---

Mantra is used by deep learning practitioners to manage their development workflow. It automatically provisions cloud instances for training, tracks and versions experiments, has a UI for training and evaluating results, and works with frameworks like PyTorch and TensorFlow.

Key Features:

• Boilerplate classes for common dataset and model types
• Command-line interface for training with parameter parsing
• Automatic provisioning of cloud instances for remote training
• UI for monitoring training, comparing experiments and storing media
• Encapsulation of datasets and models by design, enabling easy sharing

This is an alpha release. All contributions are welcome - see here for guidelines on how to contribute.

You can read the docs here.

---

Get Started

🚀 To launch your first Mantra project, execute the following to create a new project directory:

mantra launch my_project 

☁ Configure your cloud settings and API keys:

cd my_project 
mantra cloud 

💾 Get the example datasets and models from here:

mantra import https://github.com/RJT1990/mantra-examples.git

🤖 Here are the example models you can train:

mantra train relativistic_gan --dataset decks --cloud --dev --image-dim 256 256

mantra train log_reg --dataset epl_data --target home_win --features feature_1 feature_2 feature_3 

🚂 During training, you can spin up the Mantra UI to track the progress:

mantra ui

Installation

To install mantra, you can use pip:

pip install mantraml

You should also have TensorFlow or PyTorch installed depending on which framework you intend to use.

Mantra is tested on Python 3.5+. It is not currently supported on Windows, but we'll look to get support in the near future.

AWS Dependencies

You will need to install AWS CLI as a dependency.

1. Login to AWS through a browser, click your name in the menubar and click My Security Credentials.

2. Create a new Access Key and make a note of the Access Key ID and Secret Access Key.

3. From terminal enter the following:

johnsmith@computer:~$ pip install awscli
johnsmith@computer:~$ aws configure

Once prompted, enter your AWS details and your default region (e.g. us-east-1).

4. Now your credentials will be accessible by the boto3 AWS SDK library, which will allow Mantra to be used to provision cloud instances on your request.

5. Use mantra cloud from your mantra project root to configure your cloud settings.

You should also ensure you are happy with the default instance settings in mantra - you can check this in the settings.py file in your project root.

Have Fun

Arise! Awake! Approach the great and learn.