Downloads · 30 days
18
3% of all-time downloads
gpt2-paraphrase
ihgn/gpt2-paraphrase
gpt2-paraphrase is a text generation model from ihgn. Use it when you need the model to write or continue text. It is set up for transformers.
def paraphrase( question, numbeams=5, numbeamgroups=5, numreturnsequences=1, repetitionpenalty=10.0, diversitypenalty=3.0, norepeatngramsize=2, temperature=0.7, maxlength=128 ): inputids = tokenizer( f'paraphrase: {qu…
All-time downloads
645
Public
Repo size
1.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin892 MB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- t5
- Access
- Public
- Created
- Jun 11, 2023
- Updated
- Aug 18, 2023
- SHA
- 2b7c044b
Try a prompt
Model card
From the Hugging Face model README
def paraphrase( question, num_beams=5, num_beam_groups=5, num_return_sequences=1, repetition_penalty=10.0, diversity_penalty=3.0, no_repeat_ngram_size=2, temperature=0.7, max_length=128 ): input_ids = tokenizer( f'paraphrase: {question}', return_tensors="pt", padding="longest", max_length=max_length, truncation=True, ).input_ids
outputs = model.generate(
input_ids, temperature=temperature, repetition_penalty=repetition_penalty,
num_return_sequences=num_return_sequences, no_repeat_ngram_size=no_repeat_ngram_size,
num_beams=num_beams, num_beam_groups=num_beam_groups,
max_length=max_length, diversity_penalty=diversity_penalty
)
res = tokenizer.batch_decode(outputs, skip_special_tokens=True)
return res
- Task
- Text Generation
- Library
- transformers
- Type
- t5
- Languages
- en
- Created
- Jun 11, 2023
- Updated
- Aug 18, 2023