usage: ct2-fairseq-converter [-h] --model_path MODEL_PATH --data_dir DATA_DIR
[--user_dir USER_DIR] [--fixed_dictionary FIXED_DICTIONARY]
[--source_lang SOURCE_LANG] [--target_lang TARGET_LANG]
[--no_default_special_tokens] --output_dir OUTPUT_DIR
[--vocab_mapping VOCAB_MAPPING]
[--quantization {int8,int8_float32,int8_float16,int8_bfloat16,int16,float16,bfloat16,float32}]
[--force]
options:
-h, --help show this help message and exit
--model_path MODEL_PATH
Model path. (default: None)
--data_dir DATA_DIR Data directory containing the source and target vocabularies. (default:
None)
--user_dir USER_DIR Directory containing custom extensions. (default: None)
--fixed_dictionary FIXED_DICTIONARY
Fixed dictionary for multilingual models. (default: None)
--source_lang SOURCE_LANG
Source language. This argument is used to find dictionary file from
`data_dir`. (default: None)
--target_lang TARGET_LANG
Target language. This argument is used to find dictionary file from
`data_dir`. (default: None)
--no_default_special_tokens
Require all special tokens to be provided by the user during inference,
including the decoder start token. (default: False)
--output_dir OUTPUT_DIR
Output model directory. (default: None)
--vocab_mapping VOCAB_MAPPING
Vocabulary mapping file (optional). (default: None)
--quantization {int8,int8_float32,int8_float16,int8_bfloat16,int16,float16,bfloat16,float32}
Weight quantization type. (default: None)
--force Force conversion even if the output directory already exists. (default:
False)
Options (12)
| Option | Description | Source |
|---|
-h, --help | show this help message and exit | help |
--model_path MODEL_PATH | Model path. | help |
--data_dir DATA_DIR | Data directory containing the source and target vocabularies. | help |
--user_dir USER_DIR | Directory containing custom extensions. | help |
--fixed_dictionary FIXED_DICTIONARY | Fixed dictionary for multilingual models. | help |
--source_lang SOURCE_LANG | Source language. | help |
--target_lang TARGET_LANG | Target language. | help |
--no_default_special_tokens | Require all special tokens to be provided by the user during inference, including the decoder start token. | help |
--output_dir OUTPUT_DIR | Output model directory. | help |
--vocab_mapping VOCAB_MAPPING | Vocabulary mapping file (optional). | help |
--quantization int8 | Weight quantization type. | help |
--force | Force conversion even if the output directory already exists. | help |
Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.