profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/czxttkl/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.

bmaddenneu/cs4500-admin-portal-tool 1

admin portal tool backend project for cs4500

czxttkl/ARO 1

Open Source - Application Resource Optimizer (ARO)

czxttkl/czxttkl.github.io 1

http://czxttkl.github.io

czxttkl/GAE 1

Game Avatar Embedding

czxttkl/Tutorials 1

Some tutorials I experimented

fork wusuopubupt/nebula_doc

nebula 文档, gitbook版

fork in 3 hours

pull request commentfacebookresearch/ReAgent

Migrate Seq2Reward Compress Model to PyTorch Lightning

This pull request was exported from Phabricator. Differential Revision: D27663810

gji1

comment created time in 6 hours

pull request commentfacebookresearch/ReAgent

Migrate Seq2Reward Compress Model to PyTorch Lightning

This pull request was exported from Phabricator. Differential Revision: D27663810

gji1

comment created time in 10 hours

pull request commentfacebookresearch/ReAgent

Clone dqn reporter logging on tensorboard

This pull request was exported from Phabricator. Differential Revision: D27694627

MisterTea

comment created time in 11 hours

pull request commentfacebookresearch/ReAgent

Move batch_size of trainer_param to mini_batchsize of reader_options

This pull request was exported from Phabricator. Differential Revision: D27720626

gji1

comment created time in 11 hours

PR opened facebookresearch/ReAgent

Move batch_size of trainer_param to mini_batchsize of reader_options

Summary: Move the batch_size for training seq2reward trainer_param to reader_options.

Differential Revision: D27720626

+0 -3

0 comment

3 changed files

pr created time in 11 hours

push eventfacebookresearch/ReAgent

Fei Jia

commit sha 67434f458cde1f2c946237e866a73392279a7ede

Add binary-cross-entropy-with-logits loss for myopic values Summary: Adding binary-cross-entropy-with-logits loss for myopic values between 0 and 1. Reviewed By: czxttkl Differential Revision: D27712539 fbshipit-source-id: f9e5fa67cee9955d191712a4c472968086e94c91

view details

push time in 14 hours

pull request commentfacebookresearch/ReAgent

Move reagent/workflow/data -> reagent/data and reagent/data/fb

This pull request was exported from Phabricator. Differential Revision: D27692807

MisterTea

comment created time in 15 hours

pull request commentfacebookresearch/ReAgent

Clone dqn reporter logging on tensorboard

This pull request was exported from Phabricator. Differential Revision: D27694627

MisterTea

comment created time in 16 hours

PR opened facebookresearch/ReAgent

Clone dqn reporter logging on tensorboard

Summary: This diff uses the logger in pytorch lightning to recreate the graphs that were traditionally reported through the dqn_reporter. These graphs are then fed back into fblearner, eliiminating the need to report them manually.

Differential Revision: D27694627

+189 -16

0 comment

7 changed files

pr created time in 16 hours

pull request commentfacebookresearch/ReAgent

Move reagent/workflow/data -> reagent/data and reagent/data/fb

This pull request was exported from Phabricator. Differential Revision: D27692807

MisterTea

comment created time in a day

PR opened facebookresearch/ReAgent

Move reagent/workflow/data -> reagent/data and reagent/data/fb

Differential Revision: D27692807

+10 -18

0 comment

12 changed files

pr created time in a day

startedbenhoyt/mugo

started time in a day

fork veggiec/Tiger-AppStore

太阁X项目四期9组-AppStore项目

fork in 3 days

pull request commentfacebookresearch/ReAgent

move use_gpu out of model manager

Codecov Report

Merging #447 (61c647f) into master (cebd822) will increase coverage by 0.05%. The diff coverage is 96.77%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #447      +/-   ##
==========================================
+ Coverage   82.35%   82.41%   +0.05%     
==========================================
  Files         312      312              
  Lines       17818    17814       -4     
  Branches       42       42              
==========================================
+ Hits        14674    14681       +7     
+ Misses       3120     3109      -11     
  Partials       24       24              
Impacted Files Coverage Δ
reagent/model_managers/slate_q_base.py 76.38% <ø> (ø)
reagent/workflow/training.py 85.04% <ø> (ø)
reagent/training/dqn_trainer.py 85.84% <66.66%> (ø)
reagent/model_managers/actor_critic/sac.py 100.00% <100.00%> (ø)
reagent/model_managers/actor_critic/td3.py 97.43% <100.00%> (ø)
reagent/model_managers/actor_critic_base.py 94.59% <100.00%> (ø)
reagent/model_managers/discrete/discrete_c51dqn.py 100.00% <100.00%> (ø)
reagent/model_managers/discrete/discrete_crr.py 81.92% <100.00%> (ø)
reagent/model_managers/discrete/discrete_dqn.py 97.72% <100.00%> (ø)
reagent/model_managers/discrete/discrete_qrdqn.py 87.50% <100.00%> (ø)
... and 16 more

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update cebd822...61c647f. Read the comment docs.

