The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

How to loading dataset?

from datasets import load_dataset
datasets = load_dataset("minskiter/msra_dev",save_infos=True)
train,test = datasets['train'],datasets['test']
# convert label to str
print(train.features['labels'].feature.int2str(0))

Force update

from datasets import load_dataset
datasets = load_dataset("minskiter/msra_dev", download_mode="force_redownload")

Fit your train

def transform(example):
  # edit example here
  return example
for key in datasets:
  datasets[key] = datasets.map(transform)
Downloads last month
6