Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / tensorflow/agents issues and pull requests
#828 - Why collect_policy of DDQN agent seems to be unrelated to policy when I reload from checkpoint?
Issue -
State: open - Opened by aonurgiray over 1 year ago
#827 - dqn_agent.DqnAgent: epsilon_greedy not modifiable during training
Issue -
State: open - Opened by fede72bari over 1 year ago
#826 - Upgrade numpy version in 1_dqn_tutorial.ipynb
Pull Request -
State: closed - Opened by kiransair over 1 year ago
- 2 comments
#825 - Replacing CartPole with a custom simple task
Issue -
State: closed - Opened by hopskipnfall over 1 year ago
- 1 comment
#824 - Migration of suite_gym module to gymnasium 0.26
Issue -
State: closed - Opened by grizzlybearg almost 2 years ago
#823 - Updating pygame requirement for Python 3.11
Pull Request -
State: closed - Opened by sgboakes almost 2 years ago
- 1 comment
#822 - observation_and_action_constraint_splitter with LinearUCBAgent using MovieLensPyEnvironment
Issue -
State: closed - Opened by My3VM almost 2 years ago
- 1 comment
#821 - PPO Schulman's example doesn't work
Issue -
State: closed - Opened by grizzlybearg almost 2 years ago
- 1 comment
#820 - CounterV2 deprecated substituted with Dataset.Counter
Pull Request -
State: closed - Opened by tiamilani almost 2 years ago
- 5 comments
#819 - Can we change tf.compat.v1.where to tf.compat.v2.where in EpsilonGreedyPolicy._action
Issue -
State: closed - Opened by JustinACoder almost 2 years ago
- 1 comment
#818 - Checkpointer.save, can global_step become a kwarg?
Issue -
State: open - Opened by tiamilani almost 2 years ago
#817 - How can we load from specific checkpoints use checkpointer?
Issue -
State: open - Opened by wj210 almost 2 years ago
- 1 comment
#816 - CategoricalDqnAgent Distributional Training Loss Function error
Issue -
State: open - Opened by jacklu333333 almost 2 years ago
#815 - Speed Bottleneck due to dataset
Issue -
State: open - Opened by jacklu333333 almost 2 years ago
#814 - docs: use tf env instead of py env for driver
Pull Request -
State: closed - Opened by mcanevet almost 2 years ago
- 3 comments
#813 - AttributeError: module 'tree' has no attribute 'assert_same_structure'
Issue -
State: open - Opened by abbiesgame almost 2 years ago
#812 - Driver error with gpu based TFUniformReplayBuffer
Issue -
State: closed - Opened by jacklu333333 almost 2 years ago
- 1 comment
#811 - collect_step slow speed
Issue -
State: closed - Opened by jacklu333333 almost 2 years ago
- 1 comment
#808 - SAC minitaur with the Actor-Learner API demonstrator fails
Issue -
State: open - Opened by ThorAvaTahr almost 2 years ago
#807 - Errors with numpy 1.24.0
Issue -
State: open - Opened by sebastianknopf almost 2 years ago
- 10 comments
#806 - NVLink Configuration
Issue -
State: closed - Opened by jacklu333333 almost 2 years ago
- 1 comment
#805 - PPO with Mini-Batches Tutorial
Issue -
State: open - Opened by kochlisGit almost 2 years ago
#804 - Documentation: Fix broken Deep Q Network with TFA build
Pull Request -
State: closed - Opened by 8bitmp3 almost 2 years ago
- 3 comments
#803 - Error in documentation.
Issue -
State: closed - Opened by MarkosMuche almost 2 years ago
- 3 comments
#802 - BernoulliThompsonSamplingAgent error: 'tensorflow.python.eager.function' has no attribute 'register'`
Issue -
State: open - Opened by sylviawhx almost 2 years ago
#801 - replay buffer not working for contexual bandit with scalar action spec
Issue -
State: open - Opened by sylviawhx almost 2 years ago
- 2 comments
#800 - tf_agents.replay_buffers memory allocation
Issue -
State: open - Opened by soonjune almost 2 years ago
#799 - Tutorial on Multi objective optimization and Constraint Optimization
Issue -
State: open - Opened by sj31867 almost 2 years ago
#798 - Contextual Bandits High training time
Issue -
State: open - Opened by sj31867 almost 2 years ago
Labels: bandits
#797 - Actor/Learner DQN Pong: Using PyUniformReplayBuffer instead of ReverbReplayBuffer not possible
Issue -
State: closed - Opened by Sch-Stef about 2 years ago
- 2 comments
#796 - Update actor/learner, distributed docs
Pull Request -
State: closed - Opened by coreyleveen about 2 years ago
- 1 comment
#795 - ValueError: Exception encountered when calling layer "QNetwork" (Issues with Q-Networks)
Issue -
State: open - Opened by techGIAN about 2 years ago
- 1 comment
#794 - TypeError: Cannot find minimum value of <dtype: 'string'> with type <dtype: 'string'> when creating an environment
Issue -
State: closed - Opened by techGIAN about 2 years ago
- 1 comment
#793 - Data normalization for Contexual Bandits
Issue -
State: open - Opened by sj31867 about 2 years ago
- 1 comment
Labels: bandits
#792 - How to customize Actor policy in Actor-Learner setup
Issue -
State: open - Opened by JLenssen about 2 years ago
- 1 comment
#791 - Contextual Bandit Off-Policy Evaluation
Issue -
State: open - Opened by vitorkrasniqi about 2 years ago
- 2 comments
#790 - EnvironmentSteps tf_metric bug with parallel envs
Issue -
State: open - Opened by vittorione94 about 2 years ago
- 1 comment
#789 - GymWrapper wraps action_space datatype to dtype=numpy.dtype[float32]
Issue -
State: closed - Opened by Wajktor about 2 years ago
#788 - dealing with mjWARN_BADQACC
Issue -
State: open - Opened by vittorione94 about 2 years ago
- 1 comment
#787 - DDPG Parallel environment example is broken
Issue -
State: closed - Opened by vittorione94 about 2 years ago
- 3 comments
#786 - How can I save network.DistributionNetwork?
