Spaces:
Running
Running
File size: 4,007 Bytes
1ce6f57 dd452ee 1ce6f57 dd452ee 1ce6f57 dd452ee 1ce6f57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 | from fastapi import FastAPI, File, UploadFile, Query, HTTPException
from fastapi.middleware.cors import CORSMiddleware
from plantdoc_predictor import Predictor
from PIL import Image
import io
app = FastAPI(title="PlantDoc API", version="1.0.0")
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_methods=["*"],
allow_headers=["*"],
)
MODELS = [
{"name": "mobilenetv2_v1", "accuracy": 96.8, "description": "Mobile-optimized, fastest inference"},
{"name": "densenet169_v1", "accuracy": 99.68, "description": "Best classical CNN"},
{"name": "rec_add_attention_v1", "accuracy": 99.70, "description": "Best overall — Recursive Additive Attention (research model)"},
{"name": "convnext_small_v1", "accuracy": 99.50, "description": "High accuracy + efficient"},
{"name": "convnext_tiny_v1", "accuracy": 99.23, "description": "Lightweight ConvNeXt"},
{"name": "convnext_base_v1", "accuracy": 99.10, "description": "ConvNeXt base"},
{"name": "swin_tiny_patch4_window7", "accuracy": 99.10, "description": "Lightweight Swin Transformer"},
{"name": "swin_base_patch4_window7", "accuracy": 99.10, "description": "Swin Transformer"},
{"name": "vit_tiny_16_v1", "accuracy": 99.10, "description": "Lightweight Vision Transformer"},
{"name": "vit_small_16_v1", "accuracy": 99.10, "description": "Small Vision Transformer"},
{"name": "vit_base_16_v1", "accuracy": 99.10, "description": "Vision Transformer base"},
{"name": "vit_large_16_v1", "accuracy": 99.10, "description": "Vision Transformer large"},
{"name": "densenet121_v1", "accuracy": 98.68, "description": "Efficient DenseNet"},
{"name": "vgg19_v1", "accuracy": 98.98, "description": "Deep VGG"},
{"name": "inceptionv3_v1", "accuracy": 98.20, "description": "Multi-scale InceptionV3"},
{"name": "resnet50_v1", "accuracy": 97.80, "description": "Residual learning CNN"},
{"name": "efficientnetb50_v1", "accuracy": 97.80, "description": "EfficientNet balanced"},
{"name": "densenet210_v1", "accuracy": 97.00, "description": "Deep DenseNet"},
{"name": "vgg16_v1", "accuracy": 96.80, "description": "Classic CNN"},
{"name": "alexnet_v1", "accuracy": 92.80, "description": "Baseline CNN"},
]
MODEL_NAMES = [m["name"] for m in MODELS]
_predictor_cache: dict[str, Predictor] = {}
def get_predictor(model_name: str) -> Predictor:
if model_name not in _predictor_cache:
_predictor_cache[model_name] = Predictor(model_name=model_name, verbose=False)
return _predictor_cache[model_name]
@app.on_event("startup")
def warmup():
get_predictor("mobilenetv2_v1")
@app.get("/health")
def health():
return {"status": "ok", "loaded_models": list(_predictor_cache.keys())}
@app.get("/models")
def list_models():
return {"models": MODELS}
@app.post("/predict")
async def predict(
file: UploadFile = File(...),
model: str = Query(default="mobilenetv2_v1"),
):
if model not in MODEL_NAMES:
raise HTTPException(status_code=400, detail=f"Unknown model '{model}'. Call /models for the list.")
content = await file.read()
if len(content) == 0:
raise HTTPException(status_code=400, detail="Empty file uploaded.")
try:
image = Image.open(io.BytesIO(content)).convert("RGB")
except Exception:
raise HTTPException(status_code=400, detail="Invalid image file.")
try:
predictor = get_predictor(model)
result = predictor.predict(image)
plant, disease = (result["label"].split("___") + [""])[:2]
result["plant"] = plant.replace("_", " ")
result["disease"] = disease.replace("_", " ")
result["is_healthy"] = "healthy" in result["label"].lower()
return result
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
|