Spaces:
Sleeping
Sleeping
File size: 563 Bytes
f4b0576 5525b46 f4b0576 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "s3nh/mamba-gpt-3b-v3-GGML"
def load():
global model
global tokenizer
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
def generate(input_text):
input_ids = tokenizer.encode(input_text, return_tensors="pt")
output_ids = model.generate(input_ids, no_repeat_ngram_size=2, max_new_tokens=200, num_beams=2, eos_token_id=tokenizer.eos_token_id)
return tokenizer.decode(output_ids[0], skip_special_tokens=True) |