We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Feature freeze date: 16 Dec, 2019 Code freeze date: 10:30AM 24 Dec, 2019 Release Manager: @leckie-chn End Game Issue: is tracking in #1872
tests for supported NAS algorithms (Can nni offer some test metric for the example using one-shot/dart/enas model? #1598 ) @ultmaster (TBD)
cancel trials in UNKNOWN state (Feature request: UNKNOWN state - ability to cancel trial #1735 ) @chicm-ms (pending, code 1d)
fabric framework adoption @lvybriage (coding, eta. 20th, Dec)
BUG - correctly support NaN in metric data, JSON compliant (ValueError: Out of range float values are not JSON compliant #1589 ) @liuzhe-lz (code 1d) @chicm-ms to help profiling.
NaN
correctly recognize builtin tuner and customized tuner @chicm-ms Refactor dispatcher cmdline #1862 (testing on dev-refactor-dispatcher-cmdline)
dev-refactor-dispatcher-cmdline
logging in dispatcher_base @chicm-ms (bug fix after code freeze)
nnictl command auto complete @liuzhe-lz
refactor tuner working directory (doc work) @chicm-ms
properly catch tuner's error @liuzhe-lz
Fix local system as remote machine issue How to debug "no machine can be scheduled"? #1852 @chicm-ms PR Fix local system as remote machine issue #1852 #1856 submitted
/usr/bin/python3
gpu_metrics_collector.sh
The text was updated successfully, but these errors were encountered:
QuanluZhang
scarlett2018
No branches or pull requests
Feature freeze date: 16 Dec, 2019
Code freeze date: 10:30AM 24 Dec, 2019
Release Manager: @leckie-chn
End Game Issue: is tracking in #1872
Planned Features
[x] Proxyless NAS @QuanluZhang support proxylessnas with NNI NAS APIs #1863TensorFlow 2.0 support @liuzhe-lz (PR eta. Fri.)staticdynamic graph model compression (depend on Using NNI NAS interface with keras/tf.keras #1737 )more NAS algorithm support @ultmaster
C-DARTS (to be synced with partner)AutoDeepLab (TBD)more compression algorithm support @Cjkkkk @tanglang96
- [ ] separate PR for doc
- [x] BNN add quantization backward support #1854
*
TTQ (for next release)*
Trained quantization and weight sharing (TBD)Pass eslint ( #882 )
Other Features
tests for supported NAS algorithms (Can nni offer some test metric for the example using one-shot/dart/enas model? #1598 ) @ultmaster (TBD)cancel trials in UNKNOWN state (Feature request: UNKNOWN state - ability to cancel trial #1735 ) @chicm-ms (pending, code 1d)fabric framework adoption @lvybriage (coding, eta. 20th, Dec)BUG - correctly supportNaN
in metric data, JSON compliant (ValueError: Out of range float values are not JSON compliant #1589 ) @liuzhe-lz (code 1d) @chicm-ms to help profiling.correctly recognize builtin tuner and customized tuner @chicm-ms Refactor dispatcher cmdline #1862 (testing on
dev-refactor-dispatcher-cmdline
)logging in dispatcher_base @chicm-ms (bug fix after code freeze)
nnictl command auto complete @liuzhe-lz
refactor tuner working directory (doc work) @chicm-ms
properly catch tuner's error @liuzhe-lz
Fix local system as remote machine issue How to debug "no machine can be scheduled"? #1852 @chicm-ms PR Fix local system as remote machine issue #1852 #1856 submitted
Documentation
Small changes
remove hardcoded call to/usr/bin/python3
ingpu_metrics_collector.sh
( no machine can be scheduled, return TMP_NO_AVAILABLE_GPU #1712 ) @liuzhe-lzInternal feature
The text was updated successfully, but these errors were encountered: