profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/kittipatv/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.

kittipatv/recsim-no-tf 1

A Configurable Recommender Systems Simulation Platform

kittipatv/caffe2 0

Caffe2 is a lightweight, modular, and scalable deep learning framework.

kittipatv/dgl 0

Python package built to ease deep learning on graph, on top of existing DL frameworks.

kittipatv/Documentation 0

EOS.IO Documents

kittipatv/emacs-ipython-notebook 0

Jupyter notebook client in Emacs

kittipatv/governance 0

EOS Governance

kittipatv/nebula-graph 0

A distributed, fast open-source graph database featuring horizontal scalability and high availability

kittipatv/neo4j-graphql-js 0

A GraphQL to Cypher query execution layer for Neo4j and JavaScript GraphQL implementations.

kittipatv/ossci-job-dsl 0

Jenkins job definitions for OSSCI

pull request commentfacebookresearch/ReAgent

Fix discrete samplers

Hi @alexnikulkov, I created this PR on February 7, at that time this fix was not there.

wall-ed-coder

comment created time in 8 hours

pull request commentfacebookresearch/ReAgent

Fix discrete samplers

@MisterTea merged this pull request in facebookresearch/ReAgent@fd387a55d637fb4f0b16d39a47324ef64f9b3a1e.

wall-ed-coder

comment created time in 10 hours

push eventfacebookresearch/ReAgent

wall-ed-coder

commit sha fd387a55d637fb4f0b16d39a47324ef64f9b3a1e

Fix discrete samplers (#393) Summary: fix bugs: GreedyActionSampler returned one as a log prob and EpsilonGreedyActionSampler didn't work. Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/393 Test Plan: Imported from GitHub, without a `Test Plan:` line. ...but without running open source tests. Reviewed By: kaiwenw Differential Revision: D27842450 Pulled By: MisterTea fbshipit-source-id: 9b4aa85f352f2d7565473127b280d61bcc6d3b71

view details

push time in 10 hours

PR closed facebookresearch/ReAgent

Fix discrete samplers cla signed

fix bugs: GreedyActionSampler returned one as a log prob and EpsilonGreedyActionSampler didn't work.

+2 -2

3 comments

1 changed file

wall-ed-coder

pr closed time in 10 hours

pull request commentfacebookresearch/ReAgent

Make sure that trainer.train(batch) throws an error for Lightining trainers

This pull request has been merged in facebookresearch/ReAgent@d507a5703331e728d2dc7a2d90ba262e1554c54a.

alexnikulkov

comment created time in 13 hours

push eventfacebookresearch/ReAgent

Alex Nikulkov

commit sha d507a5703331e728d2dc7a2d90ba262e1554c54a

Make sure that trainer.train(batch) throws an error for Lightining trainers (#457) Summary: Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/457 trainer.train(batch) was the old, pre-Lightning ReAgent trainer API. With this diff we make sure that nobody is trying to call trainer.train(batch). trainer.train() or trainer.train(True/False) is allowed - this puts the network into training/eval mode. Reviewed By: MisterTea Differential Revision: D27862583 fbshipit-source-id: b0875e11cd4ef214c75fd1bef5b696f1cdf2b8d6

view details

push time in 13 hours

PR closed facebookresearch/ReAgent

Make sure that trainer.train(batch) throws an error for Lightining trainers cla signed fb-exported

Summary: trainer.train(batch) was the old, pre-Lightning ReAgent trainer API. With this diff we make sure that nobody is trying to call trainer.train(batch). trainer.train() or trainer.train(True/False) is allowed - this puts the network into training/eval mode.

Reviewed By: MisterTea

Differential Revision: D27862583

+11 -0

1 comment

1 changed file

alexnikulkov

pr closed time in 13 hours

pull request commentfacebookresearch/ReAgent

Oss test fixes jjg

This pull request was exported from Phabricator. Differential Revision: D27842449

MisterTea

comment created time in 15 hours

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 16 hours

pull request commentfacebookresearch/ReAgent

Fix discrete samplers

Hi @wall-ed-coder are you on master branch? We fixed the bug with the order of arguments in full_like() 2 months ago: https://github.com/facebookresearch/ReAgent/commit/5fd724322cd51bf268fa1e17263d66f74083356d

wall-ed-coder

comment created time in 16 hours

pull request commentfacebookresearch/ReAgent

Oss test fixes jjg

This pull request was exported from Phabricator. Differential Revision: D27842449

MisterTea

comment created time in 16 hours

pull request commentfacebookresearch/ReAgent

Make sure that trainer.train(batch) throws an error for Lightining trainers

This pull request was exported from Phabricator. Differential Revision: D27862583

alexnikulkov

comment created time in 18 hours

PR opened facebookresearch/ReAgent

Make sure that trainer.train(batch) throws an error for Lightining trainers

Summary: trainer.train(batch) was the old, pre-Lightning ReAgent trainer API. With this diff we make sure that nobody is trying to call trainer.train(batch). trainer.train() or trainer.train(True/False) is allowed - this puts the network into training/eval mode.

