Upload 3 files

BTD_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:269506d46e3854e3561379a7d4eb977194e66430b781b36bc56fab481e224f17
+size 178115794

Inference.py

@@ -0,0 +1,32 @@
+# inference.py
+
+import torch
+from io import BytesIO
+from PIL import Image
+from torchvision import transforms
+from TumorModel import TumorClassification
+
+# 1) Preprocessing pipeline
+_transform = transforms.Compose([
+    transforms.Grayscale(),
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize([0.5], [0.5]),
+])
+
+# 2) Load model once
+_model = TumorClassification()
+_model.load_state_dict(torch.load("BTD_model.pth", map_location="cpu"))
+_model.eval()
+
+def inference(image_bytes):
+    """
+    Hugging Face will pass the raw image bytes here.
+    Return {"label": <one of glioma, meningioma, notumor, pituitary>}.
+    """
+    img = Image.open(BytesIO(image_bytes)).convert("RGB")
+    x = _transform(img).unsqueeze(0)  # batch dimension
+    with torch.no_grad():
+        idx = torch.argmax(_model(x), dim=1).item()
+    labels = ["glioma", "meningioma", "notumor", "pituitary"]
+    return {"label": labels[idx]}

TumorModel.py

@@ -0,0 +1,34 @@
+import torch.nn as nn
+
+class TumorClassification(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.model = nn.Sequential(
+            nn.Conv2d(1, 16, kernel_size=3, stride=1, padding=1),
+            nn.ReLU(),
+            nn.MaxPool2d(2),
+            nn.Conv2d(16, 32, kernel_size=3, stride=1, padding=1),
+            nn.ReLU(),
+            nn.MaxPool2d(2),
+            nn.Flatten(),
+            nn.Linear(32 * 56 * 56, 128),
+            nn.ReLU(),
+            nn.Linear(128, 4)  # 4 classes: glioma, meningioma, notumor, pituitary
+        )
+
+    def forward(self, x):
+        return self.model(x)
+
+class GliomaStageModel(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.model = nn.Sequential(
+            nn.Linear(9, 128),
+            nn.ReLU(),
+            nn.Linear(128, 64),
+            nn.ReLU(),
+            nn.Linear(64, 4)  # 4 glioma stages
+        )
+
+    def forward(self, x):
+        return self.model(x)