Problem while pretraining on custom dataset

[mujhenahiata]

[mujhenahiata]

user@user:/opt/vakyansh-wav2vec2-experimentation/scripts/pretraining$ bash start_pretraining_base.sh ** Directory to code is: /opt/vakyansh-wav2vec2-experimentation ** Config path is: /opt/vakyansh-wav2vec2-experimentation/config/wav2vec2_base_librispeech.yaml ** Data path is: /opt/vakyansh-wav2vec2-experimentation/data/pretraining ** Checkpoint will be saved at: /opt/vakyansh-wav2vec2-experimentation/checkpoints/pretraining ** Logs will be saved at: /opt/vakyansh-wav2vec2-experimentation/logs/pretraining ** Update frequency is: 16Traceback (most recent call last): File "/user/.local/lib/python3.8/site-packages/omegaconf/omegaconf.py", line 543, in select _root, _last_key, value = cfg._select_impl( File "/user/.local/lib/python3.8/site-packages/omegaconf/base.py", line 279, in _select_impl assert ret is None or isinstance(ret, Container) AssertionError

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "/opt/conda/bin/fairseq-hydra-train", line 33, in sys.exit(load_entry_point('fairseq', 'console_scripts', 'fairseq-hydra-train')()) File "/opt/vakyansh-wav2vec2-experimentation/fairseq/fairseq_cli/hydra_train.py", line 91, in cli_main hydra_main() File "/user/.local/lib/python3.8/site-packages/hydra/main.py", line 32, in decorated_main _run_hydra( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/utils.py", line 346, in _run_hydra run_and_report( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/utils.py", line 201, in run_and_report raise ex File "/user/.local/lib/python3.8/site-packages/hydra/_internal/utils.py", line 198, in run_and_report return func() File "/user/.local/lib/python3.8/site-packages/hydra/_internal/utils.py", line 347, in lambda: hydra.run( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 100, in run cfg = self.compose_config( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 507, in compose_config cfg = self.config_loader.load_configuration( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 151, in load_configuration return self._load_configuration( File "/user/.local/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 277, in _load_configuration ConfigLoaderImpl._apply_overrides_to_config(config_overrides, cfg) File "/user/.local/lib/python3.8/site-packages/hydra/_internal/config_loader_impl.py", line 502, in _apply_overrides_to_config if OmegaConf.select( File "/user/.local/lib/python3.8/site-packages/omegaconf/omegaconf.py", line 552, in select format_and_raise(node=cfg, key=key, value=None, cause=e, msg=str(e)) File "/user/.local/lib/python3.8/site-packages/omegaconf/_utils.py", line 694, in format_and_raise _raise(ex, cause) File "/user/.local/lib/python3.8/site-packages/omegaconf/_utils.py", line 610, in _raise raise ex # set end OC_CAUSE=1 for full backtrace AssertionError: full_key: task.data reference_type=Optional[FairseqConfig] object_type=FairseqConfig

i prepared the data and then ran the pretraining part. got this error

model type used: BASE pretraining

[harveenchadha]

Can you please provide the fairseq version you are using?

[mujhenahiata]

I was not working on it for a while, now that I have resumed I came across another problem. there is another problem I came across, the flashlight has a separate class for processing text and it is in a seperate branch. how you install the flashlight library.