Uploaded model and examples

MCS-Classify.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a03e2b79be60934dd222fe016590ae587ad619d3a28a70fa3b05d25ba4ed8ff3
+size 20907727

README.md

@@ -1,8 +1,8 @@
 ---
 title: Maltese Christian Statue Classification
-emoji: 🚀
+emoji: ✝
 colorFrom: red
-colorTo: yellow
+colorTo: purple
 sdk: gradio
 sdk_version: 5.12.0
 app_file: app.py

app.py

@@ -0,0 +1,100 @@
+from typing import List, Tuple
+import gradio as gr
+from ultralytics import YOLO
+import cv2
+import os
+import torch
+import numpy as np
+
+# Check device availability
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+# Categories in English and Maltese
+categories_english = {
+    0: 'Jesus has Risen', 1: 'Jesus praying in Gethsemane', 2: 'Saint Philip of Agira',
+    3: 'Simon of Cyrene', 4: 'The Betrayal of Judas', 5: 'The Cross',
+    6: 'The Crucifixion', 7: 'The Ecce Homo', 8: 'The Flogged',
+    9: 'The Lady of Sorrows', 10: 'The Last Supper', 11: 'The Monument',
+    12: 'The Redeemer', 13: 'The Veronica'
+}
+
+categories_maltese = {
+    0: 'L-Irxoxt', 1: 'Ġesù fl-Ort tal-Ġetsemani', 2: 'San Filep ta’ Aġġira',
+    3: 'Xmun min Ċireni', 4: 'It-Tradiment ta’ Ġuda', 5: 'Is-Salib',
+    6: 'Il-Vara Il-Kbira', 7: 'L-Ecce Homo', 8: 'Il-Marbut',
+    9: 'Id-Duluri', 10: 'L-Aħħar Ċena', 11: 'Il-Monument',
+    12: 'Ir-Redentur', 13: 'Il-Veronica'
+}
+
+def predict_image(image, model, size=(244, 244)) -> List[Tuple[str, str, float]]:
+    """Predict the class of a given image and return sorted probabilities with categories."""
+    image = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
+    resized_img = cv2.resize(image, size)
+    resized_img = resized_img / 255.0  # Normalize
+    resized_img = resized_img.transpose(2, 0, 1)  # Convert to (C, H, W)
+    resized_img = resized_img[None, ...]  # Add batch dimension
+
+    # Run prediction
+    results = model.predict(image)
+    pred_probs = results[0].probs.data.cpu().numpy()
+
+    # Sort predictions by probability
+    sorted_indices = np.argsort(pred_probs)[::-1]  # Descending order
+    sorted_predictions = [
+        (
+            categories_english[i],
+            categories_maltese[i],
+            round(pred_probs[i] * 100, 2)  # Convert to percentage
+        )
+        for i in sorted_indices
+    ]
+
+    return sorted_predictions
+
+# Load the model
+model = YOLO("https://huggingface.co/mbar0075/Maltese-Christian-Statue-Classification/resolve/main/MCS-Classify.pt").to(device)
+
+example_dir = "https://huggingface.co/spaces/mbar0075/Maltese-Christian-Statue-Classification/resolve/main/examples/"
+
+# Example images
+example_images = [
+    example_dir + "1.jpg",
+    example_dir + "2.jpg",
+    example_dir + "3.jpg",
+    example_dir + "4.jpg",
+    example_dir + "5.jpg",
+    example_dir + "6.jpg",
+    example_dir + "7.jpg",
+    example_dir + "8.jpg",
+    example_dir + "9.jpg",
+    example_dir + "10.jpg",
+    example_dir + "11.jpg",
+    example_dir + "12.jpg"
+]
+
+# Define the Gradio interface
+def classify_image(input_image):
+    predictions = predict_image(input_image, model)
+    english_labels = "\n".join([f"{label}: {confidence}%" for label, _, confidence in predictions])
+    maltese_labels = "\n".join([f"{label}: {confidence}%" for _, label, confidence in predictions])
+    return english_labels, maltese_labels
+
+# Gradio interface
+gr.Interface(
+    fn=classify_image,
+    inputs=gr.Image(type="pil", label="Upload an image", tool="editor"),
+    outputs=[
+        gr.Textbox(label="Predictions (English)", lines=15),
+        gr.Textbox(label="Predictions (Maltese)", lines=15),
+    ],
+    examples=example_images,
+    live=True,
+    title="Maltese Christian Statue Image Classification ✝",
+    description=(
+        "This project aims to classify Maltese Christian statues and religious figures depicted in images. The model is trained "
+        "to recognise 14 unique categories related to the Maltese Christian heritage, presented in both English and Maltese. "
+        "Upload an image to receive predictions of the statue's category, sorted by likelihood. This initiative promotes the "
+        "preservation of Maltese culture and traditions through AI technology."
+    ),
+    article="The YOLO11s classificaiton model was trained on a dataset of Maltese Christian statues and religious figures during the period of Lent."
+).launch()

requirements.txt

@@ -0,0 +1,8 @@
+setuptools<70.0.0
+awscli==1.29.54
+gradio
+inference
+supervision
+ultralytics
+dill
+timm