Upload 10 files

app/main.py

@@ -0,0 +1,50 @@
+import PIL
+from fastapi import FastAPI, File, UploadFile
+from pydantic import BaseModel
+from fastapi.responses import JSONResponse
+from utils.model_func import class_id_to_label, load_model, transform_image
+
+model = None
+app = FastAPI()
+
+
+class ImageClass(BaseModel):
+    prediction: str
+
+class TextClass(BaseModel):
+    text: str
+
+
+@app.on_event("startup")
+async def startup_event():
+    global model
+    # Здесь используйте функцию из utils.model_func для загрузки модели
+    model = load_model()
+
+
+# @app.post('/classify')
+# async def classify_image(file: UploadFile = File(...)):
+#     # Здесь используйте функцию из utils.model_func для классификации изображения
+#     image_bytes = await file.read()
+#     prediction = transform_image(image_bytes, model)
+#     return {"prediction": prediction}
+
+@app.post('/classify')
+async def classify_image(file: UploadFile = File(...)):
+    # Use the function from utils.model_func to classify the image
+    image = PIL.Image.open(file.file)
+    adapted_image = transform_image(image)
+    pred_index = model(adapted_image.unsqeeze(0).detach().cpu().numpy().argmax())
+    imagenet_class = class_id_to_label(pred_index)     
+    response = ImageClass(prediction=imagenet_class)   
+    return response
+
+          
+                       
+
+@app.post('/clf_text')
+async def classify_text(text_data: TextClass):
+    # Здесь используйте функцию из utils.model_func для классификации текста
+    text = text_data.text
+    prediction = class_id_to_label(text, model)
+    return {"prediction": prediction}

front/app.py

@@ -0,0 +1,26 @@
+import streamlit as st
+import requests
+import json
+
+def main():
+
+    st.title("FastAPI - Streamlit Integration")
+
+    # Эндпоинт для классификации текста
+    text_input = st.text_input("Enter text for classification:")
+    if st.button("Classify Text"):
+        response = requests.post("http://127.0.0.1:8000/clf_text", json={"text": text_input})
+        result = response.json()
+        st.success(f"Classification result: {result['prediction']}")
+
+    # Эндпоинт для классификации изображений
+    image_path = st.file_uploader("Upload an image for classification:", type=["jpg", "png"])
+    if image_path is not None:
+        if st.button("Classify Image"):
+            files = {"file": image_path.read()}
+            response = requests.post("http://127.0.0.1:8000/classify", files=files)
+            result = response.json()
+            st.success(f"Classification result: {result['prediction']}")
+
+if __name__ == '__main__':
+    main()

requirements.txt

@@ -0,0 +1,7 @@
+fastapi==0.105.0
+uvicorn==0.24.0.post1
+streamlit==1.29.0
+torch== 2.1.2
+torchvision==0.16.2
+Pillow==10.1.0
+python-multipart==0.0.6

utils/model_func.py

@@ -0,0 +1,27 @@
+import torch
+from torchvision import models, transforms
+from PIL import Image
+import json
+
+def load_classes():
+    with open('utils/imagenet-simple-labels.json') as f:
+        labels = json.load(f)
+    return labels
+
+def class_id_to_label(i):
+    labels = load_classes()
+    return labels[i]
+
+def load_model():
+    model = models.mobilenet_v2(pretrained=True)
+    model.eval()
+    return model
+
+def transform_image(img):
+    transform = transforms.Compose([
+        transforms.Resize((224, 224)),
+        transforms.ToTensor(),
+        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+    ])
+    return transform(img)
+