readwrite

File reading and writing utilities for Neuroevolution fitting.

cneuromax.fitting.neuroevolution.utils.readwrite.find_existing_save_points(output_dir)

Returns a list of existing save points.

Parameters:

output_dir (str) – See output_dir.

Return type:

list[int]

Returns:

The list of existing save points.

cneuromax.fitting.neuroevolution.utils.readwrite.load_state(prev_num_gens, len_agents_batch, output_dir)

Load a previous experiment state from disk.

Parameters:

• prev_num_gens (int) – See prev_num_gens.

• len_agents_batch (int) – See len_agents_batch.

• output_dir (str) – See output_dir.

Return type:

tuple[list[list[BaseAgent]], ndarray[Shape[Pop_size, Num_pops, [fitness, num_env_steps, serialized_agent_size]], dtype[float32]] | None, Optional[int]]

Returns:

• See ~.compute_generation_results.agents_batch`.

• See

  generation_results.

• See

  total_num_env_steps.

cneuromax.fitting.neuroevolution.utils.readwrite.save_state(agents_batch, generation_results, total_num_env_steps, curr_gen, output_dir)

Dump the current experiment state to disk.

Parameters:

• agents_batch (list[list[BaseAgent]]) – See agents_batch.

• generation_results (ndarray[Shape[Pop_size, Num_pops, [fitness, num_env_steps, serialized_agent_size]], dtype[float32]] | None) – See generation_results.

• total_num_env_steps (Optional[int]) – See total_num_env_steps.

• curr_gen (int) – See curr_gen.

• output_dir (str) – See output_dir.

Return type:

None