Issue -
State: open - Opened by Rejuy about 2 years ago
- 1 comment
#785 - Fix time_step, trajectory format for tf.print
Pull Request -
State: closed - Opened by coreyleveen about 2 years ago
- 2 comments
#784 - Incorrect formatting for time_step, trajectory when using tf.print
Issue -
State: open - Opened by coreyleveen about 2 years ago
#783 - Allow passing conv_type from ActionDistributionNetwork to EncodingNetwork
Pull Request -
State: closed - Opened by kochlisGit about 2 years ago
- 2 comments
#782 - Learner.run got stuck
Issue -
State: closed - Opened by Rejuy about 2 years ago
- 3 comments
#779 - Conv1D Option for Networks
Issue -
State: open - Opened by kochlisGit about 2 years ago
- 5 comments
#778 - Using `tf-agents` for Bandits with sparse data
Issue -
State: open - Opened by ujjwal95 about 2 years ago
- 1 comment
#777 - Fix Issue 776: Incorect example to run dqn_train_eval_rnn.py
Pull Request -
State: closed - Opened by karnehm about 2 years ago
- 5 comments
#776 - Incorect example to run dqn_train_eval_rnn.py
Issue -
State: open - Opened by karnehm about 2 years ago
- 2 comments
#775 - DQN sample - AverageReturn output is same as AverageEpisodeLength
Issue -
State: open - Opened by maxima120 about 2 years ago
#774 - Implement batched observer unbatching
Pull Request -
State: closed - Opened by philstahlfeld about 2 years ago
- 1 comment
#773 - Loaded policy eval runs 4 times faster than original policy eval
Issue -
State: open - Opened by maxima120 about 2 years ago
#772 - Batching ReverbAddEpisodeObserver with variable length episodes
Issue -
State: open - Opened by philstahlfeld about 2 years ago
- 2 comments
#771 - Distributed training like SEED RL
Issue -
State: closed - Opened by philstahlfeld about 2 years ago
- 1 comment
#770 - Keras model usage
Issue -
State: open - Opened by teaglin about 2 years ago
#769 - Custom - minmax pooling - Keras - Tensorflow
Issue -
State: closed - Opened by ashishbhong82 about 2 years ago
#766 - ValueError: Given `time_step`: TimeStep
Issue -
State: closed - Opened by Shinwazu over 2 years ago
- 11 comments
#764 - Value Error
Issue -
State: open - Opened by Shinwazu over 2 years ago
- 1 comment
#763 - Add ability to pass multiple inputs to a single preprocessing layer in EncodingNetwork
Pull Request -
State: closed - Opened by boomanaiden154 over 2 years ago
- 5 comments
#760 - Does TF-Agents not support XLA?
Issue -
State: open - Opened by connor-create over 2 years ago
- 5 comments
#760 - Does TF-Agents not support XLA?
Issue -
State: open - Opened by connor-create over 2 years ago
- 5 comments
#759 - Multiple actions for PPOAgent
Issue -
State: open - Opened by DavyMorgan over 2 years ago
- 5 comments
#759 - Multiple actions for PPOAgent
Issue -
State: open - Opened by DavyMorgan over 2 years ago
- 5 comments
#759 - Multiple actions for PPOAgent
Issue -
State: open - Opened by DavyMorgan over 2 years ago
- 5 comments
#754 - Hyperlink to V2 examples
Pull Request -
State: closed - Opened by chunduriv over 2 years ago
- 1 comment
#753 - Fixing broken link in `8_networks_tutorial.ipynb`
Pull Request -
State: closed - Opened by chunduriv over 2 years ago
- 1 comment
#749 - Rename `tf_agents.policies.policy_saver.PolicySaver`
Pull Request -
State: closed - Opened by chunduriv over 2 years ago
- 2 comments
#744 - Fix TFEnvironment#reward_spec docstring
Pull Request -
State: closed - Opened by coreyleveen over 2 years ago
- 1 comment
#741 - Fixed two typos in tutorial notebooks.