Reviewed By: MisterTea

Differential Revision: D27862583

+11 -0

0 comment

1 changed file

pr created time in 18 hours

issue openedfacebookresearch/ReAgent

CPE Functionaility

Hi,

I have been reading through the code and I have a few questions regarding the CPE functionality.

In particular, with the DQN model you have separate Q-networks, both normal and target, dedicated just for CPE that I would like to better understand. At present their purpose is not really clear to me. In particular, what is the purpose of these additional networks over and above the standard Q-networks of the DQN model?

In the function _calculate_cpes, which is part of the RLTrainer class. Reading this function it seems that update the networks q_network_cpe and q_network_cpe_target to model not only the reward, but any additional metrics that could be on interest in CPE.

Am I right in thinking that performing CPE on these additional metrics is the main reason for these additional networks? Put another way, if one were only interested in performing CPE on the reward itself, would using the standard Q-networks of the DQN model suffice?

Thanks

created time in 19 hours

pull request commentfacebookresearch/ReAgent

Oss test fixes jjg

This pull request was exported from Phabricator. Differential Revision: D27842449

MisterTea

comment created time in 21 hours

issue commentfacebookresearch/ReAgent

Installation Issues

Sorry, maybe I am missing something, but don't we also have non-python dependencies. For example, I thought part of the project uses Spark.

From the error message above it was my impression that the error was coming from the spark pipeline to pre-process the data. Did I misunderstand something?

tfurmston

comment created time in a day

pull request commentfacebookresearch/ReAgent

Oss test fixes jjg

This pull request was exported from Phabricator. Differential Revision: D27842449

MisterTea

comment created time in 2 days

pull request commentfacebookresearch/ReAgent

Oss test fixes jjg

This pull request was exported from Phabricator. Differential Revision: D27842449

MisterTea

comment created time in 2 days

pull request commentfacebookresearch/ReAgent

Update the ExprTk library commit

@MisterTea merged this pull request in facebookresearch/ReAgent@6c551e938f04f53c6436b700914f6212f26cccea.

ArashPartow

comment created time in 2 days

push eventfacebookresearch/ReAgent

Arash Partow

commit sha 6c551e938f04f53c6436b700914f6212f26cccea

Update the ExprTk library commit (#384) Summary: Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/384 Test Plan: CI Tests ...but without running open source tests. Reviewed By: gji1 Differential Revision: D27842452 Pulled By: MisterTea fbshipit-source-id: 6fb192d30217d358e86a04e6bcc5a69911276e71

view details

push time in 2 days

PR closed facebookresearch/ReAgent

Update the ExprTk library commit cla signed
+1 -1

3 comments

1 changed file

ArashPartow

pr closed time in 2 days

push eventfacebookresearch/ReAgent

Wanchao Liang

commit sha edc1c24f25d0212495b241613263362006d38cdb

Remove some pyre fixmes Reviewed By: divchenko Differential Revision: D27835360 fbshipit-source-id: cbb23793ee57382e43bd65bd40cfeb2820c6eec2

view details

push time in 3 days

pull request commentfacebookresearch/ReAgent

inc. training time for sparse dqn test

This pull request has been merged in facebookresearch/ReAgent@470f86032cdfa6b96899c85b9cfefc110cab9ee4.

kaiwenw

comment created time in 3 days

push eventfacebookresearch/ReAgent

Kaiwen Wang

commit sha 470f86032cdfa6b96899c85b9cfefc110cab9ee4

inc. training time for sparse dqn test (#454) Summary: Pull Request resolved: https://github.com/facebookresearch/ReAgent/pull/454 title Reviewed By: alexnikulkov Differential Revision: D27800185 fbshipit-source-id: 406001b48f55d7304d18e06237e7bf82ed07c11b

view details

push time in 3 days

PR closed facebookresearch/ReAgent

inc. training time for sparse dqn test cla signed fb-exported

Summary: title

Differential Revision: D27800185

+1 -1

2 comments

1 changed file

kaiwenw

pr closed time in 3 days

pull request commentfacebookresearch/ReAgent

Four fixes for RASP Tutorial

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Facebook open source project. Thanks!

roelbertens

comment created time in 3 days

pull request commentfacebookresearch/ReAgent

Update the ExprTk library commit

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Facebook open source project. Thanks!

ArashPartow

comment created time in 3 days

pull request commentfacebookresearch/ReAgent

Update the ExprTk library commit

@MisterTea has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

ArashPartow

comment created time in 3 days

pull request commentfacebookresearch/ReAgent

Four fixes for RASP Tutorial

@MisterTea has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

roelbertens

comment created time in 3 days

pull request commentfacebookresearch/ReAgent

Fix discrete samplers

@MisterTea has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

wall-ed-coder

comment created time in 3 days