Upload 7 files

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+impression_sunrise.jpg filter=lfs diff=lfs merge=lfs -text
+les_demoiselles.jpg filter=lfs diff=lfs merge=lfs -text
+mona_lisa.jpg filter=lfs diff=lfs merge=lfs -text
+starry_night.jpg filter=lfs diff=lfs merge=lfs -text
+the_scream.jpg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,13 +1,33 @@
----
-title: Art Style Classifier
-emoji: 🏆
-colorFrom: red
-colorTo: indigo
-sdk: gradio
-sdk_version: 5.29.1
-app_file: app.py
-pinned: false
-short_description: Deep learning classifier for 27 art styles.
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Art Style Classifier
+
+An interactive deep learning model that identifies 27 different artistic styles from paintings.
+
+## About
+
+This project uses a fine-tuned ResNet34 model trained on the ArtWiki dataset containing over 50,000 images across 27 distinct art styles from Realism to Abstract Expressionism.
+
+## Features
+
+- Classifies paintings into 27 artistic styles
+- Provides confidence scores for top predictions
+- Includes information about each art style
+- User-friendly interface built with Gradio
+
+## Performance
+
+- 74.2% accuracy across all 27 classes
+- 76.2% precision and 74.2% recall
+- Particularly strong at recognizing distinctive styles like Pointillism (90%+ accuracy)
+
+## Technology
+
+- Transfer learning with ResNet34
+- Data augmentation techniques including rotation, flipping, and random cropping
+- Class imbalance handling through weighted sampling
+- Trained on MacBook with M3 Apple Silicon
+
+## How to Use
+
+1. Upload an image of a painting
+2. View the predicted art style with confidence scores
+3. Learn about the characteristics of the predicted style