Skip to content

Installation#

neat is distributed as a Python package. It is intended to be used in a notebook environment such as, for example, Jupyter Notebooks. Another notebook environment is the CDF notebooks.

CDF Notebooks Environment#

CDF Notebooks are a part of the Cognite Data Fusion (CDF) platform. These notebooks are a great way to get started with neat, even if you have no coding experience.

Limitations

CDF Notebooks are running in your browser. This have some limitations compared to running neat locally. The main difference is that locally neat can use a more powerful storge backend. This means that if you are working with large amounts of metadata, you might want to consider running neat locally.

  1. Go to Cognite Data Fusion
  2. Login to your account
  3. Ensure you have selected the Data Management workspace.
  4. Select Build solutions in your left menubar.
  5. Click on Jupyter Notebooks under the expanded Build solutions menu.
  6. Launch a new notebook.
  7. Install cognite-neat by running %pip install cognite-neat in a cell.
  8. Import NeatSession and CogniteClient and start using it as shown below
from cognite.client import CogniteClient
from cognite.neat import NeatSession

client = CogniteClient()

neat = NeatSession(client)

# Start using neat by typing neat.<TAB>

Local Notebook Environment#

Running neat locally requires a Python environment as well as a notebook environment. The following steps will guide youy through the installation process using Jupyter Lab as the notebook environment.

Prerequisites: Installed Python 3.10 or later, see python.org

  1. Create and enter directory for neat installation
  2. Create a virtual environment:
  3. Activate your virtual environment
  4. Install cognite-neat
  5. Install a notebook environment, pip install jupyterlab
  6. Start your notebook environment, jupyter lab
  7. Import NeatSession and get_cognite_client and start using it

mkdir neat && cd neat

python -m venv venv

venv\Scripts\activate.bat

pip install cognite-neat

pip install pip install jupyterlab

jupyter lab


mkdir neat && cd neat

python -m venv venv

source venv/bin/activate

pip install cognite-neat

pip install pip install jupyterlab

jupyter lab

In a notebook cell, you can now import NeatSession and get_cognite_client and start using it as shown below

from cognite.neat import NeatSession, get_cognite_client

client = get_cognite_client(".env")

neat = NeatSession(client)

# Start using neat by typing neat.<TAB>

Helper get_cognite_client function

The get_cognite_client function is a helper function that reads the environment variables from a .env file and creates a CogniteClient instance. This is a common pattern when working with Cognite Data Fusio through Python. Note that if you dont' have a .env file, it will prompt you to enter environment variables interactively and offer to save them to a .env file. You can instantiate a CogniteClient instance directly if you prefer.