MisterTea

comment created time in 4 days

pull request commentfacebookresearch/ReAgent

move use_gpu out of model manager

This pull request was exported from Phabricator. Differential Revision: D26627900

MisterTea

comment created time in 4 days

pull request commentfacebookresearch/ReAgent

move use_gpu out of model manager

This pull request was exported from Phabricator. Differential Revision: D26627900

MisterTea

comment created time in 4 days

pull request commentfacebookresearch/ReAgent

Simplify model manager unions

This pull request has been merged in facebookresearch/ReAgent@cebd8224cf448b1dd6b3e7e067f6b4bf54ef1823.

MisterTea

comment created time in 4 days

push eventfacebookresearch/ReAgent

Jason Gauci

commit sha cebd8224cf448b1dd6b3e7e067f6b4bf54ef1823

Simplify model manager unions (#444) Summary: Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/444 Reviewed By: kaiwenw Differential Revision: D27614614 fbshipit-source-id: ce5de96de5714eab80c1e3c6c78100663426ff66

view details

push time in 4 days

PR closed facebookresearch/ReAgent

Simplify model manager unions cla signed fb-exported

Differential Revision: D27614614

+127 -54

3 comments

8 changed files

MisterTea

pr closed time in 4 days

startednsekhar/serialcheck

started time in 4 days

pull request commentfacebookresearch/ReAgent

Train Reels LearnedVM as a residual boost

This pull request was exported from Phabricator. Differential Revision: D27264221

kaiwenw

comment created time in 4 days

PR opened facebookresearch/ReAgent

Train Reels LearnedVM as a residual boost

Summary:

  • add option to train as residual boost (on top of prod vm score)
  • net builder for MLP for better configuration of MLPScorer
  • filter out slates with 0 scores (ostensibly from precision problems); these caused nan problems in training
  • add option for orthogonal weight initialization

Reviewed By: czxttkl

Differential Revision: D27264221

+159 -76

0 comment

5 changed files

pr created time in 4 days

pull request commentfacebookresearch/ReAgent

towards RLwC

This pull request was exported from Phabricator. Differential Revision: D27495823

kaiwenw

comment created time in 4 days

pull request commentfacebookresearch/ReAgent

Migrate Seq2Reward Compress Model to PyTorch Lightning

This pull request was exported from Phabricator. Differential Revision: D27663810

gji1

comment created time in 4 days

PR opened facebookresearch/ReAgent

Migrate Seq2Reward Compress Model to PyTorch Lightning

Differential Revision: D27663810

+108 -88

0 comment

3 changed files

pr created time in 4 days

pull request commentfacebookresearch/ReAgent

Correct eval_td_loss graph for DQN

This pull request has been merged in facebookresearch/ReAgent@8f6ffe8fee177e6cb3e25d6c76eb510432bc1d21.

DavidV17

comment created time in 5 days

push eventfacebookresearch/ReAgent

David Vengerov

commit sha 8f6ffe8fee177e6cb3e25d6c76eb510432bc1d21

Correct eval_td_loss graph for DQN (#446) Summary: Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/446 Switch eval_td_loss to Tensorboard Reviewed By: bankawas Differential Revision: D27643487 fbshipit-source-id: 25c0af8f0d943abaa68b024fd2f61caf65445cd9

view details

push time in 5 days

PR closed facebookresearch/ReAgent

Correct eval_td_loss graph for DQN cla signed fb-exported

Summary: Switch eval_td_loss to Tensorboard

Differential Revision: D27643487

+2 -2

1 comment

2 changed files

DavidV17

pr closed time in 5 days

pull request commentfacebookresearch/ReAgent

move use_gpu out of model manager

This pull request was exported from Phabricator. Differential Revision: D26627900

MisterTea

comment created time in 5 days