Pull Request -
State: closed - Opened by olitheolix over 2 years ago
- 2 comments
#737 - How to use the replay buffer in tf_agents for contextual bandit, that predicts and trains on a daily basis
Issue -
State: open - Opened by tejavenkatk over 2 years ago
- 2 comments
Labels: bandits
#732 - action output and policy_step_spec structures do not match:
Issue -
State: open - Opened by PeterDomanski over 2 years ago
- 7 comments
#731 - Bias Layer: Add support for regularizer and constraint
Pull Request -
State: closed - Opened by saisua over 2 years ago
- 4 comments
#728 - Feature Proposal: Implement return_sequences=False support in RNNWrapper
Issue -
State: open - Opened by joraso over 2 years ago
- 1 comment
#722 - Support TFEnvironment in Actor
Pull Request -
State: closed - Opened by Rossil2012 over 2 years ago
- 2 comments
#721 - PPO: Unexpected output from `actor_network`. Expected `Distribution` objects, but saw output spec: TensorSpec(...)
Issue -
State: closed - Opened by PeterDomanski over 2 years ago
- 7 comments
#713 - ValueError: Exception encrounted when calling layer "QNetwork" (type QNetwork)
Issue -
State: open - Opened by dretechtips over 2 years ago
- 5 comments
#712 - Switch tf.losses.mse with tf.math.squared_difference in BehavioralCloningAgent
Pull Request -
State: closed - Opened by egordon over 2 years ago
- 3 comments
#705 - Add MultiCategoricalProjectionNetwork
Pull Request -
State: closed - Opened by sidney-tio almost 3 years ago
- 6 comments
#705 - Add MultiCategoricalProjectionNetwork
Pull Request -
State: open - Opened by sidney-tio almost 3 years ago
- 4 comments
#705 - Add MultiCategoricalProjectionNetwork
Pull Request -
State: open - Opened by sidney-tio almost 3 years ago
- 4 comments
#703 - InvalidArgumentError: Exception encountered when calling layer "dynamic_unroll" (type DynamicUnroll)
Issue -
State: open - Opened by amarshivaram almost 3 years ago
- 2 comments
#702 - Is there any agent implemented that can work with MultiBinary action space?
Issue -
State: closed - Opened by sibyjackgrove almost 3 years ago
- 7 comments
#702 - Is there any agent implemented that can work with MultiBinary action space?
Issue -
State: closed - Opened by sibyjackgrove almost 3 years ago
- 7 comments
#702 - Is there any agent implemented that can work with MultiBinary action space?
Issue -
State: closed - Opened by sibyjackgrove almost 3 years ago
- 7 comments
#672 - Offline Contextual Bandits
Issue -
State: closed - Opened by alex-seto about 3 years ago
- 3 comments
#670 - Adding New Action(s) to a Bandit Policy
Issue -
State: open - Opened by davidcereal about 3 years ago
- 3 comments
#670 - Adding New Action(s) to a Bandit Policy
Issue -
State: open - Opened by davidcereal about 3 years ago
- 3 comments
#670 - Adding New Action(s) to a Bandit Policy
Issue -
State: open - Opened by davidcereal about 3 years ago
- 3 comments
#665 - ValueError: Only scalar actions are supported now!!
Issue -
State: open - Opened by shamim237 about 3 years ago
- 7 comments
#656 - PPO policy with ActorDistributionNetwork and discrete action array
Issue -
State: open - Opened by cedavidyang about 3 years ago
- 13 comments
#621 - Feat: Add Tabular Solutions
Pull Request -
State: closed - Opened by morgandu over 3 years ago
Labels: cla: yes
#618 - auxiliary tasks with tf agents
Issue -
State: closed - Opened by npqst over 3 years ago
- 1 comment
#616 - environments/gym_wrapper forward reward_range from GymEnv to PyEnviro…
Pull Request -
State: closed - Opened by cmarlin over 3 years ago
- 1 comment
Labels: cla: yes
#616 - environments/gym_wrapper forward reward_range from GymEnv to PyEnviro…
Pull Request -
State: closed - Opened by cmarlin over 3 years ago
- 1 comment
Labels: cla: yes
#587 - PyEnvironment Methods Incompatible with TF
Issue -
State: closed - Opened by ergsfe over 3 years ago
- 7 comments
Labels: good first issue, contributions welcome
#572 - ModuleNotFoundError: No module named 'tf_agents.agents'; 'tf_agents' is not a package
Issue -
State: closed - Opened by kapilrathore42 over 3 years ago
- 2 comments
#572 - ModuleNotFoundError: No module named 'tf_agents.agents'; 'tf_agents' is not a package
Issue -
State: closed - Opened by kapilrathore42 over 3 years ago
- 2 comments
#572 - ModuleNotFoundError: No module named 'tf_agents.agents'; 'tf_agents' is not a package
Issue -
State: closed - Opened by kapilrathore42 over 3 years ago
- 2 comments
#533 - Does tf-agents support DQN agents with multiple inputs?
Issue -
State: closed - Opened by FMalerba almost 4 years ago
- 5 comments
#533 - Does tf-agents support DQN agents with multiple inputs?
Issue -
State: closed - Opened by FMalerba almost 4 years ago
- 5 comments