gremlin97: hf setup

.gitignore

@@ -0,0 +1,44 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual Environment
+.env
+.venv
+env/
+venv/
+ENV/
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+
+# Poetry
+poetry.lock
+dist/
+
+# Logs
+*.log
+
+# Local development
+.DS_Store 

README.md

@@ -1,13 +1,93 @@
----
-title: Mars Board
-emoji: 💻
-colorFrom: indigo
-colorTo: yellow
-sdk: gradio
-sdk_version: 5.23.1
-app_file: app.py
-pinned: false
-license: mit
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Mars Vision Leaderboard
+
+A comprehensive leaderboard for evaluating computer vision models on Mars-specific datasets. This leaderboard tracks performance across multiple tasks including classification, object detection, and segmentation.
+
+## Overview
+
+This leaderboard provides a standardized evaluation framework for computer vision models on Mars-specific datasets.
+
+## Tasks
+
+1. **Classification**
+   - DoMars16k - Surface Types
+   - Mars Image - Content Analysis
+   - Deep Mars - Deep Learning
+   - Dusty vs Non-dusty - Dust Analysis
+
+2. **Object Detection**
+   - Robins & Hynek - Craters
+   - Lagain - Surface Features
+   - SPOC - Surface Properties
+   - AI4MARS - Surface Analysis
+   - MarsData - General Surface
+
+3. **Segmentation**
+   - S5Mars - Surface
+   - Mars-Seg - Features
+   - Martian Landslide
+   - Martian Frost
+
+## Getting Started
+
+1. Clone the repository:
+   ```bash
+   git clone https://huggingface.co/spaces/gremlin97/mars-vision-leaderboard
+   cd mars-vision-leaderboard
+   ```
+
+2. Install dependencies using Poetry:
+   ```bash
+   poetry install
+   ```
+
+3. Run the leaderboard:
+   ```bash
+   # From the project root directory
+   poetry run python run.py
+   ```
+
+The leaderboard will be accessible at `http://localhost:7860` when running locally.
+
+## Features
+
+- Interactive Gradio interface
+- Filter models by task
+- Compare performance across datasets
+- Visualize results with plots
+- Track best performing models
+- Detailed results table
+
+## Contributing
+
+To add your model's results to the leaderboard:
+
+1. Fork this repository
+2. Add your results to the appropriate data dictionary in `app/data.py`
+3. Submit a pull request with your changes
+
+### Results Format
+
+For each task, results should be added in the following format:
+
+```python
+TASK_DATA = {
+    "Model": ["Model1", "Model2", ...],
+    "Dataset": ["Dataset1", "Dataset1", ...],
+    "Metric1": [value1, value2, ...],
+    "Metric2": [value1, value2, ...],
+}
+```
+
+## Project Structure
+
+```
+mars-vision-leaderboard/
+├── app/
+│   ├── __init__.py
+│   ├── app.py          # Main Gradio interface
+│   ├── data.py         # Dataset and model data
+│   └── leaderboard.py  # Visualization functions
+├── run.py              # Application entry point
+├── pyproject.toml      # Poetry dependencies
+└── README.md
+```

app/__init__.py

@@ -0,0 +1 @@
+"""Mars Vision Leaderboard package.""" 

