Create handler.py
Browse files- handler.py +50 -0
handler.py
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM, TextGenerationPipeline
|
2 |
+
import torch
|
3 |
+
|
4 |
+
def load_model(model_id):
|
5 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
6 |
+
model = AutoModelForCausalLM.from_pretrained(
|
7 |
+
model_id,
|
8 |
+
device_map="auto",
|
9 |
+
torch_dtype=torch.float16,
|
10 |
+
load_in_4bit=True
|
11 |
+
)
|
12 |
+
return model, tokenizer
|
13 |
+
|
14 |
+
class EndpointHandler:
|
15 |
+
def __init__(self, path=""):
|
16 |
+
self.model, self.tokenizer = load_model(path)
|
17 |
+
self.pipeline = TextGenerationPipeline(
|
18 |
+
model=self.model,
|
19 |
+
tokenizer=self.tokenizer,
|
20 |
+
max_new_tokens=512,
|
21 |
+
temperature=0.7,
|
22 |
+
top_p=0.95,
|
23 |
+
repetition_penalty=1.15,
|
24 |
+
do_sample=True
|
25 |
+
)
|
26 |
+
|
27 |
+
def __call__(self, data):
|
28 |
+
inputs = data.pop("inputs", data)
|
29 |
+
parameters = data.pop("parameters", {})
|
30 |
+
|
31 |
+
generation_kwargs = {
|
32 |
+
"max_new_tokens": 512,
|
33 |
+
"temperature": 0.7,
|
34 |
+
"top_p": 0.95,
|
35 |
+
"repetition_penalty": 1.15,
|
36 |
+
"do_sample": True
|
37 |
+
}
|
38 |
+
generation_kwargs.update(parameters)
|
39 |
+
|
40 |
+
if isinstance(inputs, str):
|
41 |
+
inputs = [inputs]
|
42 |
+
|
43 |
+
outputs = self.pipeline(
|
44 |
+
inputs,
|
45 |
+
**generation_kwargs
|
46 |
+
)
|
47 |
+
|
48 |
+
if len(outputs) == 1:
|
49 |
+
return {"generated_text": outputs[0]["generated_text"]}
|
50 |
+
return [{"generated_text": o["generated_text"]} for o in outputs]
|