Model Card for Model ID
This model is a binary classifier that predicts if a pokemon card is a collector's item. It is trained with Autogluon tabular on the ecopus/pokemon_cards dataset.
Model Details
Model Description
This model is a binary classifier that predicts if a pokemon card is a collector's item. It is trained with Autogluon tabular on the ecopus/pokemon_cards dataset.
- Developed by: Sam Der
- Model type: AutoML (AutoGluon Tabular)
- License: MIT
Uses
Direct Use
This model is intended to be used to predict if a pokemon card is a collector's item or not based on other metrics including market value, art type, and condition.
Bias, Risks, and Limitations
- small dataset may not produce accurate results
Training Details
Training Data
- dataset: ecopus/pokemon_cards
- splits:
- original (34 rows)
- augmented (300 rows)
- target column: "Collector's Item"
Training Procedure
- library: AutoGluon Tabular
- time_limit: 300 seconds
- presets: "best_quality"
Training Hyperparameters
- time_limit=300
- presets="best_quality"
Evaluation
Testing Data, Factors & Metrics
Testing Data
ecopus/pokemon_cards
Metrics
- accuracy: fraction of correctly predicted labels
- F1 (weighted): harmonic mean of precision and recall, weighted by class support
Results
Accuracy: 0.8235 | Weighted F1: 0.8135