app/app.py

@@ -0,0 +1,132 @@
+"""Main Gradio application for the Mars Vision Leaderboard."""
+
+import gradio as gr
+from .leaderboard import update_leaderboard
+from .data import TASK_DATA
+import pandas as pd
+
+TASKS = ["Classification", "Object Detection", "Segmentation"]
+
+def get_available_models(task: str) -> list:
+    """Get list of available models for a task."""
+    data, _ = TASK_DATA[task]
+    return sorted(list(set(data['Model'])))
+
+def create_interface():
+    """Create the Gradio interface."""
+    with gr.Blocks(title="Mars Vision Leaderboard", theme=gr.themes.Default()) as demo:
+        gr.Markdown(
+            """
+            # Mars Vision Leaderboard
+            A comprehensive evaluation of computer vision models on Mars-specific datasets.
+            This leaderboard tracks performance across multiple tasks including classification,
+            object detection, and segmentation.
+            """
+        )
+
+        with gr.Tabs():
+            with gr.TabItem("📊 General Datasets"):
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        gr.Markdown("""
+                        #### Surface Analysis
+                        • **Robins & Hynek** - Craters
+                        • **Lagain** - Surface Features
+                        • **SPOC** - Surface Properties
+                        • **AI4MARS** - Surface Analysis
+                        • **MarsData** - General Surface
+                        """)
+                    with gr.Column(scale=1):
+                        gr.Markdown("""
+                        #### Classification
+                        • **DoMars16k** - Surface Types
+                        • **Mars Image** - Content Analysis
+                        • **Deep Mars** - Deep Learning
+                        • **Dusty vs Non-dusty** - Dust Analysis
+                        """)
+                    with gr.Column(scale=1):
+                        gr.Markdown("""
+                        #### Segmentation
+                        • **S5Mars** - Surface
+                        • **Mars-Seg** - Features
+                        • **Martian Landslide**
+                        • **Martian Frost**
+                        """)
+            
+            with gr.TabItem("🎯 Specialized Tasks"):
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        gr.Markdown("""
+                        #### Detection Tasks
+                        • **Change Detection**
+                        • **Outlier Detection**
+                        • **Novelty Detection**
+                        """)
+                    with gr.Column(scale=1):
+                        gr.Markdown("""
+                        #### Feature Analysis
+                        • **Cone Detection**
+                        • **Dust Devil Tracks**
+                        • **Cone Segmentation**
+                        """)
+
+        with gr.Row():
+            task_dropdown = gr.Dropdown(
+                choices=TASKS,
+                value=TASKS[0],
+                label="Select Task",
+            )
+            model_multiselect = gr.Dropdown(
+                choices=get_available_models(TASKS[0]),
+                value=None,
+                label="Filter Models (Optional)",
+                multiselect=True,
+            )
+
+        with gr.Column():
+            gr.Markdown("### Best Performing Models Across Datasets")
+            best_models_output = gr.Dataframe(
+                interactive=False,
+                wrap=True,
+                headers=["Metric", "Rank", "Model", "Average Score"],
+            )
+
+            gr.Markdown("### Detailed Results")
+            table_output = gr.Dataframe(interactive=False, wrap=True)
+            
+            with gr.Row():
+                plot_output1 = gr.Plot(label="Performance Plot 1")
+                plot_output2 = gr.Plot(label="Performance Plot 2")
+
+        def update_models(task):
+            return gr.Dropdown(choices=get_available_models(task))
+
+        def update_with_filters(task, models):
+            return update_leaderboard(task, models)
+
+        # Event handlers
+        task_dropdown.change(
+            fn=update_models,
+            inputs=[task_dropdown],
+            outputs=[model_multiselect],
+        )
+
+        for component in [task_dropdown, model_multiselect]:
+            component.change(
+                fn=update_with_filters,
+                inputs=[task_dropdown, model_multiselect],
+                outputs=[table_output, plot_output1, plot_output2, best_models_output],
+            )
+
+        # Initial update
+        demo.load(
+            fn=update_with_filters,
+            inputs=[task_dropdown, model_multiselect],
+            outputs=[table_output, plot_output1, plot_output2, best_models_output],
+        )
+
+    return demo
+
+if __name__ == "__main__":
+    demo = create_interface()
+    demo.launch() 

