MCPcopy Create free account
hub / github.com/UCSC-VLAA/OpenVision / encode

Method encode

src/convert_upload/open_clip/tokenizer.py:214–220  ·  view source on GitHub ↗
(self, text)

Source from the content-addressed store, hash-verified

212 return word
213
214 def encode(self, text):
215 bpe_tokens = []
216 text = self.clean_fn(text)
217 for token in re.findall(self.pat, text):
218 token = ''.join(self.byte_encoder[b] for b in token.encode('utf-8'))
219 bpe_tokens.extend(self.encoder[bpe_token] for bpe_token in self.bpe(token).split(' '))
220 return bpe_tokens
221
222 def decode(self, tokens):
223 text = ''.join([self.decoder[token] for token in tokens])

Callers 3

__call__Method · 0.95
tokenizeMethod · 0.80

Calls 1

bpeMethod · 0.95

Tested by

no test coverage detected