[train] New training options for logging/validation based on number of steps

parlai/agents/test_agents/test_agents.py

@@ -168,7 +168,7 @@ def eval_step(self, batch):
 
 class MockTrainUpdatesAgent(MockTorchAgent):
     """
-    Simulate training updates
+    Simulate training updates.
     """
 
     def train_step(self, batch):

parlai/nn/lr_scheduler.py

@@ -17,7 +17,6 @@
 import numpy as np
 
 from parlai.core.exceptions import StopTrainException
-import parlai.utils.logging as logging
 from parlai.utils.misc import warn_once
 
 
@@ -226,17 +225,10 @@ def lr_scheduler_factory(cls, opt, optimizer, states, hard_reset=False):
         warmup_updates = opt.get('warmup_updates', -1)
         warmup_rate = opt.get('warmup_rate', 1e-4)
         max_lr_steps = opt.get('max_train_steps', -1)
-        deprecated_max_lr_steps = opt.get('max_lr_steps', -1)
-        if deprecated_max_lr_steps > 0:
-            logging.warn(
-                '**DEPRECATED: --max-lr-steps is deprecated, please only specify '
-                '--max-train-steps instead'
+        if opt.get('max_lr_steps', -1) > 0:
+            raise ValueError(
+                '--max-lr-steps is **DEPRECATED**; please set --max-train-steps directly'
             )
-            if deprecated_max_lr_steps != max_lr_steps:
-                logging.warn(
-                    f'Setting max_lr_steps from {deprecated_max_lr_steps} to {max_lr_steps}'
-                )
-                max_lr_steps = deprecated_max_lr_steps
         invsqrt_lr_decay_gamma = opt.get('invsqrt_lr_decay_gamma', -1)
 
         if opt.get('lr_scheduler') == 'none':

parlai/scripts/train_model.py

@@ -704,13 +704,27 @@ def train(self):
                 exs_per_epoch = world.num_examples()
                 self._total_exs = int(np.round(self._total_epochs * exs_per_epoch))
                 # and use the primary worker's timings for everything
-                train_time, log_time, validate_time = sync_object(
-                    (
+                if any(
+                    getattr(self, k) < float('inf')
+                    for k in [
+                        'max_train_time',
+                        'log_every_n_secs',
+                        'validation_every_n_secs',
+                    ]
+                ):
+                    train_time, log_time, validate_time = sync_object(
+                        (
+                            self.train_time.time(),
+                            self.log_time.time(),
+                            self.validate_time.time(),
+                        )
+                    )
+                else:
+                    train_time, log_time, validate_time = (
                         self.train_time.time(),
                         self.log_time.time(),
                         self.validate_time.time(),
                     )
-                )
 
                 # check counters and timers
                 if self._total_epochs >= self.max_num_epochs:

tests/test_train_model.py

@@ -114,7 +114,7 @@ def test_multitasking_id_overlap(self):
 
     def _test_opt_step_opts(self, update_freq: int):
         """
-        Test -tstep, -vstep, -lstep
+        Test -tstep, -vstep, -lstep.
 
         :param update_freq:
             update frequency