(s)
| 338 | |
| 339 | # Lowercase, trim, and remove non-letter characters |
| 340 | def normalizeString(s): |
| 341 | s = unicodeToAscii(s.lower().strip()) |
| 342 | s = re.sub(r"([.!?])", r" \1", s) |
| 343 | s = re.sub(r"[^a-zA-Z.!?]+", r" ", s) |
| 344 | s = re.sub(r"\s+", r" ", s).strip() |
| 345 | return s |
| 346 | |
| 347 | # Read query/response pairs and return a voc object |
| 348 | def readVocs(datafile, corpus_name): |
no test coverage detected