ModSpy-Data

Overview

This is your new Kedro project, which was generated using Kedro 0.18.12.

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/

Help with Compute Canada Clusters

1. Remove deafult modules: module --force purge
2. Activate StdEnv/2020 module which is a pre-requisite for gcc: module load StdEnv/2020
3. Activate following modules: module load gcc/9.3 python/3.11.5 cuda/11.8.0 arrow/12.0.1 nodejs rust/1.70.0 scipy-stack/2023b before actiavating your environment. A module list should look like this:

Currently Loaded Modules:
  1) CCconfig                 5) mii/1.1.2            9) python/3.8.10    (t)    13) gdrcopy/2.3
  2) gentoo/2020     (S)      6) gcccore/.9.3.0 (H)  10) cudacore/.11.7.0 (H,t)  14) ucx/1.8.0
  3) imkl/2020.1.217 (math)   7) gcc/9.3.0      (t)  11) cuda/11.7        (t)    15) libfabric/1.15.1
  4) StdEnv/2020     (S)      8) libffi/3.3          12) arrow/9.0.0      (t)    16) openmpi/4.0.3    (m)

4. Activate your environment: source activate <env_name>
5. Install dependancy: pip install xarray optuna wandb comet_ml lightning openpyxl loguru tqdm argparse hiplot plotly matplotlib umap-learn networkx ray[train,tune,data]
6. For FAISS, I build from source.

Next, if you want to run the jupyter notebook you can do so by requesting an interactive allocation. Here running using kedro jupyter lab to allow contextualization of the project and configuration - salloc --time=02:28:80 --ntasks=1 --cpus-per-task=1 --mem-per-cpu=8G --account=def-mtarailo kedro jupyter lab --ip $(hostname -f) --no-browser salloc --time=1:0:0 --ntasks=1 --cpus-per-task=1 --mem-per-cpu=4G --account=def-mtarailo srun $VIRTUAL_ENV/bin/jupyterlab.sh

To get allocation with GPU: salloc --time=02:59:00 --nodes=1 --ntasks=1 --mem=32G --gres=gpu:v100:1 --constraint=cascade,v100 --account=def-mtarailo srun $VIRTUAL_ENV/bin/jupyterlab.sh

Data preperation part currently relies on old modules and libraries. To run the data preperation part, you need to activate the following modules:

1. module load StdEnv/2020
2. module load gcc/9.3.0 python/3.8.10 cuda/11.7 arrow/9.0.0
3. source ~/jupyter_py3/bin/activate
4. salloc --time=8:28:80 --ntasks=1 --cpus-per-task=1 --mem-per-cpu=12G --account=def-mtarailo kedro jupyter lab --ip $(hostname -f) --no-browser

Before you start

Create a virtual environment with python 3.8.10 version.

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

I used following commands to install PyTorch related packages on ComputeCanada's Graham cluster:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 pip install torch_geometric pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.0.0+cu118.html

How to run your Kedro pipeline

You can run your Kedro project with:

kedro run

If on a compute canada cluster and want to utilize the SLURM runner, you can run the pipeline using the following command:

kedro run --runner="modspy_data.runner.SLURMRunner"

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: catalog, context, pipelines and session.

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