agent.rst

Agent interface

Initialization and termination

tensorforce.agents.TensorforceAgent.create

tensorforce.agents.TensorforceAgent.close

Main reinforcement learning interface

tensorforce.agents.TensorforceAgent.act

tensorforce.agents.TensorforceAgent.observe

Required for evaluation at episode start

tensorforce.agents.TensorforceAgent.initial_internals

Loading and saving

tensorforce.agents.TensorforceAgent.load

tensorforce.agents.TensorforceAgent.save

Get and assign variables

tensorforce.agents.TensorforceAgent.get_variables

tensorforce.agents.TensorforceAgent.get_variable

tensorforce.agents.TensorforceAgent.assign_variable

Advanced functions for specialized use cases

tensorforce.agents.TensorforceAgent.experience

tensorforce.agents.TensorforceAgent.update

tensorforce.agents.TensorforceAgent.pretrain

Others

tensorforce.agents.TensorforceAgent.reset

tensorforce.agents.TensorforceAgent.get_output_tensors

tensorforce.agents.TensorforceAgent.get_available_summaries