1

app.py

@@ -0,0 +1,198 @@
+import streamlit as st
+from PIL import Image
+import os
+import json
+import numpy as np
+import pickle
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score
+
+# Directories
+UPLOAD_DIR = "uploaded_images"
+MODEL_PATH = os.path.join(UPLOAD_DIR, "model.pkl")
+LABELS_PATH = os.path.join(UPLOAD_DIR, "labels.json")
+os.makedirs(UPLOAD_DIR, exist_ok=True)
+
+# Initialize model and labels
+if os.path.exists(MODEL_PATH):
+    with open(MODEL_PATH, "rb") as f:
+        model = pickle.load(f)
+else:
+    model = RandomForestClassifier()
+
+if os.path.exists(LABELS_PATH):
+    with open(LABELS_PATH, "r") as f:
+        labels = json.load(f)
+else:
+    labels = {}
+
+# Helper function to save uploaded images
+def save_uploaded_image(uploaded_file):
+    file_path = os.path.join(UPLOAD_DIR, uploaded_file.name)
+    with open(file_path, "wb") as f:
+        f.write(uploaded_file.getbuffer())
+    return file_path
+
+# Feature extraction function
+def extract_features(file_path):
+    try:
+        with Image.open(file_path) as img:
+            return np.array(img.resize((64, 64))).flatten()
+    except Exception as e:
+        print(f"Error processing image {file_path}: {e}")
+        return None
+
+# Train the model
+def train_model(training_data):
+    global model, labels
+
+    features, targets = [], []
+    for file_path, label in training_data.items():
+        feature = extract_features(file_path)
+        if feature is not None:
+            features.append(feature)
+            targets.append(label)
+
+    if features:
+        features = np.array(features)
+        targets = np.array(targets)
+
+        model.fit(features, targets)
+
+        with open(MODEL_PATH, "wb") as f:
+            pickle.dump(model, f)
+
+        with open(LABELS_PATH, "w") as f:
+            json.dump(labels, f)
+    else:
+        print("No valid features found for training.")
+
+# Classify an image
+def classify_image(file_path):
+    global model
+    features = extract_features(file_path)
+    if features is not None:
+        features = features.reshape(1, -1)
+        return model.predict(features)[0]
+    else:
+        return "Invalid Image"
+
+# Streamlit app
+def main():
+    st.title("Human or Alien Identification")
+
+    st.markdown(
+        """
+        Welcome to the **Human or Alien Identification App**! Here's what you can do:
+
+        - **Identify Image:** Upload an image and classify it as "Human" or "Alien." The classifications you save will be added to the training data.
+        - **Train Model:** Review and manage the images already classified as "Human" or "Alien." Upload additional images to improve the training dataset.
+        """
+    )
+
+    tab1, tab2 = st.tabs(["Identify Image", "Train Model"])
+
+    with tab1:
+        st.header("Identify Image")
+        uploaded_files = st.file_uploader("Upload Images to Identify", type=["jpg", "jpeg", "png"], accept_multiple_files=True)
+
+        if uploaded_files:
+            results = {}
+
+            for uploaded_file in uploaded_files:
+                try:
+                    st.image(uploaded_file, caption=f"Uploaded Image: {uploaded_file.name}", use_container_width=True)
+                    file_path = save_uploaded_image(uploaded_file)
+
+                    prediction = classify_image(file_path)
+                    results[file_path] = prediction
+                except Exception as e:
+                    st.error(f"Error processing file {uploaded_file.name}: {e}")
+
+            st.subheader("Classification Results")
+            for file_path, label in results.items():
+                st.markdown(
+                    f"<p style='font-size:20px; color:blue; background-color:lightyellow; padding:10px; border-radius:5px;'>Image: {os.path.basename(file_path)} - Classified as: {label}</p>",
+                    unsafe_allow_html=True,
+                )
+
+    with tab2:
+        st.header("Train Model")
+
+        st.subheader("Upload Images for Training")
+        training_files = st.file_uploader("Upload Training Images", type=["jpg", "jpeg", "png"], accept_multiple_files=True)
