Upload hello_doctor.py

hello_doctor.py

@@ -0,0 +1,82 @@
+# -*- coding: utf-8 -*-
+"""Hello doctor skin.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1OI0xTyanplOVAVOE0OmaZBoPmrFSS9tb
+"""
+
+!pip install --upgrade openai transformers gradio huggingface_hub
+
+import os
+from getpass import getpass
+from huggingface_hub import login
+
+# Securely input your keys
+openai_api_key = getpass("Enter your OpenAI API key: ")
+hf_token = getpass("Enter your Hugging Face token: ")
+
+# Set environment variables (no keys shown in code)
+os.environ["OPENAI_API_KEY"] = openai_api_key
+
+# Login to Hugging Face
+login(token=hf_token)
+
+import openai
+
+def analyze_symptoms(symptoms):
+    try:
+        client = openai.OpenAI(api_key=os.environ["OPENAI_API_KEY"])
+        response = client.chat.completions.create(
+            model="gpt-4-turbo",
+            messages=[
+                {"role": "system", "content": "You are a helpful medical assistant."},
+                {"role": "user", "content": f"I have these symptoms: {symptoms}. What might be the cause?"}
+            ]
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"Error analyzing symptoms: {e}"
+
+from transformers import AutoProcessor, AutoModelForImageClassification
+from PIL import Image
+import torch
+
+# Load public image model
+image_model_id = "microsoft/beit-base-patch16-224-pt22k-ft22k"
+processor = AutoProcessor.from_pretrained(image_model_id, token=HF_TOKEN)
+model = AutoModelForImageClassification.from_pretrained(image_model_id, token=HF_TOKEN)
+
+def classify_image(image):
+    try:
+        img = image.convert("RGB")
+        inputs = processor(images=img, return_tensors="pt")
+        with torch.no_grad():
+            outputs = model(**inputs)
+            predicted_class = outputs.logits.argmax(-1).item()
+            label = model.config.id2label[predicted_class]
+        return f"Predicted skin condition: {label}"
+    except Exception as e:
+        return f"Error classifying image: {e}"
+
+import gradio as gr
+
+iface = gr.Interface(
+    fn=lambda symptoms, img: (
+        analyze_symptoms(symptoms),
+        classify_image(img) if img else "No image uploaded"
+    ),
+    inputs=[
+        gr.Textbox(label="Describe your symptoms"),
+        gr.Image(type="pil", label="Upload skin image")
+    ],
+    outputs=[
+        gr.Textbox(label="Symptom Analysis"),
+        gr.Textbox(label="Image Diagnosis")
+    ],
+    title="AI Doctor",
+    description="Enter your symptoms and/or upload a skin image to get medical insights. Powered by GPT-4 and Hugging Face vision transformers."
+)
+
+iface.launch(share=True)