Skip to content

Latest commit

 

History

History
188 lines (147 loc) · 9.84 KB

README.md

File metadata and controls

188 lines (147 loc) · 9.84 KB

Transforms

The transformation framework is designed to operate on rows of columnar data, generally contained in parquet files and read as pyarrow tables.

Transforms are written to process the table to, for example:

  • Annotate the tables to add additional data such as document quality score, language, etc.
  • Filter the table to remove or edit rows and/or columns, for example to remove rows from blocked domain.

While these transformation modules were originally built for pre-training, they are also useful for fine-tuning data preparation.

Annotating Transforms

Annotating transforms examine 1 or more columns of data, typically a content column containing a document to be annotated. The content is often spoken/text or programming language, generally to build a large language model (LLM). Examples of annotation might include:

  • Language identification - an additional string column is added to identify the language of the document content.
  • Document quality - an additional float column is added to associated a quality score with the document.
  • Block listing - an addtional boolean column is added that indicates if the content source url (in one of the columns) is from a blocked domain.

Filtering Transforms

Filtering transforms modify the rows and/or columns, usually based on associated column values.
For example,

  • Language selection - remove rows that do not match the desired language
  • Document quality threshold - remove rows that do not meet a minimum document quality value.
  • Block listing - remove rows that have been flagged as having been sourced from undesirable domains.

Transform Organization

This directory hierarchy of transforms is organized as follows:

  • universal - transforms applicable across code and language model data include
  • language - spoken language model specific transforms
  • code - programming language specific transforms.

Each of the universal, language and code directories contains a directory for a specific transform. Each transform is expected to be a standalone entity that generally runs at scale from within a docker image. As such they each have their own virtual environments for development.

Transform Project Conventions

The transform projects all try to use a common set of conventions including code layout, build, documentation and IDE recommendations. For a transformed named xyz, it is expected to have its project located under one of

transforms/code/xyz
transforms/language/xyz, OR
transforms/universal/xyz.

Makefile

The Makefile is the primary entry point for performing most functions for the build and management of a transform. This includes cleanup, testing, creating the virtual environment, building a docker image and more. Use make help in any directory with a Makefile to see the available targets. Each Makefile generally requires the following macro definitions:

  • REPOROOT - specifies a relative path to the local directory that is the root of the repository.
  • TRANSFORM_NAME - specifies the simple name of the transform that will be used in creating pypi artifacts and docker images.
  • DOCKER_IMAGE_VERSION - sets the version of the docker image and is usually set from one of the macros in .make.versions at the top of the repository

These are used with the project conventions outlined below to build and manage the transform.

Runtime Organization

Transforms support one or more runtimes (e.,g python, Ray, Spark, KFP, etc). Each runtime implementation is placed in a sub-directory under the transform's primary directory, for example:

transforms/universal/xyz/python
transforms/universal/xyz/ray
transforms/universal/xyz/spark
transforms/universal/xyz/kfp

A transform only need implement the python runtime, and the others generally build on this.

All runtime projects are structured as a standard python project with the following:

  • src - directory contains all implementation code
  • test - directory contains test code
  • test-data - directory containing data used in the tests
  • pyproject.toml or requirements.txt (the latter is being phased out)
  • Makefile- runs most operations, try make help to see a list of targets.
  • Dockerfile to build the transform and runtime into a docker image
  • output - temporary directory capturing any test/local run output. Ignored by .gitignore.

A virtual environment is created for the runtime project using make venv.

In general, all runtime-specific python files use an _<runtime>.py> suffix, and docker images use a -<runtime> suffix in their names. For example,

  • noop_transform_python.py
  • test_noop_spark.py
  • dpk-noop-transform-ray

Finally, the command make conventions run from within a runtime directory will examine the runtime project structure and make recommendations.

Python Runtime

The python runtime project contains the core transform implementation and its configuration, along with the python-runtime classes to launch the transform. The following organization and naming conventions are strongly recommended and in some cases required for the Makefile to do its work.

  1. src directory contain python source for the transform with the following naming conventions/requirements.

    • xyz_transform.py generally contains the core transform implementation:
      • XYZTransform class implementing the transformation
      • XYXTransformConfiguration class that defines CLI configuration for the transform
    • xyz_transform_python.py - runs the transform on input using the python runtime
      • XYZPythonTransformConfiguration class
      • main() to start the PythonTransformLauncher with the above.
  2. test directory contains pytest test sources

    • test_xyz.py - a standalone (non-ray launched) transform test. This is best for initial debugging.
      • Inherits from an abstract test class so that to test one needs only to provide test data.
    • test_xyz_python.py - runs the transform via the Python launcher.
      • Again, inherits from an abstract test class so that to test one needs only to provide test data.

    Tests are expected to be run from anywhere and so need to use __file__ location to create absolute directory paths to the data in the ../test-data directory.
    From the command line, make test sets up the virtual environment and PYTHONPATH to include src From the IDE, you must add the src directory to the project's Sources Root (see below). Do not add sys.path.append(...) in the test python code. All test data should be referenced as ../test-data.

Ray/Spark Runtimes

These projects are structured in a similar way and replace the python runtime source and test files with the following:

src/xyz_transform_[ray|spark].py * [Ray|Spark]TransformRuntimeConfiguration - runtime configuration class * contains a main() that launches the runtime test/test_xyz_[ray|spark].py - tests the transform running in the given runtime.

Configuration and command line options

A transform generally accepts a dictionary of configuration to control its operation. For example, the size of a table, the location of a model, etc. These are set either explicitly in dictionaries (e.g. during testing) or from the command line when run from a Ray launcher.

When specified on the command line, transform xyz should use an xyz prefix with --xyz_ (dash dash) to define its command line options. For example, --xyz_some_cfg somevalue sets the value for the xyz_some_cfg configuration key value to somevalue. To avoid potential collisions with options for the Ray launcher, Data Access Factory and others, it is strongly encouraged to not use single dash options with a single or small number of characters (e.g. -n).

Release process

The transform versions are managed in a central file named .make.versions. This file is where the versions are automatically propagated to the Makefile rules when building and pushing the transform images. When a new transform version is created, the tag of the transform should be updated in this file. If there is no entry for the transform in the file yet, create a new one and add a reference to it in the transform Makefile, following the format used for other transforms. ore specifically, the entry should be of the following format: <transform image name>_<RUNTIME>_VERSION=<version>, for example: FDEDUP_RAY_VERSION=0.2.77

Building the docker image

Generally to build a docker image, one uses the make image command, which uses the Dockerfile, which in turn uses the src and requirements.txt to build the image. Note that the Makefile defines the TRANSFORM_NAME and DOCKER_IMAGE_VERSION and should be redefined if copying from another transform project.

To build individual transform image use make -C <path to transform directory>, for example: make -C universal/fdedup image. To push all the images run make push, or make -C <path to transform directory> push for individual transform.

IDE Setup

When running in an IDE, such as PyCharm or VS Code, the following are generally required:

  • From the command line, build the venv using make venv.
  • In the IDE
    • Set your project/run configuration to use the venv/bin/python as your runtime virtual environment.
      • In PyCharm, this can be done through the PyCharm->Settings->Project...->Python Interpreter page
      • In VS Code, click on the current Python Interpreter in the bottom right corner and make sure that the Interpreter path is venv/bin/python
    • Mark the src as a source root so that it is included in your PYTHONPATH when running .py files in the IDE
      • In Pycharm this can be done by selecting the src directory, and then selecting Mark Directory as -> Sources Root