Views
No views yet
| Model | CoLA | SST | MRPC | STS | QQP | MNLI | QNLI | RTE | Avg. |
|---|---|---|---|---|---|---|---|---|---|
| ELECTRA-Small-OWT(original) | 56.8 | 88.3 | 87.4 | 86.8 | 88.3 | 78.9 | 87.9 | 68.5 | 80.36 |
| ELECTRA-RoFormer-Small-OWT (this) | 55.76 | 90.45 | 87.3 | 86.64 | 89.61 | 81.17 | 88.85 | 62.71 | 80.31 |
1import torch
2from transformers import ElectraTokenizer,RoFormerModel
3tokenizer = ElectraTokenizer.from_pretrained("junnyu/roformer_small_discriminator")
4model = RoFormerModel.from_pretrained("junnyu/roformer_small_discriminator")
5inputs = tokenizer("Beijing is the capital of China.", return_tensors="pt")
6with torch.no_grad():
7 outputs = model(**inputs)
8 print(outputs[0].shape)