Hi,
I am try to run and fine-tune the models for NERT base models in NER task.
I downloaded 3 tsv file from the paper of the models for English. en - Google Drive
I run the code to get the data.
from datasets import load_dataset
train = load_dataset('train_en.tsv')
The train.tsv and SystemA.ipynb are in the same folder.
The error is:
---------------------------------------------------------------------------
FileNotFoundError Traceback (most recent call last)
SystemA.ipynb Cell 9 line 2
1 from datasets import load_dataset
----> 2 train = load_dataset('train_en.tsv')
3 train
File c:\Users\cnp\Anaconda3\envs\ML2023\Lib\site-packages\datasets\load.py:2128, in load_dataset(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, **config_kwargs)
2123 verification_mode = VerificationMode(
2124 (verification_mode or VerificationMode.BASIC_CHECKS) if not save_infos else VerificationMode.ALL_CHECKS
2125 )
2127 # Create a dataset builder
-> 2128 builder_instance = load_dataset_builder(
2129 path=path,
2130 name=name,
2131 data_dir=data_dir,
2132 data_files=data_files,
2133 cache_dir=cache_dir,
2134 features=features,
2135 download_config=download_config,
2136 download_mode=download_mode,
2137 revision=revision,
2138 token=token,
2139 storage_options=storage_options,
2140 **config_kwargs,
2141 )
2143 # Return iterable dataset in case of streaming
2144 if streaming:
File c:\Users\cnp\Anaconda3\envs\ML2023\Lib\site-packages\datasets\load.py:1814, in load_dataset_builder(path, name, data_dir, data_files, cache_dir, features, download_config, download_mode, revision, token, use_auth_token, storage_options, **config_kwargs)
1812 download_config = download_config.copy() if download_config else DownloadConfig()
1813 download_config.storage_options.update(storage_options)
-> 1814 dataset_module = dataset_module_factory(
1815 path,
1816 revision=revision,
1817 download_config=download_config,
1818 download_mode=download_mode,
1819 data_dir=data_dir,
1820 data_files=data_files,
1821 )
1822 # Get dataset builder class from the processing script
1823 builder_kwargs = dataset_module.builder_kwargs
File c:\Users\cnp\Anaconda3\envs\ML2023\Lib\site-packages\datasets\load.py:1507, in dataset_module_factory(path, revision, download_config, download_mode, dynamic_modules_path, data_dir, data_files, **download_kwargs)
1505 raise e1 from None
1506 if isinstance(e1, FileNotFoundError):
-> 1507 raise FileNotFoundError(
1508 f"Couldn't find a dataset script at {relative_to_absolute_path(combined_path)} or any data file in the same directory. "
1509 f"Couldn't find '{path}' on the Hugging Face Hub either: {type(e1).__name__}: {e1}"
1510 ) from None
1511 raise e1 from None
1512 else:
FileNotFoundError: Couldn't find a dataset script at train_en.tsv\train_en.tsv.py or any data file in the same directory. Couldn't find 'train_en.tsv' on the Hugging Face Hub either: FileNotFoundError: Dataset 'train_en.tsv' doesn't exist on the Hub. If the repo is private or gated, make sure to log in with `huggingface-cli login`.