`lmp.script.gen_txt`#

Use pre-trained language model checkpoint to generate continual text of given text segment.

One must first run the script lmp.script.train_model before running this script. This script use pre-trained language model checkpoint to generate continual text of given text segment. Most inference (generation) methods are stochastic process, only some are deterministic.

See also

lmp.infer: All available inference methods.
lmp.model: All available language models.
lmp.script.train_model: Train language model.

Examples

The following example use "Hello world" as conditioned text segment to generate continual text with pre-trained language model experiment my_model_exp. It use top-1 inference method to generate continual text.

python -m lmp.script.gen_txt top-1 \
  --ckpt 5000 \
  --exp_name my_model_exp \
  --max_seq_len 128 \
  --txt "Hello world"

The following example use the same conditioned text segment but inferencing with top-k inference method.

python -m lmp.script.gen_txt top-1 \
  --ckpt 5000 \
  --exp_name my_model_exp \
  --k 10 \
  --max_seq_len 128 \
  --txt "Hello world"

You can use -h or --help options to get a list of available inference methods.

python -m lmp.script.gen_txt -h

You can use -h or --help options on a specific inference method to get a list of supported CLI arguments.

python -m lmp.script.gen_txt top-k -h

lmp.script.gen_txt.main(argv: List[str]) → None[source]

Script entry point.

Parameters: argv (list[str]) – List of CLI arguments.
Return type: None

lmp.script.gen_txt.parse_args(argv: List[str]) → Namespace[source]

Parse CLI arguments.

Parameters: argv (list[str]) – List of CLI arguments.

lmp.script.gen_txt#

`lmp.script.gen_txt`#