Spaces:
Sleeping
Sleeping
File size: 2,482 Bytes
ae467e7 |
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 |
from transformers import ViTImageProcessor, ViTForImageClassification
from PIL import Image
import os
import torch
from .logging import get_logger
logger = get_logger(__name__)
class ViTBrainTumorClassifier:
CLASS_LABELS = {0: "Glioma", 1: "Meningioma", 2: "No Tumor", 3: "Pituitary"}
def __init__(self, device: str = "cpu", model_name: str = "codeby-hp/vit-brain-tumor-classifier"):
self.device = device
self.model_name = model_name
self.model = None
self.processor = None
self._load_model()
def _load_model(self):
try:
logger.info(f"Downloading model from HuggingFace Hub: {self.model_name}")
# Download from HuggingFace Hub
self.processor = ViTImageProcessor.from_pretrained(self.model_name)
self.model = ViTForImageClassification.from_pretrained(self.model_name)
self.model.to(self.device)
self.model.eval()
logger.info(f"Model loaded successfully on {self.device}")
except Exception as e:
logger.error(f"Model loading failed: {e}")
raise
def predict(self, image_path: str) -> dict:
try:
image = Image.open(image_path).convert("RGB")
inputs = self.processor(images=image, return_tensors="pt")
inputs = {k: v.to(self.device) for k, v in inputs.items()}
with torch.no_grad():
outputs = self.model(**inputs)
logits = outputs.logits
probabilities = torch.nn.functional.softmax(logits, dim=-1)
predicted_class = torch.argmax(probabilities, dim=-1).item()
confidence = probabilities[0, predicted_class].item()
result = {
"predicted_class": self.CLASS_LABELS.get(predicted_class, "Unknown"),
"confidence": round(confidence * 100, 2),
"all_predictions": {
self.CLASS_LABELS[i]: round(probabilities[0, i].item() * 100, 2)
for i in range(len(self.CLASS_LABELS))
}
}
logger.info(f"Prediction: {result['predicted_class']} ({result['confidence']}%)")
return result
except Exception as e:
logger.error(f"Prediction error: {e}")
raise |