Upload 3 files

README.md

@@ -1,14 +1,103 @@
 ---
-title: Skin Ai
-emoji: 🌍
-colorFrom: purple
-colorTo: indigo
+title: Skin Disease Detection Demo
+emoji: 🩺
+colorFrom: blue
+colorTo: pink
 sdk: gradio
-sdk_version: 6.22.0
-python_version: '3.12'
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
 license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🩺 Skin Disease Detection — Demo
+
+A demo Space where users can:
+- **Upload a skin image** (or click a bundled test image) and get an AI classifier's
+  prediction across common skin-lesion categories, plus a plain-language explanation
+  from an LLM.
+- **Chat with an assistant** about skin-health topics in general terms.
+
+> ⚠️ **This is an educational demo, not a medical device.** It does not diagnose
+> anyone. Always see a licensed dermatologist or doctor for real concerns.
+
+## How it works
+
+| Component | What it does | Default model |
+|---|---|---|
+| Image classifier | Predicts a skin-lesion category from a photo | [`Anwarkh1/Skin_Cancer-Image_Classification`](https://huggingface.co/Anwarkh1/Skin_Cancer-Image_Classification) (ViT, trained on a HAM10000-derived dataset) |
+| Chat / explanation | Turns predictions into plain language, answers general questions | `HuggingFaceH4/zephyr-7b-beta` via the Hugging Face Inference API |
+
+Both model IDs are configurable via Space variables — see below.
+
+## 🚀 Deploy in 3 steps
+
+1. **Create a new Space**
+   Go to [huggingface.co/new-space](https://huggingface.co/new-space) →
+   choose **Gradio** as the SDK → CPU basic hardware is fine for the classifier.
+
+2. **Upload these files**
+   Upload `app.py`, `requirements.txt`, this `README.md`, and the `examples/`
+   folder (if you generated test images — see below) to the Space repo,
+   either via the web UI ("Add file") or:
+   ```bash
+   git clone https://huggingface.co/spaces/<your-username>/<your-space-name>
+   cp -r app.py requirements.txt README.md examples <your-space-name>/
+   cd <your-space-name>
+   git add . && git commit -m "Initial commit" && git push
+   ```
+
+3. **Add an `HF_TOKEN` secret (to enable the chatbot)**
+   In your Space → **Settings → Variables and secrets** → **New secret**:
+   - Name: `HF_TOKEN`
+   - Value: a Hugging Face access token (create one at
+     [huggingface.co/settings/tokens](https://huggingface.co/settings/tokens),
+     "Read" scope is enough)
+
+   Without this secret, image classification still works — only the
+   LLM-generated explanations and the chat tab need the token.
+
+That's it — the Space will build and the app will be live.
+
+## 🖼️ Adding test images (optional but recommended)
+
+So users have something to click without needing their own photo, generate a
+few sample images once, locally, before you push:
+
+```bash
+pip install datasets pillow
+python scripts/download_examples.py
+```
+
+This saves a handful of JPEGs into `examples/`. Include that folder when you
+push to your Space. You can also just drop in your own sample `.jpg`/`.png`
+files instead.
+
+## ⚙️ Configuration
+
+Set these as Space **variables** (not secret, unless noted) to customize:
+
+- `IMAGE_MODEL_ID` — any Hugging Face image-classification model compatible
+  with `transformers.pipeline("image-classification", ...)`.
+- `CHAT_MODEL_ID` — any chat-completion-capable model available via the HF
+  Inference API.
+- `HF_TOKEN` *(secret)* — required for the chat tab and the LLM explanations.
+
+## 🩹 Limitations & responsible use
+
+- The bundled classifier was trained on a research dataset (HAM10000-derived
+  dermatoscopic images) and has real, published accuracy limits — treat its
+  output as a talking point, not a result.
+- Performance depends heavily on image quality, lighting, and skin tone
+  representation in the training data.
+- This app must not be used as a substitute for professional medical
+  evaluation, and the chatbot is instructed to avoid giving diagnoses,
+  treatments, or dosages.
+
+## Local development
+
+```bash
+pip install -r requirements.txt
+export HF_TOKEN=your_token_here   # optional, for chat
+python app.py
+```

app.py

@@ -0,0 +1,205 @@
+"""
+Skin Disease Detection — Demo
+------------------------------
+A Gradio app for Hugging Face Spaces with two tabs:
+  1. Image Diagnosis — upload (or pick a test image) and get a classifier
+     prediction plus a plain-language, LLM-generated explanation.
