V1.8.x #19262
V1.8.x #19262
Conversation
* add zero grad for npi_unique (apache#18080) * fix np.clip scalar input case (apache#17788) * fix true_divide (apache#18393) Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com>
* Fix Windows GPU CI (apache#17962) Update Windows CI to use VS 2019 and enable x64 bit toolchain. Previously we are using an older 32 bit toolchain causing OOM errors during linking. Switching to x64 bit toolchain on the older VS version previously used by the CI was attempted in apache#17912 and did not work. Update to Cuda 10.2 as it is required by VS 2019. Switch to ninja-build on Windows to speed up build as ninja-build is now preinstalled. Remove logic to install cmake 3.16 on every PR as cmake 3.17 is now preinstalled. Add build retrials due to cuda thrust + VS2019 flakyness. Co-authored-by: vexilligera <vexilligera@gmail.com> * backport mixed type Co-authored-by: Leonard Lausen <lausen@amazon.com> Co-authored-by: vexilligera <vexilligera@gmail.com>
… variable input shapes (apache#18632) (apache#18703) * Fix the monitor_callback invalid issue during calibration with variable input shapes * retrigger CI * Add UT for monitor check and disable codecov Co-authored-by: Tao Lv <tao.a.lv@intel.com>
leezu
requested review from
aaronmarkham,
anirudh2290,
eric-haibin-lin,
marcoabreu and
szha
as code owners
|
Hey @leezu , Thanks for submitting the PR
CI supported jobs: [website, centos-cpu, edge, miscellaneous, windows-cpu, unix-gpu, clang, sanity, unix-cpu, centos-gpu, windows-gpu] Note: |
|
Also CC @samskalicky @sandeep-krishnamurthy |
|
Thank you so much @leezu |
|
In fact, should we ensure that each release needs to include all the commits of the previous release? @samskalicky @sandeep-krishnamurthy |
I like the idea of comparing to the previous release to ensure no PRs were missing (backwards comparing all previous releases seems to be a bit too much though). But theres already a lot of work required by the release manager: https://cwiki.apache.org/confluence/display/MXNET/Release+Process . Its already a significant time commitment. Maybe we should consider adding the instructions for committers/release-managers to verify that commits to release branches (ie. 1.7.x or 1.8.x) are just cherry-pick/porting PRs from the base branch (ie. 1.x) to avoid the problem in the first place. |
|
@samskalicky I agree. We should revise the guideline for cherry-picking the commits. That means, we need to both cherry-pick to the specific branch and also to 1.x / 2.x in the future. |
* Update to thrust 1.9.8 on Windows * Remove debug logic
Updating thrust alone did not help. Similar issues (though less often) still occur with updated thrust, and also with nvidia cub. Tracked upstream at NVIDIA/thrust#1090
|
@mxnet-bot run ci [unix-gpu] |
|
Jenkins CI successfully triggered : [unix-gpu] |
|
@mxnet-bot run ci [unix-gpu] |
|
Jenkins CI successfully triggered : [unix-gpu] |
|
@leezu how do you wanna handle this, should we merge this PR so we have a single commit we can cherry-pick to v1.x? Or do you wanna open a separate PR there first before merging this one? Just want to make sure we dont merge this into a feature branch without having it in the main v1.x branch again... And what reviews do you need before merging this PR? |
* * Fix einsum gradient (apache#18482) * [v1.7.x] Backport PRs of numpy features (apache#18653) * add zero grad for npi_unique (apache#18080) * fix np.clip scalar input case (apache#17788) * fix true_divide (apache#18393) Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com> * [v1.7.x] backport mixed type binary ops to v1.7.x (apache#18649) * Fix Windows GPU CI (apache#17962) Update Windows CI to use VS 2019 and enable x64 bit toolchain. Previously we are using an older 32 bit toolchain causing OOM errors during linking. Switching to x64 bit toolchain on the older VS version previously used by the CI was attempted in apache#17912 and did not work. Update to Cuda 10.2 as it is required by VS 2019. Switch to ninja-build on Windows to speed up build as ninja-build is now preinstalled. Remove logic to install cmake 3.16 on every PR as cmake 3.17 is now preinstalled. Add build retrials due to cuda thrust + VS2019 flakyness. Co-authored-by: vexilligera <vexilligera@gmail.com> * backport mixed type