Update app/main.py
Browse files- app/main.py +15 -30
app/main.py
CHANGED
@@ -1,50 +1,35 @@
|
|
1 |
import PIL
|
2 |
from fastapi import FastAPI, File, UploadFile
|
3 |
from pydantic import BaseModel
|
4 |
-
from fastapi.responses import JSONResponse
|
5 |
from utils.model_func import class_id_to_label, load_model, transform_image
|
6 |
|
7 |
-
model = None
|
8 |
app = FastAPI()
|
9 |
|
10 |
|
|
|
11 |
class ImageClass(BaseModel):
|
12 |
prediction: str
|
13 |
|
14 |
-
|
15 |
-
text: str
|
16 |
-
|
17 |
-
|
18 |
@app.on_event("startup")
|
19 |
-
|
20 |
global model
|
21 |
-
# Здесь используйте функцию из utils.model_func для загрузки модели
|
22 |
model = load_model()
|
23 |
|
|
|
|
|
|
|
24 |
|
25 |
-
# @app.post('/classify')
|
26 |
-
# async def classify_image(file: UploadFile = File(...)):
|
27 |
-
# # Здесь используйте функцию из utils.model_func для классификации изображения
|
28 |
-
# image_bytes = await file.read()
|
29 |
-
# prediction = transform_image(image_bytes, model)
|
30 |
-
# return {"prediction": prediction}
|
31 |
|
32 |
@app.post('/classify')
|
33 |
-
|
34 |
-
# Use the function from utils.model_func to classify the image
|
35 |
image = PIL.Image.open(file.file)
|
36 |
adapted_image = transform_image(image)
|
37 |
-
pred_index = model(adapted_image.
|
38 |
-
imagenet_class = class_id_to_label(pred_index)
|
39 |
-
response = ImageClass(
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
@app.post('/clf_text')
|
46 |
-
async def classify_text(text_data: TextClass):
|
47 |
-
# Здесь используйте функцию из utils.model_func для классификации текста
|
48 |
-
text = text_data.text
|
49 |
-
prediction = class_id_to_label(text, model)
|
50 |
-
return {"prediction": prediction}
|
|
|
1 |
import PIL
|
2 |
from fastapi import FastAPI, File, UploadFile
|
3 |
from pydantic import BaseModel
|
|
|
4 |
from utils.model_func import class_id_to_label, load_model, transform_image
|
5 |
|
6 |
+
model = None
|
7 |
app = FastAPI()
|
8 |
|
9 |
|
10 |
+
# Create class of answer: only class name
|
11 |
class ImageClass(BaseModel):
|
12 |
prediction: str
|
13 |
|
14 |
+
# Load model at startup
|
|
|
|
|
|
|
15 |
@app.on_event("startup")
|
16 |
+
def startup_event():
|
17 |
global model
|
|
|
18 |
model = load_model()
|
19 |
|
20 |
+
@app.get('/')
|
21 |
+
def return_info():
|
22 |
+
return 'Hello FastAPI'
|
23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
@app.post('/classify')
|
26 |
+
def classify(file: UploadFile = File(...)):
|
|
|
27 |
image = PIL.Image.open(file.file)
|
28 |
adapted_image = transform_image(image)
|
29 |
+
pred_index = model(adapted_image.unsqueeze(0)).detach().cpu().numpy().argmax()
|
30 |
+
imagenet_class = class_id_to_label(pred_index)
|
31 |
+
response = ImageClass(
|
32 |
+
prediction=imagenet_class
|
33 |
+
)
|
34 |
+
|
35 |
+
return response
|
|
|
|
|
|
|
|
|
|
|
|
|
|