+  2. Chat Assistant — a cautious chatbot that discusses skin-health topics
+     in general terms.
+
+IMPORTANT: This is an educational / demo tool only. It is NOT a medical
+device and does NOT provide real diagnoses. See DISCLAIMER below.
+"""
+
+import os
+
+import gradio as gr
+from transformers import pipeline
+from huggingface_hub import InferenceClient
+
+# ---------------------------------------------------------------------------
+# Config — override any of these via Space "Variables and secrets"
+# ---------------------------------------------------------------------------
+IMAGE_MODEL_ID = os.environ.get("IMAGE_MODEL_ID", "Anwarkh1/Skin_Cancer-Image_Classification")
+CHAT_MODEL_ID = os.environ.get("CHAT_MODEL_ID", "HuggingFaceH4/zephyr-7b-beta")
+HF_TOKEN = os.environ.get("HF_TOKEN")  # add as a Space SECRET to enable the chatbot
+
+DISCLAIMER = (
+    "⚠️ **Demo only — not a medical device.** This tool does not provide a real "
+    "diagnosis. Always consult a licensed dermatologist or doctor about any skin concern, "
+    "especially a spot that is new, changing, bleeding, or growing."
+)
+
+# Friendly descriptions for the HAM10000-style label set used by the default model
+CLASS_INFO = {
+    "actinic keratoses": "A rough, scaly patch caused by sun damage; can occasionally progress to skin cancer if untreated.",
+    "basal cell carcinoma": "The most common type of skin cancer; usually slow-growing and rarely spreads.",
+    "benign keratosis-like lesions": "Non-cancerous growths such as seborrheic keratoses or solar lentigines.",
+    "dermatofibroma": "A common benign skin nodule, often on the legs; usually harmless.",
+    "melanocytic nevi": "Ordinary moles; almost always benign.",
+    "melanoma": "The most serious type of skin cancer; early evaluation matters.",
+    "vascular lesions": "Blood-vessel-related marks such as angiomas; usually benign.",
+}
+
+SYSTEM_PROMPT = (
+    "You are a friendly, cautious skin-health assistant inside a DEMO application. "
+    "You are NOT a doctor and must never claim to diagnose anyone. "
+    "You can explain general skin-condition concepts, explain what classifier labels mean, "
+    "and encourage users to see a licensed dermatologist for anything concerning. "
+    "Never give a definitive diagnosis, never prescribe treatment or dosages. "
+    "If a user describes an urgent symptom (rapid growth, bleeding, severe pain, sudden "
+    "changes to a mole), tell them to seek in-person medical care promptly. "
+    "Keep answers concise — 3 to 6 sentences."
+)
+
+# ---------------------------------------------------------------------------
+# Lazy-loaded models (so the Space boots fast and only loads on first use)
+# ---------------------------------------------------------------------------
+_classifier = None
+
+
+def get_classifier():
+    global _classifier
+    if _classifier is None:
+        _classifier = pipeline("image-classification", model=IMAGE_MODEL_ID)
+    return _classifier
+
+
+def get_chat_client():
+    if not HF_TOKEN:
+        return None
+    return InferenceClient(model=CHAT_MODEL_ID, token=HF_TOKEN)
+
+
+# ---------------------------------------------------------------------------
+# Image diagnosis tab
+# ---------------------------------------------------------------------------
+def diagnose_image(image):
+    """Run the image classifier and, if a chat model is configured, ask it
+    for a short plain-language explanation of the top prediction."""
+    if image is None:
+        return "Please upload an image or pick one of the test images first."
+
+    try:
+        clf = get_classifier()
+        preds = clf(image, top_k=5)
+    except Exception as e:
+        return f"**Model error:** {e}\n\nMake sure `IMAGE_MODEL_ID` is a valid image-classification model."
