Sentence Similarity
sentence-transformers
Safetensors
English
bert
mteb
Sentence Transformers
Eval Results (legacy)
text-embeddings-inference
Instructions to use rounit786757/e5-large-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use rounit786757/e5-large-v2 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("rounit786757/e5-large-v2") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Checkpoint System
This directory stores runtime checkpoint files used to resume index building after interruptions.
Files
| File | Purpose |
|---|---|
experiences_checkpoint.json |
Saves parsed catalog records after data ingestion (from scripts/build_index.py) |
embedding_checkpoint.npy |
Saves partial E5 embeddings during batch encoding (from core/indexer.py) |
Auto-Checkpoint Behavior
Experiences Checkpoint
- Created after all data files are parsed into memory
- Resume - if
experiences_checkpoint.jsonexists on next run, data ingestion is skipped entirely - Removed automatically after a successful full build
Embedding Checkpoint
- Created after each batch during the E5 encoding loop
- Resume - if
embedding_checkpoint.npyexists, encoding resumes from the last completed batch - Removed automatically after a successful full build
Manual Control
To force a complete fresh build from scratch:
# Remove checkpoints manually
Remove-Item -LiteralPath "index\experiences_checkpoint.json" -ErrorAction SilentlyContinue
Remove-Item -LiteralPath "index\embedding_checkpoint.npy" -ErrorAction SilentlyContinue
Then re-run:
python scripts/build_index.py