Dataset Viewer
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Deep Hybrid Recommender β Demo Artifacts
Model artifacts for the Deep Hybrid Recommender thesis project (DATN).
Quick Start
pip install huggingface_hub
python scripts/download_demo_artifacts.py
Or manually:
from huggingface_hub import snapshot_download
local_dir = snapshot_download(repo_id="Clync/datn-deep-hybrid-recommender-data", repo_type="dataset")
Contents
artifacts/early_fusion_v3/β Early Fusion V3 model weights + user/item IDs (main deployed model, Recall@10=0.9004)artifacts/sbert/β SBERT v2 item embeddings (768-d, 2.46M items) + FAISS index IDsartifacts/factors/β ALS & BPR factors (32-d) + user/item IDsartifacts/faiss/β FAISS index for SBERT ANN searchartifacts/item_metadata.dbβ SQLite item metadata (725MB)artifacts/user_history.dbβ SQLite user history (343MB)artifacts/item_popularity.npyβ Item popularity scores
Models
- Early Fusion V3 (deployed): Recall@10=0.9004, NDCG@10=0.7305, MRR=0.6819
- RGH v4 (research): HR@10=0.7983 (holdout N=7500)
Source
- GitHub: https://github.com/TnClync/DATN
- Kaggle kernels: minhteonguyn/train-early-fusion-v3-datn, minhteonguyn/train-sbert-v2-datn
- Downloads last month
- 45