app/data.py

@@ -0,0 +1,148 @@
+"""Leaderboard data for different tasks."""
+
+CLASSIFICATION_DATA = {
+    "Model": [
+        "ResNet-50",
+        "ViT-Base",
+        "Swin-T",
+        "InceptionV3",
+        "SqueezeNet",
+        "ResNet-50",
+        "ViT-Base",
+        "Swin-T",
+        "InceptionV3",
+        "SqueezeNet",
+        "ResNet-50",
+        "ViT-Base",
+        "Swin-T",
+        "InceptionV3",
+        "SqueezeNet",
+    ],
+    "Dataset": [
+        "DoMars16",
+        "DoMars16",
+        "DoMars16",
+        "DoMars16",
+        "DoMars16",
+        "Atmospheric Dust",
+        "Atmospheric Dust",
+        "Atmospheric Dust",
+        "Atmospheric Dust",
+        "Atmospheric Dust",
+        "Martian Frost",
+        "Martian Frost",
+        "Martian Frost",
+        "Martian Frost",
+        "Martian Frost",
+    ],
+    "Accuracy": [
+        92.5, 94.2, 95.8, 93.1, 89.7,
+        88.3, 90.1, 91.5, 89.8, 87.2,
+        85.6, 87.9, 88.4, 86.7, 84.3,
+    ],
+    "F1-Score": [
+        91.8, 93.5, 94.9, 92.4, 88.6,
+        87.5, 89.2, 90.7, 88.9, 86.3,
+        84.8, 86.9, 87.5, 85.8, 83.4,
+    ],
+}
+
+DETECTION_DATA = {
+    "Model": [
+        "Faster R-CNN",
+        "YOLOv5",
+        "DETR",
+        "RetinaNet",
+        "SSD",
+        "Faster R-CNN",
+        "YOLOv5",
+        "DETR",
+        "RetinaNet",
+        "SSD",
+        "Faster R-CNN",
+        "YOLOv5",
+        "DETR",
+        "RetinaNet",
+        "SSD",
+    ],
+    "Dataset": [
+        "Mars Crater",
+        "Mars Crater",
+        "Mars Crater",
+        "Mars Crater",
+        "Mars Crater",
+        "Rover Component",
+        "Rover Component",
+        "Rover Component",
+        "Rover Component",
+        "Rover Component",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+    ],
+    "mAP": [
+        78.5, 80.2, 82.1, 79.3, 77.8,
+        75.6, 77.3, 78.9, 76.7, 75.1,
+        73.4, 75.1, 76.7, 74.5, 73.0,
+    ],
+    "IoU": [
+        0.72, 0.74, 0.76, 0.73, 0.71,
+        0.69, 0.71, 0.73, 0.70, 0.68,
+        0.67, 0.69, 0.71, 0.68, 0.67,
+    ],
+}
+
+SEGMENTATION_DATA = {
+    "Model": [
+        "U-Net",
+        "DeepLabV3+",
+        "Mask R-CNN",
+        "SegFormer",
+        "HRNet",
+        "U-Net",
+        "DeepLabV3+",
+        "Mask R-CNN",
+        "SegFormer",
+        "HRNet",
+        "U-Net",
+        "DeepLabV3+",
+        "Mask R-CNN",
+        "SegFormer",
+        "HRNet",
+    ],
+    "Dataset": [
+        "Mars Terrain",
+        "Mars Terrain",
+        "Mars Terrain",
+        "Mars Terrain",
+        "Mars Terrain",
+        "Dust Storm",
+        "Dust Storm",
+        "Dust Storm",
+        "Dust Storm",
+        "Dust Storm",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+        "Geological Feature",
+    ],
+    "Dice Score": [
+        0.85, 0.87, 0.88, 0.86, 0.84,
+        0.82, 0.84, 0.85, 0.83, 0.82,
+        0.81, 0.83, 0.84, 0.82, 0.81,
+    ],
+    "IoU": [
+        0.74, 0.76, 0.78, 0.75, 0.73,
+        0.70, 0.72, 0.74, 0.71, 0.70,
+        0.68, 0.70, 0.72, 0.69, 0.68,
+    ],
+}
+
+TASK_DATA = {
+    "Classification": (CLASSIFICATION_DATA, ["Accuracy", "F1-Score"]),
+    "Object Detection": (DETECTION_DATA, ["mAP", "IoU"]),
+    "Segmentation": (SEGMENTATION_DATA, ["Dice Score", "IoU"]),
+} 

