|
from typing import Dict, List, Any |
|
from ctransformers import AutoModelForCausalLM |
|
|
|
|
|
class EndpointHandler(): |
|
def __init__(self, path=""): |
|
model_id = "djomo/MISTRALllux1000-7b-v5-GGUF" |
|
model_file="mistralllux1000-7b-v5.gguf.q5_k_m.bin" |
|
config = {'context_length' : 3048,'max_new_tokens': 856, 'repetition_penalty': 1.1,'temperature': 0.1, 'stream': True} |
|
llm = AutoModelForCausalLM.from_pretrained( |
|
model_id, |
|
model_file=model_file, |
|
model_type="mistral", |
|
gpu_layers=130, |
|
**config |
|
) |
|
self.pipeline = llm |
|
|
|
|
|
def __call__(self, data: Any) -> List[List[Dict[str, float]]]: |
|
""" |
|
Args: |
|
data (:obj:): |
|
includes the input data and the parameters for the inference. |
|
Return: |
|
A :obj:`list`:. The object returned should be a list of one list like [[{"label": 0.9939950108528137}]] containing : |
|
- "label": A string representing what the label/class is. There can be multiple labels. |
|
- "score": A score between 0 and 1 describing how confident the model is for this label/class. |
|
""" |
|
inputs = data.pop("inputs", data) |
|
parameters = data.pop("parameters", None) |
|
|
|
|
|
if parameters is not None: |
|
prediction = self.pipeline(inputs, stream=False) |
|
else: |
|
prediction = self.pipeline(inputs, stream=False) |
|
|
|
return prediction |