vandanavk commented Dec 3, 2018

roywei commented Dec 14, 2018

roywei left a comment

pengxin99 commented Dec 16, 2018

pengzhao-intel commented Dec 19, 2018

TaoLv Dec 24, 2018

TaoLv Dec 24, 2018

pengxin99 Dec 25, 2018

TaoLv Dec 24, 2018

TaoLv Dec 24, 2018

TaoLv Dec 24, 2018

TaoLv Dec 24, 2018

TaoLv left a comment

TaoLv left a comment

TaoLv commented Jan 8, 2019

TaoLv commented Jan 11, 2019

Code modification for testcases of various network models in directory example #12498

Code modification for testcases of various network models in directory example #12498

Conversation

luobao-intel commented Sep 10, 2018 • edited Loading

Description

Checklist

Essentials

Changes

add CPU option

small bugs fix

file location change

Comments

kalyc commented Sep 14, 2018

vandanavk commented Sep 25, 2018

vrakesh commented Oct 9, 2018

sandeep-krishnamurthy left a comment

Choose a reason for hiding this comment

luobao-intel commented Oct 18, 2018

sandeep-krishnamurthy left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ankkhedia commented Oct 30, 2018

kalyc commented Nov 12, 2018

pengzhao-intel commented Nov 13, 2018

juliusshufan commented Nov 13, 2018

juliusshufan commented Nov 15, 2018

pengxin99 commented Dec 3, 2018

vandanavk commented Dec 3, 2018

roywei commented Dec 14, 2018

roywei left a comment

Choose a reason for hiding this comment

pengxin99 commented Dec 16, 2018

pengzhao-intel commented Dec 19, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

TaoLv left a comment

Choose a reason for hiding this comment

TaoLv left a comment

Choose a reason for hiding this comment

TaoLv commented Jan 8, 2019

TaoLv commented Jan 11, 2019

luobao-intel commented Sep 10, 2018 •

edited

Loading