app/leaderboard.py

@@ -0,0 +1,71 @@
+"""Leaderboard visualization functions."""
+
+import pandas as pd
+import plotly.express as px
+from typing import Dict, List, Tuple
+from plotly.graph_objects import Figure
+
+from app.data import TASK_DATA
+
+def create_leaderboard_table(data: Dict[str, List], selected_models: List[str] = None) -> pd.DataFrame:
+    """Create a formatted DataFrame for the leaderboard."""
+    df = pd.DataFrame(data)
+    
+    if selected_models:
+        df = df[df['Model'].isin(selected_models)]
+    
+    # Format numeric columns to 1 decimal place
+    numeric_cols = df.select_dtypes(include=['float64', 'int64']).columns
+    df[numeric_cols] = df[numeric_cols].apply(lambda x: x.apply(lambda y: f"{y:.1f}"))
+    
+    return df
+
+def create_performance_plot(data: Dict[str, List], task: str, metric: str, selected_models: List[str] = None) -> Figure:
+    """Create a bar plot showing model performance."""
+    df = pd.DataFrame(data)
+    
+    if selected_models:
+        df = df[df['Model'].isin(selected_models)]
+    
+    fig = px.bar(
+        df,
+        x="Model",
+        y=metric,
+        color="Dataset",
+        title=f"{task} - {metric}",
+        barmode="group",
+    )
+    fig.update_layout(
+        xaxis_title="Model",
+        yaxis_title=metric,
+        showlegend=True,
+    )
+    return fig
+
+def get_best_models(data: Dict[str, List], metrics: List[str]) -> pd.DataFrame:
+    """Get the best performing models for each metric across datasets."""
+    df = pd.DataFrame(data)
+    best_models = []
+    
+    for metric in metrics:
+        top_models = df.groupby('Model')[metric].mean().sort_values(ascending=False).head(3)
+        best_models.extend([
+            {
+                'Metric': metric,
+                'Rank': rank,
+                'Model': model,
+                'Average Score': f"{score:.1f}"
+            }
+            for rank, (model, score) in enumerate(top_models.items(), 1)
+        ])
+    
+    return pd.DataFrame(best_models)
+
+def update_leaderboard(task: str, selected_models: List[str] = None) -> Tuple[pd.DataFrame, Figure, Figure, pd.DataFrame]:
+    """Update the leaderboard based on selected task and models."""
+    data, metrics = TASK_DATA[task]
+    return (
+        create_leaderboard_table(data, selected_models),
+        *[create_performance_plot(data, task, metric, selected_models) for metric in metrics],
+        get_best_models(data, metrics)
+    ) 

pyproject.toml

@@ -0,0 +1,37 @@
+[tool.poetry]
+name = "mars-vision-leaderboard"
+version = "0.1.0"
+description = "A comprehensive leaderboard for evaluating computer vision models on Mars-specific datasets"
+authors = ["Your Name <[email protected]>"]
+readme = "README.md"
+packages = [{include = "app"}]
+
+[tool.poetry.dependencies]
+python = ">=3.9,<3.13"
+gradio = "^4.19.2"
+pandas = "^2.2.0"
+numpy = "^1.26.0"
+plotly = "^5.18.0"
+
+[tool.poetry.group.dev.dependencies]
+black = "^24.1.0"
+isort = "^5.13.0"
+flake8 = "^7.0.0"
+pytest = "^8.0.0"
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.poetry.scripts]
+mars-vision-leaderboard = "run:main"
+
+[tool.black]
+line-length = 88
+target-version = ['py39']
+include = '\.pyi?$'
+
+[tool.isort]
+profile = "black"
+multi_line_output = 3
+line_length = 88 

requirements.txt

@@ -0,0 +1,4 @@
+gradio>=4.19.2
+pandas>=2.2.0
+numpy>=1.26.0
+plotly>=5.18.0 

run.py

@@ -0,0 +1,11 @@
+"""Entry point for the Mars Vision Leaderboard application."""
+
+from app.app import create_interface
+
+def main():
+    """Run the Mars Vision Leaderboard application."""
+    demo = create_interface()
+    demo.launch()
+
+if __name__ == "__main__":
+    main()