t5-small-spoken-typo / train /dataset_info.json
willwade's picture
retrained with new data from opensubs. qwerty subs
5d53663
raw
history blame
1.29 kB
{
"builder_name": "csv",
"citation": "",
"config_name": "default",
"dataset_name": "csv",
"dataset_size": 472235292,
"description": "",
"download_checksums": {
"/home/ubuntu/wwdrive2/14March/clean_train.csv": {
"num_bytes": 369392984,
"checksum": null
},
"/home/ubuntu/wwdrive2/14March/clean_eval.csv": {
"num_bytes": 71786407,
"checksum": null
}
},
"download_size": 441179391,
"features": {
"input_ids": {
"feature": {
"dtype": "int32",
"_type": "Value"
},
"_type": "Sequence"
},
"attention_mask": {
"feature": {
"dtype": "int8",
"_type": "Value"
},
"_type": "Sequence"
},
"labels": {
"feature": {
"dtype": "int64",
"_type": "Value"
},
"_type": "Sequence"
}
},
"homepage": "",
"license": "",
"size_in_bytes": 913414683,
"splits": {
"train": {
"name": "train",
"num_bytes": 394316099,
"num_examples": 4507525,
"dataset_name": "csv"
},
"eval": {
"name": "eval",
"num_bytes": 77919193,
"num_examples": 1127410,
"dataset_name": "csv"
}
},
"version": {
"version_str": "0.0.0",
"major": 0,
"minor": 0,
"patch": 0
}
}