Skip to content

Latest commit

 

History

History
69 lines (54 loc) · 3.51 KB

README.md

File metadata and controls

69 lines (54 loc) · 3.51 KB
Raw

Language Identification Transform

Please see the set of transform project conventions for details on general project conventions, transform configuration, testing and IDE set up.

Summary

This transform will identify language of each text with confidence score with fasttext language identification model. ref

Configuration and command line Options

The set of dictionary keys holding LangIdentificationTransform configuration for values are as follows:

Key name Default Description
model_credential unset specifies the credential you use to get model. This will be huggingface token. Guide to get huggingface token
model_kind unset specifies what kind of model you want to use for language identification. Currently, only fasttext is available.
model_url unset specifies url that model locates. For fasttext, this will be repo nme of the model, like facebook/fasttext-language-identification
content_column_name contents specifies name of the column containing documents
output_lang_column_name lang specifies name of the output column to hold predicted language code
output_score_column_name score specifies name of the output column to hold score of prediction

Running

Launched Command Line Options

The following command line arguments are available in addition to the options provided by the python launcher.

  --lang_id_model_credential LANG_ID_MODEL_CREDENTIAL   the credential you use to get model. This will be huggingface token.
  --lang_id_model_kind LANG_ID_MODEL_KIND   what kind of model you want to use for language identification. Currently, only `fasttext` is available.
  --lang_id_model_url LANG_ID_MODEL_URL   url that model locates. For fasttext, this will be repo name of the model, like `facebook/fasttext-language-identification`
  --lang_id_content_column_name LANG_ID_CONTENT_COLUMN_NAME   A name of the column containing documents
  --lang_id_output_lang_column_name LANG_ID_OUTPUT_LANG_COLUMN_NAME   Column name to store identified language
  --lang_id_output_score_column_name LANG_ID_OUTPUT_SCORE_COLUMN_NAME   Column name to store the score of language identification

These correspond to the configuration keys described above.

Running the samples

To run the samples, use the following make targets

  • run-cli-sample - runs src/lang_id_transform.py using command line args
  • run-local-sample - runs src/lang_id_local.py

These targets will activate the virtual environment and set up any configuration needed. Use the -n option of make to see the detail of what is done to run the sample.

For example,

make run-cli-sample
...

Then

ls output

To see results of the transform.

Troubleshooting guide

For M1 Mac user, if you see following error during make command, error: command '/usr/bin/clang' failed with exit code 1, you may better follow this step

Transforming data using the transform image

To use the transform image to transform your data, please refer to the running images quickstart, substituting the name of this transform image and runtime as appropriate.