Delete classy elastic trainer and dependency to pet checkpoint api (pytorch#80)

Summary:
Pull Request resolved: pytorch#80

Pull Request resolved: facebookresearch/ClassyVision#464

Obsolete code as of D20787422.

Reviewed By: vreis

Differential Revision: D20787751

fbshipit-source-id: d0bba49902467e9f117ec6ac8199a7fc05b91ab4

Author: Kiuk Chung

test/p2p/elastic_trainer_test_base.py

@@ -624,60 +624,6 @@ def test_sync_retryable_exception(self):
         # are retryable / non-fatal
         self.assertEqual([410, 410, 410, 410], sums)
 
-    def test_checkpoint(self):
-        """
-        Test with 4 trainers:
-            - Save checkpoint every train_step
-            - Trainers suicide at 3rd step
-            - Restart training (from checkpoint)
-        """
-
-        def process_crash():
-            log.warning("Suicide, pid:{}".format(os.getpid()))
-            os.kill(os.getpid(), signal.SIGKILL)
-
-        hooks = {"process_crash": process_crash}
-        run_id = self._generate_run_id()
-
-        nprocs = 4
-
-        # Before training, there is no checkpoint
-        checkpoint_manager = FileSystemCheckpointManager(self.test_dir.name)
-        self.assertEqual(0, len(checkpoint_manager.list_checkpoints()))
-
-        for _ in range(0, nprocs):
-            _, qout, qerr = self._spawn(
-                self._train_with_checkpoint, run_id, _train_step, hooks
-            )
-
-        # wait all training process complete
-        # clean up for next run
-        self._wait_all_and_clean()
-
-        # we run 2 steps before suicide, expect two checkpoints be saved
-        self.assertEqual(2, len(checkpoint_manager.list_checkpoints()))
-
-        qouts = []
-        qerrs = []
-        # start next run
-        for _ in range(0, nprocs):
-            _, qout, qerr = self._spawn(
-                self._train_with_checkpoint, run_id, _train_step, None
-            )
-            qouts.append(qout)
-            qerrs.append(qerr)
-
-        # Gather all nums and sums from final states, they should match the input
-        sums = []
-        for i in range(0, nprocs):
-            state = _get_or_raise(qouts[i], qerrs[i])
-            # Everyone reads 3 samples after recovering from checkpoint:
-            self.assertEqual(3, len(state.nums))
-            sums.append(state.total_sum)
-
-        # The job should be completely recovered through checkpoints / crashes:
-        self.assertEqual([410, 410, 410, 410], sums)
-
     def test_process_crash(self):
         """
         Test 4 trainers, 2 of which SIGKILL themselves and terminate.

torchelastic/train_loop.py

@@ -11,7 +11,6 @@
 import warnings
 
 import torchelastic
-from torchelastic.checkpoint import CheckpointUtil
 from torchelastic.coordinator import NonRetryableException, StopException
 from torchelastic.metrics import get_elapsed_time_ms, publish_metric
 
@@ -69,8 +68,6 @@ def run_train(coordinator, train_step_gen, state):
     failure_count = 0
     rank = 0
 
-    checkpoint_util = CheckpointUtil(coordinator)
-
     while not coordinator.should_stop_training():
         # See: https://github.com/pytorch/elastic/issues/7
         if failure_count >= MAX_FAILURES:
@@ -90,17 +87,14 @@ def run_train(coordinator, train_step_gen, state):
             # does not sync.
             coordinator.barrier()
 
-            # load checkpoint if necessary
-            state = checkpoint_util.load_checkpoint(state, rank)
-
             state_sync_start_time = time.time()
             state.sync(world_size, rank)
             publish_metric(
                 "torchelastic",
                 "state_sync.duration.ms",
                 get_elapsed_time_ms(state_sync_start_time),
             )
-            checkpoint_util.set_checkpoint_loaded()
+
             coordinator.barrier()
             log.info("Rank {0} synced state with other nodes".format(rank))
         except StopException:
@@ -140,7 +134,6 @@ def run_train(coordinator, train_step_gen, state):
 
                 coordinator.monitor_progress(state, worker_stats)
 
-                checkpoint_util.save_checkpoint(state, rank)
                 if coordinator.should_rendezvous(state):
                     log.info("Rank {0} will re-rendezvous".format(rank))
                     # Executor told us, for whatever reason, to re-rendezvous.