Full feature upload.

.gitignore

@@ -0,0 +1,2 @@
+.venv
+.env

app.py

@@ -1,4 +1,91 @@
 import streamlit as st
+import tempfile
+from PIL import Image
+import os
+import tensorflow as tf
+import numpy as np
+import io
+import json
+from google import genai
+from dotenv import dotenv_values, load_dotenv
 
-x = st.slider('Select a value')
-st.write(x, 'squared is', x * x)
+load_dotenv()
+
+config = dotenv_values(".env")
+
+MODEL_ID = "gemini-2.0-flash" 
+model_id = MODEL_ID
+client = genai.Client(api_key = config["GEMINI_API_KEY"])
+
+# Load labels from JSON file
+with open('model/labels.json', 'r') as f:
+    labels = json.load(f)
+
+# Load TensorFlow Lite model
+def load_tflite_model():
+    interpreter = tf.lite.Interpreter(model_path="model/model.tflite")
+    interpreter.allocate_tensors()
+    return interpreter
+
+# Get input and output details
+interpreter = load_tflite_model()
+input_details = interpreter.get_input_details()
+output_details = interpreter.get_output_details()
+
+def main():
+    st.title("Skin Cancer Classifier")
+    img_file = None
+    image_path = ""
+
+    # Upload an image
+    img_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
+
+    if img_file is not None:
+        # Save the image to a temporary file
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as temp_file:
+            temp_file.write(img_file.read())
+            image_path = temp_file.name
+            st.write("Image saved to:", image_path) 
+            st.image(image_path, use_container_width=True)
+
+    if st.button("Classify"):
+        with st.spinner("Processing..."):
+            image = Image.open(io.BytesIO(img_file.getbuffer())).convert('RGB')
+            
+            # Get model input shape
+            input_shape = input_details[0]['shape']
+            
+            # Preprocess the image
+            image = image.resize((input_shape[1], input_shape[2]))
+            image_array = np.array(image, dtype=np.float32)
+            image_array = image_array / 255.0
+            image_array = np.expand_dims(image_array, axis=0)
+            
+            # Make prediction
+            interpreter.set_tensor(input_details[0]['index'], image_array)
+            interpreter.invoke()
+            
+            # Get prediction results
+            prediction = interpreter.get_tensor(output_details[0]['index'])
+            
+            # Get the predicted class index
+            predicted_class_index = np.argmax(prediction[0])
+            
+            # Get the corresponding label information
+            predicted_label = labels[predicted_class_index]
+
+            genai_response = client.models.generate_content(
+                model=MODEL_ID,
+                contents=[
+                    "You are a medical encyclopedia. You are given a skin cancer image and you need to provide a detailed explanation of the disease and its treatment.",
+                    f"The predicted class is {predicted_label['name']}. Provide a detailed explanation of the disease, its treatment, and prevention.",
+                    "Start with the title: Classfification of Skin Cancer",
+                ]
+            )
+
+            # Display Gemini response
+            st.markdown(genai_response.text)
+
+
+if __name__ == "__main__":
+    main()  

model/labels.json

@@ -0,0 +1,30 @@
+[
+    {
+        "label": "akiec",
+        "name": "Actinic Keratoses and Intraepithelial Carcinomae"
+    },
+    {
+        "label": "bcc",
+        "name": "Basal Cell Carcinoma"
+    },
+    {
+        "label": "bkl",
+        "name": "Benign Keratosis"
+    },
+    {
+        "label": "df",
+        "name": "Dermatofibroma"
+    },
+    {
+        "label": "mel",
+        "name": "Melanoma"
+    },
+    {
+        "label": "nv",
+        "name": "Melanocytic Nevi"
+    },
+    {
+        "label": "vasc",
+        "name": "Vascular Skin Lesions"
+    }
+]

model/labels.txt

@@ -0,0 +1,7 @@
+akiec
+bcc
+bkl
+df
+mel
+nv
+vasc

model/model.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:247b237b98b121939ec335d716b2bf9bc36d8c4e34d86729b7fb676a2ea6a590
+size 10336348

requirements.txt

@@ -0,0 +1,8 @@
+flask
+tensorflow
+numpy
+pillow
+python-dotenv
+google-genai
+streamlit
+python-dotenv