+
+        if training_files:
+            training_data = {}
+
+            for training_file in training_files:
+                try:
+                    st.image(training_file, caption=f"Uploaded Training Image: {training_file.name}", use_container_width=True)
+                    file_path = save_uploaded_image(training_file)
+
+                    label = st.radio(
+                        f"Classify {training_file.name}", ["Human", "Alien"], index=0, key=training_file.name
+                    )
+                    training_data[file_path] = label
+                except Exception as e:
+                    st.error(f"Error processing file {training_file.name}: {e}")
+
+            if st.button("Save Training Data"):
+                labels.update(training_data)
+                train_model(training_data)
+                st.success("Training data has been saved and the model updated.")
+
+        st.subheader("Bulk Train with Existing Dataset")
+        dataset_file = st.file_uploader("Upload a JSON file containing labeled data", type=["json"])
+
+        if dataset_file:
+            dataset = json.load(dataset_file)
+            training_data = {}
+
+            for label, image_paths in dataset.items():
+                for image_path in image_paths:
+                    if os.path.exists(image_path):
+                        training_data[image_path] = label
+
+            if st.button("Train Model with Dataset"):
+                labels.update(training_data)
+                train_model(training_data)
+                st.success("Model has been trained with the uploaded dataset.")
+
+        subtab1, subtab2 = st.tabs(["Humans", "Aliens"])
+
+        with subtab1:
+            st.subheader("Human Images")
+            human_images = [
+                img for img, lbl in labels.items() if lbl == "Human"
+            ]
+
+            if human_images:
+                st.info("These images have already been trained.")
+                for image_path in human_images:
+                    try:
+                        st.image(image_path, caption=f"Image: {os.path.basename(image_path)}", use_container_width=True)
+                    except Exception as e:
+                        st.error(f"Error displaying image {os.path.basename(image_path)}: {e}")
+            else:
+                st.warning("No human images found for training.")
+
+        with subtab2:
+            st.subheader("Alien Images")
+            alien_images = [
+                img for img, lbl in labels.items() if lbl == "Alien"
+            ]
+
+            if alien_images:
+                st.info("These images have already been trained.")
+                for image_path in alien_images:
+                    try:
+                        st.image(image_path, caption=f"Image: {os.path.basename(image_path)}", use_container_width=True)
+                    except Exception as e:
+                        st.error(f"Error displaying image {os.path.basename(image_path)}: {e}")
+            else:
+                st.warning("No alien images found for training.")
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,2 @@
+streamlit
+pillow

uploaded_images/labels.json

@@ -0,0 +1 @@
+{"uploaded_images\\tamil-woman-close-up-of-happy-face-ECNPHF.jpg": "Human", "uploaded_images\\ung-cheerful-beautiful-girl-long-hair-casual-shirt-smiling-looking-172927805.jpg": "Human", "uploaded_images\\vladimir-putin-smiling-face-png-11646750895sk2xyu6id1.png": "Human", "uploaded_images\\web3-happy-people-outside-smile-sun-nature-eduardo-dutra-620857-unsplash.jpg": "Human", "uploaded_images\\94.jpg": "Alien", "uploaded_images\\95.jpg": "Alien", "uploaded_images\\96.jpg": "Alien"}

uploaded_images/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:425e5e787d6520e1c913c617f7eadb38c4d7eabcd3df9444e9ec07dd9cae70c7
+size 55529

uploaded_images/sample_dataset.json

@@ -0,0 +1 @@
+{"Human": ["uploaded_images/sample_human1.jpg", "uploaded_images/sample_human2.jpg"], "Alien": ["uploaded_images/sample_alien1.jpg", "uploaded_images/sample_alien2.jpg"]}