Upload 5 files

.gitattributes

@@ -1,35 +1 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,12 +1,8 @@
----
-title: Corrobotv2
-emoji: 🐢
-colorFrom: green
-colorTo: yellow
-sdk: gradio
-sdk_version: 5.44.1
-app_file: app.py
-pinned: false
----
+# Corrosion Classifier (ViT-B/16 · CPU)
+Space Gradio pronto per Zero GPU. Carica `vit_b16_best.pth` nella root.
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+## Uso
+1. Crea uno Space Gradio su Hugging Face.
+2. Carica: app.py, requirements.txt, classes.json, vit_b16_best.pth.
+3. Runtime: CPU.
+4. Avvia.

app.py

@@ -0,0 +1,97 @@
+
+import os
+import json
+import torch
+import torchvision.transforms as T
+from PIL import Image
+import numpy as np
+import gradio as gr
+
+MODEL_WEIGHTS = os.getenv("MODEL_WEIGHTS_PATH", "vit_b16_best.pth")
+CLASSES_PATH = os.getenv("CLASSES_PATH", "classes.json")
+IMAGE_SIZE = 224
+
+def load_classes(path: str):
+    with open(path, "r", encoding="utf-8") as f:
+        return json.load(f)
+
+def build_transforms(img_size=IMAGE_SIZE):
+    return T.Compose([
+        T.Resize((img_size, img_size), interpolation=T.InterpolationMode.BICUBIC),
+        T.ToTensor(),
+        T.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+    ])
+
+def create_model(num_classes: int):
+    import timm
+    model = timm.create_model("vit_base_patch16_224", pretrained=False, num_classes=num_classes)
+    return model
+
+def load_model(weights_path: str, classes):
+    device = torch.device("cpu")
+    model = create_model(num_classes=len(classes))
+    state = torch.load(weights_path, map_location=device)
+    if isinstance(state, dict) and "state_dict" in state:
+        state = state["state_dict"]
+    cleaned = {k.replace("module.", "").replace("model.", ""): v for k, v in state.items()}
+    model.load_state_dict(cleaned, strict=False)
+    model.eval()
+    return model
+
+_model = None
+_classes = None
+_tfm = None
+
+def setup():
+    global _model, _classes, _tfm
+    if _classes is None:
+        _classes = load_classes(CLASSES_PATH)
+    if _tfm is None:
+        _tfm = build_transforms()
+    if _model is None:
+        if not os.path.exists(MODEL_WEIGHTS):
+            raise FileNotFoundError(f"File pesi non trovato: {MODEL_WEIGHTS}")
+        _model = load_model(MODEL_WEIGHTS, _classes)
+
+@torch.inference_mode()
+def predict(image: Image.Image):
+    setup()
+    if image.mode != "RGB":
+        image = image.convert("RGB")
+    x = _tfm(image).unsqueeze(0)
+    logits = _model(x)
+    probs = torch.softmax(logits, dim=1).cpu().numpy().squeeze(0)
+    top_idx = np.argsort(-probs)[:3]
+    top_labels = [_classes[i] for i in top_idx]
+    top_scores = [float(probs[i]) for i in top_idx]
+    pred_label = top_labels[0]
+    pred_conf = top_scores[0]
+    result = {
+        "prediction": pred_label,
+        "confidence": round(pred_conf * 100, 2),
+        "top3": [
+            {"label": top_labels[j], "confidence": round(top_scores[j] * 100, 2)}
+            for j in range(3)
+        ]
+    }
+    human = f"Tipo: {pred_label} — Affidabilità: {result['confidence']}%"
+    return human, result
+
+title = "Corrosion Classifier (ViT-B/16 • CPU)"
+description = "Carica o scatta una foto del pezzo corroso. Predizione su CPU."
+
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown(f"# {title}")
+    gr.Markdown(description)
+    with gr.Row():
+        with gr.Column():
+            inp = gr.Image(label="Immagine", type="pil", sources=["upload", "camera"], image_mode="RGB")
+            analyze_btn = gr.Button("Analizza immagine", variant="primary")
+        with gr.Column():
+            out_text = gr.Textbox(label="Risultato", interactive=False)
+            out_json = gr.JSON(label="Dettagli (top-3)")
+    analyze_btn.click(fn=predict, inputs=[inp], outputs=[out_text, out_json])
+    gr.Examples(examples=[], inputs=[inp])
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)

classes.json

@@ -0,0 +1,11 @@
+[
+  "crevice_corrosion",
+  "erosion_corrosion",
+  "galvanic_corrosion",
+  "mic_corrosion",
+  "no_corrosion",
+  "pitting_corrosion",
+  "stress_corrosion",
+  "under_insulation_corrosion",
+  "uniform_corrosion"
+]

requirements.txt

@@ -0,0 +1,6 @@
+gradio>=4.44.0
+torch>=2.3.0
+torchvision>=0.18.0
+timm>=0.9.12
+pillow>=10.3.0
+numpy>=1.26.4