File size: 1,708 Bytes
16917d3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 |
from typing import Any, Dict
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, pipeline
dtype = torch.bfloat16 if torch.cuda.get_device_capability()[0] == 8 else torch.float16
class EndpointHandler:
def __init__(self, path=""):
quantization_config = BitsAndBytesConfig(load_in_8bit=True)
tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code = True)
model = AutoModelForCausalLM.from_pretrained(
path,
return_dict = True,
device_map = "auto",
torch_dtype = dtype,
trust_remote_code = True,
quantization_config=quantization_config
)
gen_config = model.generation_config
gen_config.max_new_tokens = 256
gen_config.num_return_sequences = 1
gen_config.pad_token_id = tokenizer.eos_token_id
gen_config.eos_token_id = tokenizer.eos_token_id
self.generation_config = gen_config
self.pipeline = pipeline(
'text-generation', model=model, tokenizer=tokenizer
)
def __call__(self, data: Dict[dict, Any]) -> Dict[str, Any]:
prompt = data.pop("inputs", data)
instruction = "Create a list of chords,a corresponding scale to improve with, title, and style along with an example in ABC notation based on this input in JSON format."
full_prompt = f"""<s>
### Instruction:
{instruction}
### Input:
{prompt}
### Response:
"""
result = self.pipeline(full_prompt, generation_config = self.generation_config)[0]
return result |