Upload dataset_viz.py

dataset_viz.py

@@ -0,0 +1,50 @@
+import streamlit as st
+import os
+from PIL import Image
+import random
+
+# Dataset directory (update this path as needed)
+DATASET_DIR = "Dataset_final/test"
+
+# Mapping numeric folder names to emotion labels
+emotion_labels = {
+    "1": "Surprise",
+    "2": "Disgust",
+    "3": "Happiness",
+    "4": "Sadness",
+    "5": "Anger",
+    "6": "Neutral"
+}
+
+def show_sample_images_page():
+    st.title("Face Emotion Dataset Visualization")
+
+    # Slider to control number of images per emotion
+    num_images = st.slider("Number of images to display per emotion:", min_value=1, max_value=20, value=5)
+
+    # Check dataset path
+    if not os.path.isdir(DATASET_DIR):
+        st.error(f"Dataset path '{DATASET_DIR}' not found.")
+    else:
+        # Only process folders that are in the defined emotion_labels
+        valid_folders = [f for f in os.listdir(DATASET_DIR) if f in emotion_labels]
+
+        if not valid_folders:
+            st.warning("No valid emotion folders (1–6) found in the dataset directory.")
+        else:
+            for folder in sorted(valid_folders):
+                emotion_name = emotion_labels[folder]
+                emotion_path = os.path.join(DATASET_DIR, folder)
+                image_files = [f for f in os.listdir(emotion_path) if f.lower().endswith(('.png', '.jpg', '.jpeg'))]
+
+                st.subheader(f"{emotion_name} ({len(image_files)} images)")
+                selected_images = random.sample(image_files, min(num_images, len(image_files)))
+
+                cols = st.columns(min(5, len(selected_images)))
+                for i, img_file in enumerate(selected_images):
+                    img_path = os.path.join(emotion_path, img_file)
+                    try:
+                        image = Image.open(img_path)
+                        cols[i % len(cols)].image(image, use_container_width = True, caption=img_file)
+                    except Exception as e:
+                        st.error(f"Failed to load {img_path}: {e}")