Views
No views yet
| Metric | Value |
|---|---|
| ROC AUC | 0.9899 |
| Precision @ 80% | 80.2% |
| Recall @ 80% | 78.3% |
| F1 (migrated) | 0.66 |
| Training samples | 135,617 |
| Features | 36 |
| Trained | Daily at 6am UTC |
entity_total_migrations — how many times entity has migrated beforeentity_migration_rate — entity success rateinactivity_gap_seconds — dead time between tradesis_quality_dev — archetype classificationdistinct_sellerstotal_sellsis_bot_cluster — bot detectioncurve_pct_filled — bonding curve progressentity_prior — Bayesian prior scoredistinct_buyers1import xgboost as xgb
2from huggingface_hub import hf_hub_download
3
4# Download model
5model_path = hf_hub_download(repo_id="jrhode666/gpt-scorer", filename="xgb_scorer.json")
6
7# Load
8model = xgb.XGBClassifier()
9model.load_model(model_path)
10
11# Predict (36 features required — see xgb_meta.json)
12probability = model.predict_proba(features_df)[:, 1]xgb_meta.json for the full feature list. Core categories:1 if token migrated to DEX (PumpSwap, Raydium, Meteora, Jupiter), 0 otherwise.