CPEN455-24W2 Course Project Leaderboard
๐ PixelCNN++ for Image Classification Leaderboard
Submit your model's predictions on the test set to see how it performs against others! The leaderboard shows results on a hidden test dataset to ensure fair comparison.
The use of the leaderboard is completely optional. You can submit your results without using the leaderboard. However, we highly recommend you to use the leaderboard to iteratively improve your model and write your report.
Please refer to this repo for project details.
Columns:
- Model Name: Name of submitted model
- Accuracy: Percentage of correct predictions
- F1 Score: Harmonic mean of precision & recall
- Submission Date: Date of submission
{
- "headers": [
- "eval_name",
- "Model Name",
- "Accuracy",
- "F1 Score",
- "Student ID",
- "Submission Date"
- "data": [
- [
- "34762484_pixelCNN++",
- "pixelCNN++",
- 0.879,
- 0.879,
- "34762484",
- "2025-03-10T20:50:45Z"
- [
- "Test_dummy-pixelCNN++",
- "dummy-pixelCNN++",
- 0.249,
- 0.099,
- "Test",
- "2025-03-10T09:47:19Z"
- [
- "metadata": null
Evaluation Process
- Participants submit their models' predictions (a CSV file) on image classification tasks
- Results are computed using accuracy and F1-score metrics
- Top performances are shown on this leaderboard
Please refer to this file for a template of the CSV file.
Some Tips for Submission
- Please refer to this file for a template of the CSV file. Please do not change the first column content as it specifies the image ID.
- You may submit your results as many times as you want before the deadline. However, we encourage you to submit your results as early as possible to avoid last-minute server issues.