+
+    lines = ["### 🔬 Classifier prediction (demo only — not a real diagnosis)\n"]
+    for p in preds:
+        label = p["label"]
+        score = p["score"] * 100
+        info = CLASS_INFO.get(label.lower(), "")
+        info_txt = f"  \n  _{info}_" if info else ""
+        lines.append(f"- **{label}** — {score:.1f}%{info_txt}")
+
+    client = get_chat_client()
+    if client is not None:
+        top = preds[0]["label"]
+        others = ", ".join(p["label"] for p in preds[1:3])
+        prompt = (
+            f"An image classifier in a DEMO app predicted the top possible skin-lesion "
+            f"category as '{top}' ({preds[0]['score']*100:.1f}% confidence), with other "
+            f"candidates including {others}. In 3-4 short, plain-language sentences, explain "
+            f"generally what the top category means, make clear this is not a diagnosis, and "
+            f"suggest seeing a dermatologist, especially if the spot is new, changing, or bleeding."
+        )
+        try:
+            resp = client.chat_completion(
+                messages=[
+                    {"role": "system", "content": SYSTEM_PROMPT},
+                    {"role": "user", "content": prompt},
+                ],
+                max_tokens=250,
+            )
+            explanation = resp.choices[0].message.content
+            lines.append(f"\n### 🤖 Assistant explanation\n{explanation}")
+        except Exception as e:
+            lines.append(f"\n_(Chat explanation unavailable: {e})_")
+    else:
+        lines.append(
+            "\n_Add an `HF_TOKEN` secret to this Space to also get a plain-language "
+            "explanation from the chat model here._"
+        )
+
+    lines.append(f"\n---\n{DISCLAIMER}")
+    return "\n".join(lines)
+
+
+# ---------------------------------------------------------------------------
+# Chat tab
+# ---------------------------------------------------------------------------
+def chat_respond(message, history):
+    client = get_chat_client()
+    if client is None:
+        return (
+            "⚠️ Chat isn't configured yet. Add an `HF_TOKEN` secret to this Space "
+            "(Settings → Variables and secrets) to enable the chatbot."
+        )
+
+    messages = [{"role": "system", "content": SYSTEM_PROMPT}]
+    for turn in history:
+        # supports both the legacy (user, bot) tuple format and the
+        # newer {"role": ..., "content": ...} messages format
+        if isinstance(turn, dict):
+            messages.append(turn)
+        else:
+            user_msg, bot_msg = turn
+            messages.append({"role": "user", "content": user_msg})
+            if bot_msg:
+                messages.append({"role": "assistant", "content": bot_msg})
+    messages.append({"role": "user", "content": message})
+
+    try:
+        resp = client.chat_completion(messages=messages, max_tokens=400)
+        return resp.choices[0].message.content
+    except Exception as e:
+        return f"Sorry, I hit an error talking to the model: {e}"
+
+
+# ---------------------------------------------------------------------------
+# UI
+# ---------------------------------------------------------------------------
+EXAMPLES_DIR = "examples"
+example_images = []
+if os.path.isdir(EXAMPLES_DIR):
+    example_images = [
+        os.path.join(EXAMPLES_DIR, f)
+        for f in sorted(os.listdir(EXAMPLES_DIR))
+        if f.lower().endswith((".jpg", ".jpeg", ".png"))
+    ]
+
+with gr.Blocks(title="Skin Disease Detection — Demo") as demo:
+    gr.Markdown("# 🩺 Skin Disease Detection — Demo")
+    gr.Markdown(DISCLAIMER)
+
+    with gr.Tab("📷 Image Diagnosis"):
+        with gr.Row():
+            with gr.Column():
+                img_in = gr.Image(type="pil", label="Upload a skin image")
+                if example_images:
+                    gr.Examples(examples=example_images, inputs=img_in, label="Or try a test image")
+                else:
+                    gr.Markdown(
+                        "_No bundled test images found. Run `scripts/download_examples.py` "
+                        "before deploying, or just upload your own image._"
+                    )
+                analyze_btn = gr.Button("Analyze image", variant="primary")
+            with gr.Column():
+                result_md = gr.Markdown()
+        analyze_btn.click(diagnose_image, inputs=img_in, outputs=result_md)
+
+    with gr.Tab("💬 Chat Assistant"):
+        gr.ChatInterface(
+            fn=chat_respond,
+            description="Ask general questions about skin conditions. " + DISCLAIMER,
+        )
+
+    gr.Markdown(
+        "---\nBuilt with 🤗 Transformers + Gradio. "
+        f"Image model: `{IMAGE_MODEL_ID}`  ·  Chat model: `{CHAT_MODEL_ID}`"
+    )
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,5 @@
+gradio>=4.44.0
+transformers>=4.42.0
+torch>=2.2.0
+pillow>=10.0.0
+huggingface_hub>=0.24.0