Pin PTL, bump omegaconf (#1049)

* Pin PTL, bump omegaconf Signed-off-by: smajumdar <titu1994@gmail.com> * Patch config preservation Signed-off-by: smajumdar <titu1994@gmail.com> * Patch config preservation Signed-off-by: smajumdar <titu1994@gmail.com> * Patch config preservation Signed-off-by: smajumdar <titu1994@gmail.com> * Correct preservation of decoder config after update Signed-off-by: smajumdar <titu1994@gmail.com> * Bump hydra-core Signed-off-by: smajumdar <titu1994@gmail.com> * Hyper parameter saving patch Signed-off-by: smajumdar <titu1994@gmail.com>
NVIDIA · Aug 20, 2020 · fcc1d99 · fcc1d99
1 parent cf0d554
commit fcc1d99
Show file tree

Hide file tree

Showing 5 changed files with 21 additions and 11 deletions.
diff --git a/nemo/collections/asr/models/classification_models.py b/nemo/collections/asr/models/classification_models.py
@@ -16,7 +16,7 @@
 from typing import Dict, List, Optional, Union
 
 import torch
-from omegaconf import DictConfig, ListConfig
+from omegaconf import DictConfig, ListConfig, OmegaConf
 from pytorch_lightning import Trainer
 
 from nemo.collections.asr.data.audio_to_text import AudioLabelDataset
@@ -246,7 +246,10 @@ def change_labels(self, new_labels: List[str]):
 
             # Update config
             self._cfg.labels = new_labels
-            self._cfg.decoder.params = new_decoder_config
+
+            OmegaConf.set_struct(self._cfg.decoder, False)
+            self._cfg.decoder = new_decoder_config
+            OmegaConf.set_struct(self._cfg.decoder, True)
 
             if 'train_ds' in self._cfg and self._cfg.train_ds is not None:
                 self._cfg.train_ds.labels = new_labels

diff --git a/nemo/collections/asr/models/ctc_bpe_models.py b/nemo/collections/asr/models/ctc_bpe_models.py
@@ -256,17 +256,20 @@ def change_vocabulary(self, new_tokenizer_dir: str, new_tokenizer_type: str):
         # Override number of classes if placeholder provided
         logging.info(
             "\nReplacing old number of classes ({}) with new number of classes - {}".format(
-                decoder_config.params['num_classes'], len(vocabulary)
+                decoder_config['params']['num_classes'], len(vocabulary)
             )
         )
-        decoder_config.params['num_classes'] = len(vocabulary)
+        decoder_config['params']['num_classes'] = len(vocabulary)
 
         del self.decoder
         self.decoder = EncDecCTCModelBPE.from_config_dict(decoder_config)
         self._wer = WERBPE(tokenizer=self.tokenizer, batch_dim_index=0, use_cer=False, ctc_decode=True)
 
         # Update config
-        self._cfg.decoder.params = decoder_config
+        OmegaConf.set_struct(self._cfg.decoder, False)
+        self._cfg.decoder = decoder_config
+        OmegaConf.set_struct(self._cfg.decoder, True)
+
         logging.info(f"Changed tokenizer to {self.decoder.vocabulary} vocabulary.")
 
 

diff --git a/nemo/collections/asr/models/ctc_models.py b/nemo/collections/asr/models/ctc_models.py
@@ -19,7 +19,7 @@
 from typing import Dict, List, Optional, Union
 
 import torch
-from omegaconf import DictConfig
+from omegaconf import DictConfig, OmegaConf
 from pytorch_lightning import Trainer
 
 from nemo.collections.asr.data.audio_to_text import AudioToCharDataset, TarredAudioToCharDataset
@@ -154,7 +154,10 @@ def change_vocabulary(self, new_vocabulary: List[str]):
             self._wer = WER(vocabulary=self.decoder.vocabulary, batch_dim_index=0, use_cer=False, ctc_decode=True)
 
             # Update config
-            self._cfg.decoder.params = new_decoder_config
+            OmegaConf.set_struct(self._cfg.decoder, False)
+            self._cfg.decoder = new_decoder_config
+            OmegaConf.set_struct(self._cfg.decoder, True)
+
             logging.info(f"Changed decoder to output to {self.decoder.vocabulary} vocabulary.")
 
     def _setup_dataloader_from_config(self, config: Optional[Dict]):

diff --git a/nemo/core/classes/modelPT.py b/nemo/core/classes/modelPT.py
@@ -78,7 +78,7 @@ def __init__(self, cfg: DictConfig, trainer: Trainer = None):
 
         self._cfg = config
 
-        self.save_hyperparameters(self._cfg)
+        self.save_hyperparameters(OmegaConf.to_container(self._cfg, resolve=True))
         self._train_dl = None
         self._validation_dl = None
         self._test_dl = None
@@ -231,6 +231,7 @@ def load_from_checkpoint(
         Loads ModelPT from checkpoint, with some maintenance of restoration.
         For documentation, please refer to LightningModule.load_from_checkpoin() documentation.
         """
+        # TODO (@titu1994): When PTL 0.9+ is supported, add `strict=False` flag to constructor
         checkpoint = None
         try:
             cls.__set_model_restore_state(is_being_restored=True)

diff --git a/requirements/requirements.txt b/requirements/requirements.txt
@@ -1,12 +1,12 @@
 numpy>=1.18.2
 onnx>=1.7.0
-pytorch-lightning>=0.8.5
+pytorch-lightning==0.8.5
 python-dateutil
 torch
 wget
 wrapt
 ruamel.yaml
 scikit-learn
-omegaconf==2.0.1rc11
-hydra-core==1.0.0rc3
+omegaconf==2.0.1rc12
+hydra-core==1.0.0rc4
 transformers>=2.11.0