Upgrade PyTorch Lightning to 1.0.2 (#7852)

Co-authored-by: Sam Shleifer <sshleifer@gmail.com>

examples/lightning_base.py

@@ -337,7 +337,7 @@ def add_generic_args(parser, root_dir) -> None:
 def generic_train(
     model: BaseTransformer,
     args: argparse.Namespace,
-    early_stopping_callback=False,
+    early_stopping_callback=None,
     logger=True,  # can pass WandbLogger() here
     extra_callbacks=[],
     checkpoint_callback=None,
@@ -355,6 +355,8 @@ def generic_train(
         checkpoint_callback = pl.callbacks.ModelCheckpoint(
             filepath=args.output_dir, prefix="checkpoint", monitor="val_loss", mode="min", save_top_k=1
         )
+    if early_stopping_callback:
+        extra_callbacks.append(early_stopping_callback)
     if logging_callback is None:
         logging_callback = LoggingCallback()
 
@@ -376,7 +378,6 @@ def generic_train(
         callbacks=[logging_callback] + extra_callbacks,
         logger=logger,
         checkpoint_callback=checkpoint_callback,
-        early_stop_callback=early_stopping_callback,
         **train_params,
     )
 

examples/requirements.txt

@@ -5,7 +5,7 @@ psutil
 sacrebleu
 rouge-score
 tensorflow_datasets
-pytorch-lightning==0.9.0
+pytorch-lightning==1.0.4
 matplotlib
 git-python==1.0.3
 faiss-cpu

examples/seq2seq/callbacks.py

@@ -102,7 +102,6 @@ def get_checkpoint_callback(output_dir, metric, save_top_k=1, lower_is_better=Fa
         monitor=f"val_{metric}",
         mode="min" if "loss" in metric else "max",
         save_top_k=save_top_k,
-        period=0,  # maybe save a checkpoint every time val is run, not just end of epoch.
     )
     return checkpoint_callback
 

examples/seq2seq/finetune.py

@@ -182,7 +182,6 @@ def validation_step(self, batch, batch_idx) -> Dict:
         return self._generative_step(batch)
 
     def validation_epoch_end(self, outputs, prefix="val") -> Dict:
-
         self.step_count += 1
         losses = {k: torch.stack([x[k] for x in outputs]).mean() for k in self.loss_names}
         loss = losses["loss"]
@@ -252,7 +251,7 @@ def get_dataset(self, type_path) -> Seq2SeqDataset:
     def get_dataloader(self, type_path: str, batch_size: int, shuffle: bool = False) -> DataLoader:
         dataset = self.get_dataset(type_path)
 
-        if self.hparams.sortish_sampler and type_path != "test":
+        if self.hparams.sortish_sampler and type_path != "test" and type_path != "val":
             sampler = dataset.make_sortish_sampler(batch_size, distributed=self.hparams.gpus > 1)
             return DataLoader(
                 dataset,
@@ -263,7 +262,7 @@ def get_dataloader(self, type_path: str, batch_size: int, shuffle: bool = False)
                 sampler=sampler,
             )
 
-        elif self.hparams.max_tokens_per_batch is not None and type_path != "test":
+        elif self.hparams.max_tokens_per_batch is not None and type_path != "test" and type_path != "val":
             batch_sampler = dataset.make_dynamic_sampler(
                 self.hparams.max_tokens_per_batch, distributed=self.hparams.gpus > 1
             )

examples/seq2seq/test_bash_script.py

@@ -144,14 +144,14 @@ def test_opus_mt_distill_script(self):
                 f"--num_train_epochs={epochs}",
                 "--warmup_steps=10",
                 "--val_check_interval=1.0",
+                "--do_predict",
             ]
         )
         with patch.object(sys, "argv", testargs):
             parser = argparse.ArgumentParser()
             parser = pl.Trainer.add_argparse_args(parser)
             parser = BartSummarizationDistiller.add_model_specific_args(parser, os.getcwd())
             args = parser.parse_args()
-            args.do_predict = False
             # assert args.gpus == gpus THIS BREAKS for multigpu
 
             model = distill_main(args)

examples/seq2seq/test_seq2seq_examples_multi_gpu.py

@@ -176,7 +176,6 @@ def convert(k, v):
         print(metrics)
         last_step_stats = metrics["val"][-1]
         self.assertGreaterEqual(last_step_stats["val_avg_gen_time"], 0.01)
-        self.assertGreaterEqual(1.0, last_step_stats["val_avg_gen_time"])
         self.assertIsInstance(last_step_stats[f"val_avg_{val_metric}"], float)
         self.assertEqual(len(metrics["test"]), 1)
         desired_n_evals = int(args_d["max_epochs"] * (1 / args_d["val_check_interval"]) / 2 + 1)

examples/text-classification/run_pl_glue.py

@@ -192,7 +192,7 @@ def main():
 
     # Optionally, predict on dev set and write to output_dir
     if args.do_predict:
-        checkpoints = list(sorted(glob.glob(os.path.join(args.output_dir, "checkpointepoch=*.ckpt"), recursive=True)))
+        checkpoints = list(sorted(glob.glob(os.path.join(args.output_dir, "checkpoint-epoch=*.ckpt"), recursive=True)))
         model = model.load_from_checkpoint(checkpoints[-1])
         return trainer.test(model)
 

examples/token-classification/run_pl_ner.py

@@ -207,9 +207,9 @@ def add_model_specific_args(parser, root_dir):
 
     if args.do_predict:
         # See https://github.com/huggingface/transformers/issues/3159
-        # pl use this format to create a checkpoint:
+        # pl use this default format to create a checkpoint:
         # https://github.com/PyTorchLightning/pytorch-lightning/blob/master\
-        # /pytorch_lightning/callbacks/model_checkpoint.py#L169
-        checkpoints = list(sorted(glob.glob(os.path.join(args.output_dir, "checkpointepoch=*.ckpt"), recursive=True)))
+        # /pytorch_lightning/callbacks/model_checkpoint.py#L322
+        checkpoints = list(sorted(glob.glob(os.path.join(args.output_dir, "checkpoint-epoch=*.ckpt"), recursive=True)))
         model = model.load_from_checkpoint(checkpoints[-1])
         trainer.test(model)