{"library":"bpemb","type":"library","category":null,"description":"BPEmb provides byte-pair encodings (BPE) from raw text and maps subword units to pre-trained embeddings for 275 languages. It's designed for NLP tasks requiring efficient subword tokenization and embedding. The current version is 0.3.6, with releases occurring sporadically based on updates to models or features.","language":"python","status":"active","version":"0.3.6","tags":["embeddings","nlp","bpe","subword","language-models","vector-representations"],"last_verified":"Tue May 26","install":[{"cmd":"pip install bpemb","imports":["from bpemb import BPEmb"]}],"homepage":"https://nlp.h-its.org/bpemb","github":null,"docs":null,"changelog":null,"pypi":"https://pypi.org/project/bpemb/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"compatibility":{"summary":{"python_range":"3.10–3.9","success_rate":50,"avg_install_s":9.5,"avg_import_s":0.27,"wheel_type":"wheel"},"url":"https://checklist.day/v1/registry/bpemb/compatibility"}}