You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

README.md 1.7 kB

8 years ago
7 years ago
7 years ago
12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758
  1. # FastNLP
  2. ```
  3. FastNLP
  4. │ LICENSE
  5. │ README.md
  6. │ requirements.txt
  7. │ setup.py
  8. |
  9. ├─docs (documentation)
  10. |
  11. └─tests (unit tests, intergrating tests, system tests)
  12. | │ test_charlm.py
  13. | │ test_loader.py
  14. | │ test_trainer.py
  15. | │ test_word_seg.py
  16. | │
  17. | └─data_for_tests (test data used by models)
  18. | charlm.txt
  19. | cws_test
  20. | cws_train
  21. |
  22. └─fastNLP
  23. ├─action (model independent process)
  24. │ │ action.py (base class)
  25. │ │ README.md
  26. │ │ tester.py (model testing, for deployment and validation)
  27. │ │ trainer.py (main logic for model training)
  28. │ │ __init__.py
  29. │ │
  30. |
  31. ├─loader (file loader for all loading operations)
  32. │ | base_loader.py (base class)
  33. │ | config_loader.py (model-specific configuration/parameter loader)
  34. │ | dataset_loader.py (data set loader, base class)
  35. │ | embed_loader.py (embedding loader, base class)
  36. │ | __init__.py
  37. ├─model (definitions of PyTorch models)
  38. │ │ base_model.py (base class, abstract)
  39. │ │ char_language_model.py (derived class, to implement abstract methods)
  40. │ │ word_seg_model.py
  41. │ │ __init__.py
  42. │ │
  43. ├─reproduction (code library for paper reproduction)
  44. │ ├─Char-aware_NLM
  45. │ │
  46. │ ├─CNN-sentence_classification
  47. │ │
  48. │ └─HAN-document_classification
  49. ├─saver (file saver for all saving operations)
  50. │ base_saver.py
  51. │ logger.py
  52. │ model_saver.py
  53. ```

一款轻量级的自然语言处理(NLP)工具包,目标是减少用户项目中的工程型代码,例如数据处理循环、训练循环、多卡运行等