参考：
cola
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/cola')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
1063 |
'train' |
8551 |
'validation' |
1043 |
- 特征：
{
"sentence": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"unacceptable",
"acceptable"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
sst2
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/sst2')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
1821 |
'train' |
67349 |
'validation' |
872 |
- 特征：
{
"sentence": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"negative",
"positive"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
mrpc
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/mrpc')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
1725 |
'train' |
3668 |
'validation' |
408 |
- 特征：
{
"sentence1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"sentence2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"not_equivalent",
"equivalent"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
qqp
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/qqp')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
390965 |
'train' |
363846 |
'validation' |
40430 |
- 特征：
{
"question1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"question2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"not_duplicate",
"duplicate"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
stsb
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/stsb')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
1379 |
'train' |
5749 |
'validation' |
1500 |
- 特征：
{
"sentence1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"sentence2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"dtype": "float32",
"id": null,
"_type": "Value"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
mnli
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/mnli')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test_matched' |
9796 |
'test_mismatched' |
9847 |
'train' |
392702 |
'validation_matched' |
9815 |
'validation_mismatched' |
9832 |
- 特征：
{
"premise": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"hypothesis": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 3,
"names": [
"entailment",
"neutral",
"contradiction"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
mnli_mismatched
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/mnli_mismatched')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
9847 |
'validation' |
9832 |
- 特征：
{
"premise": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"hypothesis": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 3,
"names": [
"entailment",
"neutral",
"contradiction"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
mnli_matched
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/mnli_matched')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
9796 |
'validation' |
9815 |
- 特征：
{
"premise": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"hypothesis": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 3,
"names": [
"entailment",
"neutral",
"contradiction"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
qnli
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/qnli')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
5463 |
'train' |
104743 |
'validation' |
5463 |
- 特征：
{
"question": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"sentence": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"entailment",
"not_entailment"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
rte
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/rte')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
3000 |
'train' |
2490 |
'validation' |
277 |
- 特征：
{
"sentence1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"sentence2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"entailment",
"not_entailment"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
wnli
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/wnli')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
146 |
'train' |
635 |
'validation' |
71 |
- 特征：
{
"sentence1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"sentence2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
"not_entailment",
"entailment"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
ax
使用以下命令在 TFDS 中加载此数据集：
ds = tfds.load('huggingface:glue/ax')
- 说明：
GLUE, the General Language Understanding Evaluation benchmark
(https://gluebenchmark.com/) is a collection of resources for training,
evaluating, and analyzing natural language understanding systems.
- 许可：无已知许可
- 版本：1.0.0
- 拆分：
| 拆分 | 样本 |
|---|---|
'test' |
1104 |
- 特征：
{
"premise": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"hypothesis": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 3,
"names": [
"entailment",
"neutral",
"contradiction"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}