Closed manishshettym closed 8 months ago
Ideal presentation:
# https://github.com/opium-sh/prl/blob/3e21f8c7c87cfc7aee84d9e264c3a8b2bc549076/prl/callbacks/callbacks.py#L345
# def on_iteration_end(self, agent: AgentABC):
# agent_logs = agent_logger.flush(self.agent_logger_cid)
# mean_test_reward = np.mean(agent_logs[0]["test_episode_total_reward"])
mean_test_episode_length = np.mean(agent_logs[0]["test_episode_length"])
# history_summary = (mean_test_reward, mean_test_episode_length)
# if self.on_screen:
# self._log_on_screen(agent.iteration_count, history_summary)
# if self.to_file:
# self._log_to_file(agent.iteration_count, history_summary)
Highlight the "idiomatic" partial program in context of the overall function.
https://github.com/tart-proj/codescholar/blob/9914b8f1c2079ef7a709d6e6d0f6b06132006e8e/codescholar/search/search.py#L42
This function currently takes a bunch of idioms generated and saves the partial programs (partial sasts transformed) python files. This is not ideal:
clarity
-- a partial prog, esp., for a small-sized idiom, has very less info.context
-- a partial prog is concise but loses a lot of the context of the problem itself (surrounding code)provenance
-- a partial prog (presented as an idiom) has separated from the original function; hence the promise of provenance (to a github program) is lost.