Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,9 +1,15 @@
|
|
| 1 |
from fastapi import FastAPI, File, UploadFile
|
| 2 |
-
from fastapi.responses import JSONResponse
|
| 3 |
from transformers import AutoProcessor, AutoModelForVision2Seq
|
| 4 |
from PIL import Image
|
| 5 |
import io
|
| 6 |
import torch
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
|
| 8 |
app = FastAPI()
|
| 9 |
|
|
@@ -13,17 +19,31 @@ model = AutoModelForVision2Seq.from_pretrained("HuggingFaceTB/SmolVLM-256M-Instr
|
|
| 13 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
| 14 |
model = model.to(device)
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
@app.post("/predict")
|
| 17 |
async def predict_gender(file: UploadFile = File(...)):
|
| 18 |
image_bytes = await file.read()
|
| 19 |
image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
|
| 20 |
|
| 21 |
prompt = "Is the person on this ID male or female?"
|
| 22 |
-
|
| 23 |
-
# Prepare inputs
|
| 24 |
inputs = processor(images=image, text=prompt, return_tensors="pt").to(device)
|
| 25 |
-
|
| 26 |
-
# Generate response
|
| 27 |
outputs = model.generate(**inputs, max_new_tokens=32)
|
| 28 |
answer = processor.batch_decode(outputs, skip_special_tokens=True)[0].strip()
|
| 29 |
|
|
|
|
| 1 |
from fastapi import FastAPI, File, UploadFile
|
| 2 |
+
from fastapi.responses import JSONResponse, HTMLResponse
|
| 3 |
from transformers import AutoProcessor, AutoModelForVision2Seq
|
| 4 |
from PIL import Image
|
| 5 |
import io
|
| 6 |
import torch
|
| 7 |
+
import os
|
| 8 |
+
|
| 9 |
+
# Make sure cache is writable
|
| 10 |
+
os.environ["HF_HOME"] = "/app/cache"
|
| 11 |
+
os.environ["TRANSFORMERS_CACHE"] = "/app/cache"
|
| 12 |
+
os.environ["HF_HUB_CACHE"] = "/app/cache/hub"
|
| 13 |
|
| 14 |
app = FastAPI()
|
| 15 |
|
|
|
|
| 19 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
| 20 |
model = model.to(device)
|
| 21 |
|
| 22 |
+
@app.get("/")
|
| 23 |
+
def home():
|
| 24 |
+
return {"message": "API is running. Use POST /predict with an image, or visit /upload to test in browser."}
|
| 25 |
+
|
| 26 |
+
@app.get("/upload", response_class=HTMLResponse)
|
| 27 |
+
def upload_form():
|
| 28 |
+
return """
|
| 29 |
+
<html>
|
| 30 |
+
<body>
|
| 31 |
+
<h2>Upload an ID Image</h2>
|
| 32 |
+
<form action="/predict" enctype="multipart/form-data" method="post">
|
| 33 |
+
<input name="file" type="file">
|
| 34 |
+
<input type="submit" value="Upload">
|
| 35 |
+
</form>
|
| 36 |
+
</body>
|
| 37 |
+
</html>
|
| 38 |
+
"""
|
| 39 |
+
|
| 40 |
@app.post("/predict")
|
| 41 |
async def predict_gender(file: UploadFile = File(...)):
|
| 42 |
image_bytes = await file.read()
|
| 43 |
image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
|
| 44 |
|
| 45 |
prompt = "Is the person on this ID male or female?"
|
|
|
|
|
|
|
| 46 |
inputs = processor(images=image, text=prompt, return_tensors="pt").to(device)
|
|
|
|
|
|
|
| 47 |
outputs = model.generate(**inputs, max_new_tokens=32)
|
| 48 |
answer = processor.batch_decode(outputs, skip_special_tokens=True)[0].strip()
|
| 49 |
|