tracebloc-package 0.7.0

Build AI with anyone. On data that can't move. SDK for the tracebloc collaborative AI workspace.

tracebloc

Build AI with anyone. On data that can't move.

tracebloc is a collaborative AI workspace you deploy on your own infrastructure. Invite researchers, partners, vendors, or your own teams to train, fine-tune, and benchmark models on your private data — without the data ever leaving your environment.

Install

Pick the extra for your ML framework — the default install contains the core SDK only (~140 MB, ~30 sec) instead of every framework (~8 GB).

pip install "tracebloc_package[pytorch]"      # most users
pip install "tracebloc_package[tensorflow]"   # TensorFlow
pip install "tracebloc_package[sklearn]"      # scikit-learn only
pip install "tracebloc_package[boosting]"     # XGBoost / CatBoost / LightGBM
pip install "tracebloc_package[survival]"     # lifelines / scikit-survival
pip install "tracebloc_package[all]"          # everything

Upgrading from 0.6.32 or earlier? See MIGRATION.md.

Quick Start

from tracebloc_package.user import User

# 1. Log in to your workspace
user = User()
user.login()

# 2. Upload your model to a use case
user.upload_model(model_name="my_model")

# 3. Link your model to the dataset — returns a training-plan facade
training_plan = user.link_model_dataset(dataset_id="<your-dataset-id>")

# 4. Start training
training_plan.start()

For a full walkthrough, open the Quickstart Notebook on Google Colab.

Supported Frameworks

Framework	Use Cases
PyTorch	Image classification, object detection, semantic segmentation, tabular, text classification, time series, keypoint detection, survival analysis
TensorFlow	Image classification, tabular classification
scikit-learn	Tabular classification, tabular regression
XGBoost	Tabular classification, tabular regression
CatBoost	Tabular classification, tabular regression
LightGBM	Tabular classification, tabular regression
lifelines	Survival analysis (time-to-event)
scikit-survival	Survival analysis (time-to-event)

How It Works

1. Deploy a tracebloc workspace on any machine or Kubernetes cluster
2. Define a use case — select datasets, set evaluation metrics
3. Invite anyone — researchers, partners, your own teams across locations
4. Build — contributors train models inside your environment using this SDK
5. Compare — every submission benchmarked under identical conditions on one leaderboard

Local development setup

To work on the tracebloc_package and test changes with the "start-training" Jupyter Notebook (also running locally), clone the start-training repository and set up a virtual environment using a Python version compatible with TensorFlow (ideally 3.11 or 3.12).

Set up virtual environment:

python3.12 -m venv {venv_location}
source {venv_location}/bin/activate
pip install jupyter

Or with Pyenv:

pyenv virtualenv 3.12 start-training
pyenv activate start-training
pip install jupyter

Register the venv as a Jupyter kernel so the notebook uses it:

python -m ipykernel install --user --name=start-training --display-name "start-training"

Option 1: Editable install (recommended for iterating on source)

pip install -e {path_to}/tracebloc_package

Source changes are picked up immediately — no reinstall needed. Best for day-to-day development.

Option 2: Local install (non-editable)

pip install {path_to}/tracebloc_package

Installs a snapshot into site-packages. You must re-run this command after every change. Useful for verifying the package works when installed normally.

Option 3: Install from a built distribution

Build a wheel or sdist first, then install it:

cd {path_to}/tracebloc_package
python -m build
cd {path_to}/start-training
pip install {path_to}/tracebloc_package/dist/tracebloc_package-*.whl

Use this to catch packaging issues (missing files, incorrect metadata, incomplete install_requires) before publishing to PyPI.

Switching between options

pip install on the same package name replaces the previous installation automatically. To remove the package entirely:

pip uninstall tracebloc_package -y

Running the notebook

jupyter notebook notebooks/traceblocTrainingGuide.ipynb

Select the "start-training" kernel if it isn't already selected.

Note: skip the !pip install tracebloc_package[pytorch] cell in the notebook — the package is already installed locally via one of the options above.

Links

• Documentation
• Quickstart Notebook
• Website
• GitHub

License

MIT