Logging

hive/__init__.py

@@ -1,5 +1,11 @@
+import logging
 import os
 
+logging.basicConfig(
+    format="[%(levelname)s %(asctime)s %(filename)s:%(lineno)s] %(message)s",
+    level=logging.INFO,
+)
+
 from hive import agents, envs, replays, runners, utils
 from hive.utils.registry import Registrable, registry
 

hive/agents/__init__.py

@@ -1,3 +1,5 @@
+import logging
+
 from hive.agents import qnets
 from hive.agents.agent import Agent
 from hive.agents.dqn import DQNAgent
@@ -16,4 +18,6 @@
     },
 )
 
+logging.info("Registered agents.")
+
 get_agent = getattr(registry, f"get_{Agent.type_name()}")

hive/agents/qnets/__init__.py

@@ -1,3 +1,5 @@
+import logging
+
 from hive.utils.registry import registry
 from hive.agents.qnets.atari import NatureAtariDQNModel
 from hive.agents.qnets.base import FunctionApproximator
@@ -13,4 +15,6 @@
     },
 )
 
+logging.info("Registered function approximators.")
+
 get_qnet = getattr(registry, f"get_{FunctionApproximator.type_name()}")

hive/agents/qnets/utils.py

@@ -1,3 +1,4 @@
+import logging
 import math
 
 import torch
@@ -138,4 +139,6 @@ def type_name(cls):
     },
 )
 
+logging.info("Registered PyTorch initialization functions.")
+
 get_optimizer_fn = getattr(registry, f"get_{InitializationFn.type_name()}")

hive/configs/gym/dqn.yml

@@ -1,6 +1,6 @@
 run_name: &run_name 'gym-dqn'
 train_steps: 50000
-test_frequency: 200
+test_frequency: 5000
 test_episodes: 10
 max_steps_per_episode: 1000
 stack_size: &stack_size 1

hive/envs/__init__.py

@@ -1,3 +1,5 @@
+import logging
+
 from hive.envs.base import BaseEnv, ParallelEnv
 from hive.envs.env_spec import EnvSpec
 from hive.envs.gym_env import GymEnv
@@ -41,4 +43,6 @@
     },
 )
 
+logging.info('Registered environments.')
+
 get_env = getattr(registry, f"get_{BaseEnv.type_name()}")

hive/replays/__init__.py

@@ -1,3 +1,4 @@
+import logging
 from hive.replays.circular_replay import CircularReplayBuffer, SimpleReplayBuffer
 from hive.replays.legal_moves_replay import LegalMovesBuffer
 from hive.replays.prioritized_replay import PrioritizedReplayBuffer
@@ -14,4 +15,6 @@
     },
 )
 
+logging.info("Registered replays.")
+
 get_replay = getattr(registry, f"get_{BaseReplayBuffer.type_name()}")

hive/runners/base.py

@@ -1,8 +1,11 @@
+import logging
 from abc import ABC
+from asyncio.log import logger
+import pprint
 from typing import List
+
 from hive.agents.agent import Agent
 from hive.envs.base import BaseEnv
-
 from hive.runners.utils import Metrics
 from hive.utils import schedule
 from hive.utils.experiment import Experiment
@@ -124,6 +127,8 @@ def run_episode(self):
     def run_training(self):
         """Run the training loop."""
         self.train_mode(True)
+        logging.info("Starting train loop")
+
         while self._train_schedule.get_value():
             # Run training episode
             if not self._training:
@@ -135,13 +140,20 @@ def run_training(self):
 
             # Run test episodes
             if self._run_testing:
+                logging.info(
+                    f"{self._train_schedule._steps}/"
+                    f"{self._train_schedule._flip_step} training steps completed."
+                )
+                logger.info("Running testing.")
                 test_metrics = self.run_testing()
                 self._logger.update_step("test")
                 self._logger.log_metrics(test_metrics, "test")
                 self._run_testing = False
+                logging.info(f"Testing results: {pprint.pformat(test_metrics)}")
 
             # Save experiment state
             if self._save_experiment:
+                logger.info("Saving run.")
                 self._experiment_manager.save()
                 self._save_experiment = False
 

hive/runners/multi_agent_loop.py

@@ -1,5 +1,7 @@
 import argparse
 import copy
+import logging
+import pprint
 
 from hive import agents as agent_lib
 from hive import envs
@@ -274,6 +276,11 @@ def main():
         args.logger_config,
     )
     runner = set_up_experiment(config)
+    logging.info(
+        f"Using config: \n{pprint.pformat(runner._experiment_manager._config, compact=True)}"
+    )
+    logging.info("Created runner. Starting run!")
+
     runner.run_training()
 
 

hive/runners/single_agent_loop.py

@@ -1,5 +1,7 @@
 import argparse
 import copy
+import logging
+import pprint
 
 from hive import agents as agent_lib
 from hive import envs
@@ -209,6 +211,11 @@ def main():
         args.logger_config,
     )
     runner = set_up_experiment(config)
+    logging.info(
+        f"Using config: \n{pprint.pformat(runner._experiment_manager._config, compact=False)}"
+    )
+    logging.info("Created runner. Starting run!")
+
     runner.run_training()
 
 

hive/utils/loggers.py

@@ -1,5 +1,6 @@
 import abc
 import copy
+import logging
 import os
 from typing import List
 
@@ -450,4 +451,6 @@ def load(self, dir_name):
     },
 )
 
+logging.info("Registered loggers.")
+
 get_logger = getattr(registry, f"get_{Logger.type_name()}")

hive/utils/schedule.py

@@ -1,4 +1,5 @@
 import abc
+import logging
 
 from hive.utils.registry import Registrable, registry
 
@@ -200,4 +201,6 @@ def __repr__(self):
     },
 )
 
+logging.info("Registered schedules.")
+
 get_schedule = getattr(registry, f"get_{Schedule.type_name()}")

hive/utils/torch_utils.py

@@ -1,3 +1,4 @@
+import logging
 import numpy as np
 import torch
 from torch import optim
@@ -193,6 +194,8 @@ def step(self, closure=None):
     },
 )
 
+logging.info("Registered PyTorch optimizers.")
+
 registry.register_all(
     LossFn,
     {
@@ -218,5 +221,7 @@ def step(self, closure=None):
     },
 )
 
+logging.info("Registered PyTorch losses.")
+
 get_optimizer_fn = getattr(registry, f"get_{OptimizerFn.type_name()}")
 get_loss_fn = getattr(registry, f"get_{LossFn.type_name()}")