You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
speed_comparisons_streaming_10k.py: Streams and filters 100,001 records from a Hugging Face dataset, then benchmarks SymRank, NumPy, and scikit-learn cosine similarity on 10,000 real embeddings.
speed_comparisons_streaming_10k_batch.py: Same data streaming and preparation, but benchmarks batch-mode performance for SymRank, NumPy, and scikit-learn.
Both scripts provide realistic, large-scale performance comparisons using real-world embedding data.