Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / thiagopbueno/tf-plan issues and pull requests

#43 - Add CEC-based online stochastic planner

Issue - State: open - Opened by thiagopbueno almost 4 years ago
Labels: refactor

#42 - chore(deps): bump tensorflow-cpu from 1.15 to 2.1.2

Pull Request - State: closed - Opened by dependabot[bot] almost 4 years ago - 1 comment
Labels: dependencies

#41 - Dump nonfluents in CSV file before running any planner

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#40 - Report experiment summary

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#39 - Report trajectory results for each trial

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#38 - Apply discount factor over rewards for online planning

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#35 - Create separate click app for tensorplan

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: refactor

#33 - Add the total reward for each online step to trajectory stats in stochastic planners

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#31 - Add training epoch scheduler

Pull Request - State: closed - Opened by thiagopbueno over 4 years ago

#30 - Add planning horizon and warm-start hyperparameters

Pull Request - State: closed - Opened by thiagopbueno over 4 years ago

#29 - Report training losses curves for each evaluation timestep

Issue - State: closed - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#28 - Add training epochs scheduler

Issue - State: closed - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#27 - Add regularization term over the l2-norm of the policy variables

Issue - State: open - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#26 - Add warm start in re-planning

Issue - State: closed - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#25 - Separate planning horizon and evaluation horizon

Issue - State: closed - Opened by thiagopbueno over 4 years ago
Labels: enhancement

#24 - chore(deps): Update tensorflow requirement from <2.0.0 to <3.0.0

Pull Request - State: closed - Opened by dependabot[bot] almost 5 years ago - 2 comments
Labels: dependencies

#23 - Bump tensorflow from 1.5.0 to 1.12.2

Pull Request - State: closed - Opened by dependabot[bot] almost 5 years ago - 1 comment
Labels: dependencies

#22 - Configure tf.ConfigProto.gpu_options

Issue - State: open - Opened by thiagopbueno almost 5 years ago

#21 - Control and log the device placement in RNN models

Issue - State: open - Opened by thiagopbueno almost 5 years ago

#20 - Add time profiling for stochastic planners

Issue - State: open - Opened by thiagopbueno almost 5 years ago
Labels: enhancement

#19 - Add summary ops and tensorboard support

Issue - State: closed - Opened by thiagopbueno almost 5 years ago

#18 - Add logging capabilities for trajectory debugging

Issue - State: open - Opened by thiagopbueno almost 5 years ago

#17 - Add tqdm in training loop

Issue - State: closed - Opened by thiagopbueno almost 5 years ago
Labels: enhancement

#15 - Fix sampled noise inputs between online execution steps

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: bug

#14 - Remove dependencies to tf-rddlsim package

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: refactor

#13 - Unroll trajectory graph with actions and noise as inputs

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement, good first issue

#12 - Add arguments to ActionOptimizer to select and configure inner optimizer

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#11 - Unify test/evaluator.py and planners/environment.py modules

Issue - State: closed - Opened by thiagopbueno about 5 years ago - 1 comment
Labels: refactor

#10 - Refactor ActionOptimizer class to define general interface for Planner class

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: refactor

#9 - Create Planner base class

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: refactor

#8 - Build the hindsight value function ops

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#7 - Build straight-line value function ops

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#6 - Sample a batch of trajectory scenarios via re-parameterization

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#5 - Add separate sub-parsers for online and offline tfplan script arguments and options

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#4 - Refactor online planner to avoid re-building the RNN dynamic unrolling ops each timestep

Issue - State: closed - Opened by thiagopbueno about 5 years ago
Labels: enhancement

#3 - GPU vs CPU performance

Issue - State: open - Opened by meneguzzi almost 6 years ago - 1 comment

#2 - Issue with tf-plan

Issue - State: closed - Opened by ramonpereira almost 6 years ago - 9 comments

#1 - Merge pull request #1 from thiagopbueno/master

Pull Request - State: closed - Opened by songziya almost 6 years ago