Kedro Tf Utils 🧰

Provides Kedro pipeline components for multi-modal fusion and training for text and image models for healthcare as in X-rays and radiology reports. Use this template to import these pipelines and perform multi-modal ML.

How to install

pip install git+https://github.com/dermatologist/kedro-tf-utils.git

How to build fusion model

from from kedro_tf_utils.pipelines.fusion.pipelines import create_fusion_pipeline

# Add any number of models after parameters
fusion_inputs = {
    "parameters": "params:fusion",
    "tabular_model": "tabular_model",
    "bert_model": "bert_model",
    "text_model": "text_model"
    "image_model": "chexnet_model",
}
fusion_pipeline = create_fusion_pipeline(**fusion_inputs)

# In pipeline_registry
"__default__": other_pipelines + fusion_pipeline

How to build training pipeline

from from kedro_tf_utils.pipelines.train.pipelines import create_train_pipeline

# model and outputs are required
args = {"parameters": "params:train", "model": "fusion_model",
                  "bert_data": "text_data", "image_data": "image_data", "outputs": "trained_model"}
train_pipeline = create_train_pipeline(**args)
# In pipeline_registry
"__default__": train_pipeline

How to build embedding pipeline

(For use with kedro-graph)

from from kedro_tf_utils.pipelines.embedding.pipelines import create_embedding_pipeline

# model and outputs are required
# model and data should be of same type.
# output is a PickleDataset
args = {"parameters": "params:embedding", "model": "tabular_model",
                  "tabular_data": "tabular_data", "outputs": "embeddings"}
embedding_pipeline = create_embedding_pipeline(**args)
# In pipeline_registry
"__default__": create_embedding_pipeline

Catalogue

params:fusion

VISUALIZE: /path/to/save/model/structure.png

params:train

CALLBACKS: /path/to/save/checkpoints
SERVING: /path/to/save/new/model/with/tf/serving/b64/image/layer

TODO

Checkpoints and model wrapped for TF serving b64 image string support only supports local folders currently.

Troubleshoot

BERT model has some issues related to tensorflow_text

Downloaded BERT models will not copy vocab.txt in assets folder to the newly created fusion model. This has to be manually copied.
The class_num in TfModelWeights must be equal to to NCLASSES during training. Otherwise it throws an error: Tensorflow estimator ValueError: logits and labels must have the same shape ((?, 1) vs (?,))

Contributors

Bell Eapen |

Kedro Overview

This is your new Kedro project, which was generated using Kedro 0.18.1.

Take a look at the Kedro documentation to get started.

Rules and guidelines

In order to get the best out of the template:

Don't remove any lines from the .gitignore file we provide
Make sure your results can be reproduced by following a data engineering convention
Don't commit data to your repository
Don't commit any credentials or your local configuration to your repository. Keep all your credentials and local configuration in conf/local/

How to install dependencies

Declare any dependencies in src/requirements.txt for pip installation and src/environment.yml for conda installation.

To install them, run:

pip install -r src/requirements.txt

How to run your Kedro pipeline

You can run your Kedro project with:

kedro run

How to test your Kedro project

Have a look at the file src/tests/test_run.py for instructions on how to write your tests. You can run your tests as follows:

kedro test

To configure the coverage threshold, go to the .coveragerc file.

Project dependencies

To generate or update the dependency requirements for your project:

kedro build-reqs

This will pip-compile the contents of src/requirements.txt into a new file src/requirements.lock. You can see the output of the resolution by opening src/requirements.lock.

After this, if you'd like to update your project requirements, please update src/requirements.txt and re-run kedro build-reqs.

Further information about project dependencies

How to work with Kedro and notebooks

Note: Using kedro jupyter or kedro ipython to run your notebook provides these variables in scope: context, catalog, and startup_error.

Jupyter, JupyterLab, and IPython are already included in the project requirements by default, so once you have run pip install -r src/requirements.txt you will not need to take any extra steps before you use them.

Jupyter

To use Jupyter notebooks in your Kedro project, you need to install Jupyter:

pip install jupyter

After installing Jupyter, you can start a local notebook server:

kedro jupyter notebook

JupyterLab

To use JupyterLab, you need to install it:

pip install jupyterlab

You can also start JupyterLab:

kedro jupyter lab

IPython

And if you want to run an IPython session:

kedro ipython

How to convert notebook cells to nodes in a Kedro project

You can move notebook code over into a Kedro project structure using a mixture of cell tagging and Kedro CLI commands.

By adding the node tag to a cell and running the command below, the cell's source code will be copied over to a Python file within src/<package_name>/nodes/:

kedro jupyter convert <filepath_to_my_notebook>

Note: The name of the Python file matches the name of the original notebook.

Alternatively, you may want to transform all your notebooks in one go. Run the following command to convert all notebook files found in the project root directory and under any of its sub-folders:

kedro jupyter convert --all

How to ignore notebook output cells in `git`

To automatically strip out all output cell contents before committing to git, you can run kedro activate-nbstripout. This will add a hook in .git/config which will run nbstripout before anything is committed to git.

Note: Your output cells will be retained locally.

Package your Kedro project

Further information about building project documentation and packaging your project

dermatologist/kedro-tf-utils