Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,7 +1,56 @@
|
|
1 |
-
from fastapi import FastAPI
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
app = FastAPI()
|
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
@app.get("/")
|
6 |
-
def
|
7 |
-
return {"
|
|
|
1 |
+
# from fastapi import FastAPI
|
2 |
+
|
3 |
+
# app = FastAPI()
|
4 |
+
|
5 |
+
# @app.get("/")
|
6 |
+
# def greet_json():
|
7 |
+
# return {"Hello": "World!"}
|
8 |
+
|
9 |
+
from fastapi import FastAPI, HTTPException
|
10 |
+
from pydantic import BaseModel
|
11 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
12 |
+
import torch
|
13 |
|
14 |
app = FastAPI()
|
15 |
|
16 |
+
# Check if CUDA is available
|
17 |
+
if torch.cuda.is_available():
|
18 |
+
device = torch.device("cuda:0")
|
19 |
+
else:
|
20 |
+
device = torch.device("cpu")
|
21 |
+
|
22 |
+
# Load the tokenizer and model
|
23 |
+
tokenizer = AutoTokenizer.from_pretrained("kmack/malicious-url-detection")
|
24 |
+
model = AutoModelForSequenceClassification.from_pretrained("kmack/malicious-url-detection")
|
25 |
+
model = model.to(device)
|
26 |
+
|
27 |
+
# Define the request model
|
28 |
+
class URLRequest(BaseModel):
|
29 |
+
url: str
|
30 |
+
|
31 |
+
# Prediction function
|
32 |
+
def get_prediction(input_text: str) -> dict:
|
33 |
+
label2id = model.config.label2id
|
34 |
+
inputs = tokenizer(input_text, return_tensors='pt', truncation=True)
|
35 |
+
inputs = inputs.to(device)
|
36 |
+
outputs = model(**inputs)
|
37 |
+
logits = outputs.logits
|
38 |
+
sigmoid = torch.nn.Sigmoid()
|
39 |
+
probs = sigmoid(logits.squeeze().cpu())
|
40 |
+
probs = probs.detach().numpy()
|
41 |
+
for i, k in enumerate(label2id.keys()):
|
42 |
+
label2id[k] = probs[i]
|
43 |
+
label2id = {k: float(v) for k, v in sorted(label2id.items(), key=lambda item: item[1].item(), reverse=True)}
|
44 |
+
return label2id
|
45 |
+
|
46 |
+
# Define the API endpoint for URL prediction
|
47 |
+
@app.post("/predict")
|
48 |
+
async def predict(url_request: URLRequest):
|
49 |
+
url_to_check = url_request.url
|
50 |
+
result = get_prediction(url_to_check)
|
51 |
+
return {"prediction": result}
|
52 |
+
|
53 |
+
# Health check endpoint
|
54 |
@app.get("/")
|
55 |
+
async def read_root():
|
56 |
+
return {"message": "API is up and running"}
|