[tune] Fix directory naming regression (#6839)

doc/source/rllib-training.rst

@@ -180,7 +180,7 @@ Tune will schedule the trials to run in parallel on your Ray cluster:
 Custom Training Workflows
 ~~~~~~~~~~~~~~~~~~~~~~~~~
 
-In the `basic training example <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_env.py>`__, Tune will call ``train()`` on your trainer once per iteration and report the new training results. Sometimes, it is desirable to have full control over training, but still run inside Tune. Tune supports `custom trainable functions <tune-usage.html#trainable-api>`__ that can be used to implement `custom training workflows (example) <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_train_fn.py>`__.
+In the `basic training example <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_env.py>`__, Tune will call ``train()`` on your trainer once per training iteration and report the new training results. Sometimes, it is desirable to have full control over training, but still run inside Tune. Tune supports `custom trainable functions <tune-usage.html#trainable-api>`__ that can be used to implement `custom training workflows (example) <https://github.com/ray-project/ray/blob/master/rllib/examples/custom_train_fn.py>`__.
 
 For even finer-grained control over training, you can use RLlib's lower-level `building blocks <rllib-concepts.html>`__ directly to implement `fully customized training workflows <https://github.com/ray-project/ray/blob/master/rllib/examples/rollout_worker_custom_workflow.py>`__.
 

python/ray/tune/tests/test_trial_runner.py

@@ -1,3 +1,4 @@
+import os
 import sys
 import unittest
 
@@ -57,7 +58,7 @@ class TrialRunnerTest(unittest.TestCase):
             trial_generator.add_configurations({name: spec})
             for trial in trial_generator.next_trials():
                 trial_executor.start_trial(trial)
-                self.assertLessEqual(len(trial.logdir), 200)
+                self.assertLessEqual(len(os.path.basename(trial.logdir)), 200)
                 trial_executor.stop_trial(trial)
 
     def testExtraResources(self):

python/ray/tune/tests/test_tune_save_restore.py

@@ -47,6 +47,8 @@ class SerialTuneRelativeLocalDirTest(unittest.TestCase):
         _register_all()
 
     def _get_trial_dir(self, absoulte_exp_dir):
+        print("looking for", self.MockTrainable._name)
+        print("in", os.listdir(absoulte_exp_dir))
         trial_dirname = next(
             (child_dir for child_dir in os.listdir(absoulte_exp_dir)
              if (os.path.isdir(os.path.join(absoulte_exp_dir, child_dir))

python/ray/tune/trial.py

@@ -270,7 +270,8 @@ class Trial:
         """Init logger."""
         if not self.result_logger:
             if not self.logdir:
-                self.logdir = Trial.create_logdir(str(self), self.local_dir)
+                self.logdir = Trial.create_logdir(
+                    str(self) + "_" + self.experiment_tag, self.local_dir)
             else:
                 os.makedirs(self.logdir, exist_ok=True)
 

python/ray/tune/tune.py

@@ -362,7 +362,7 @@ def run_experiments(experiments,
                     reuse_actors=False,
                     trial_executor=None,
                     raise_on_failed_trial=True,
-                    concurrent=False):
+                    concurrent=True):
     """Runs and blocks until all trials finish.
 
     Examples:

rllib/agents/trainer.py

@@ -210,7 +210,7 @@ COMMON_CONFIG = {
     # Whether to LZ4 compress individual observations
     "compress_observations": False,
     # Wait for metric batches for at most this many seconds. Those that
-    # have not returned in time will be collected in the next iteration.
+    # have not returned in time will be collected in the next train iteration.
     "collect_metrics_timeout": 180,
     # Smooth metrics over this many episodes.
     "metrics_smoothing_episodes": 100,
@@ -224,10 +224,10 @@ COMMON_CONFIG = {
     # but optimal value could be obtained by measuring your environment
     # step / reset and model inference perf.
     "remote_env_batch_wait_ms": 0,
-    # Minimum time per iteration
+    # Minimum time per train iteration
     "min_iter_time_s": 0,
     # Minimum env steps to optimize for per train call. This value does
-    # not affect learning, only the length of iterations.
+    # not affect learning, only the length of train iterations.
     "timesteps_per_iteration": 0,
     # This argument, in conjunction with worker_index, sets the random seed of
     # each worker, so that identically configured trials will have identical