Co-authored-by: Leonard Lausen <lausen@amazon.com> Co-authored-by: vexilligera <vexilligera@gmail.com> * revise activations (apache#18700) * [v1.6] Fix the monitor_callback invalid issue during calibration with variable input shapes (apache#18632) (apache#18703) * Fix the monitor_callback invalid issue during calibration with variable input shapes * retrigger CI * Add UT for monitor check and disable codecov Co-authored-by: Tao Lv <tao.a.lv@intel.com> * Fail build_windows.py if all retries failed (apache#18177) * Update to thrust 1.9.8 on Windows (apache#18218) * Update to thrust 1.9.8 on Windows * Remove debug logic * Re-enable build retries on MSVC (apache#18230) Updating thrust alone did not help. Similar issues (though less often) still occur with updated thrust, and also with nvidia cub. Tracked upstream at NVIDIA/thrust#1090 Co-authored-by: Ke Han <38852697+hanke580@users.noreply.github.com> Co-authored-by: Xingjian Shi <xshiab@connect.ust.hk> Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com> Co-authored-by: Yijun Chen <chenyijun0902@gmail.com> Co-authored-by: vexilligera <vexilligera@gmail.com> Co-authored-by: ciyong <ciyong.chen@intel.com> Co-authored-by: Tao Lv <tao.a.lv@intel.com>
* * Fix einsum gradient (#18482) * [v1.7.x] Backport PRs of numpy features (#18653) * add zero grad for npi_unique (#18080) * fix np.clip scalar input case (#17788) * fix true_divide (#18393) Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com> * [v1.7.x] backport mixed type binary ops to v1.7.x (#18649) * Fix Windows GPU CI (#17962) Update Windows CI to use VS 2019 and enable x64 bit toolchain. Previously we are using an older 32 bit toolchain causing OOM errors during linking. Switching to x64 bit toolchain on the older VS version previously used by the CI was attempted in #17912 and did not work. Update to Cuda 10.2 as it is required by VS 2019. Switch to ninja-build on Windows to speed up build as ninja-build is now preinstalled. Remove logic to install cmake 3.16 on every PR as cmake 3.17 is now preinstalled. Add build retrials due to cuda thrust + VS2019 flakyness. Co-authored-by: vexilligera <vexilligera@gmail.com> * backport mixed type Co-authored-by: Leonard Lausen <lausen@amazon.com> Co-authored-by: vexilligera <vexilligera@gmail.com> * revise activations (#18700) * [v1.6] Fix the monitor_callback invalid issue during calibration with variable input shapes (#18632) (#18703) * Fix the monitor_callback invalid issue during calibration with variable input shapes * retrigger CI * Add UT for monitor check and disable codecov Co-authored-by: Tao Lv <tao.a.lv@intel.com> * Fail build_windows.py if all retries failed (#18177) * Update to thrust 1.9.8 on Windows (#18218) * Update to thrust 1.9.8 on Windows * Remove debug logic * Re-enable build retries on MSVC (#18230) Updating thrust alone did not help. Similar issues (though less often) still occur with updated thrust, and also with nvidia cub. Tracked upstream at NVIDIA/thrust#1090 Co-authored-by: Ke Han <38852697+hanke580@users.noreply.github.com> Co-authored-by: Xingjian Shi <xshiab@connect.ust.hk> Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com> Co-authored-by: Yijun Chen <chenyijun0902@gmail.com> Co-authored-by: vexilligera <vexilligera@gmail.com> Co-authored-by: ciyong <ciyong.chen@intel.com> Co-authored-by: Tao Lv <tao.a.lv@intel.com> Co-authored-by: Leonard Lausen <lausen@amazon.com> Co-authored-by: Ke Han <38852697+hanke580@users.noreply.github.com> Co-authored-by: Xingjian Shi <xshiab@connect.ust.hk> Co-authored-by: Hao Jin <hjjn.amzn@gmail.com> Co-authored-by: Xi Wang <xidulu@gmail.com> Co-authored-by: Yijun Chen <chenyijun0902@gmail.com> Co-authored-by: vexilligera <vexilligera@gmail.com> Co-authored-by: ciyong <ciyong.chen@intel.com> Co-authored-by: Tao Lv <tao.a.lv@intel.com>
Use magit-cherry mode to check for commits present in v1.7.x but missing from v1.8.x.
-indicates that the commit is also present in v1.8.x, whereas+indicates that the commit is missing.There are a couple of false positives (declared missing but actually present), as those commits were forward ported to v1.8.x in a squashed form. I went through all
+marked commits and applied them to thev1.8.xbranch.Only conflict to resolve was 8c7c2f1 which had to take into account the change by ce0a518