Step 6: Using Activeloop Storage
Storing and loading datasets from Deep Lake Storage.
You can store your Deep Lake Datasets with Activeloop by first creating an account in the Deep Lake App or in the CLI using:
activeloop register
In order for the Python API to authenticate with your account, you should log in from the CLI using:
!activeloop login
# Alternatively, you can directly input your username and password in the same line:
# activeloop login -u <your_username> -p <your_password>
You can then access or create Deep Lake Datasets by passing the Deep Lake path to
deeplake.dataset()import deeplake
deeplake_path = 'hub://organization_name/dataset_name'
#'hub://jane_smith/my_awesome_dataset'
ds = deeplake.dataset(deeplake_path)
When you create an account in Deep Lake, a default organization is created that has the same name as your username. You can also create other organizations that represent companies, teams, or other collections of multiple users.
Public datasets such as
'hub://activeloop/mnist-train' can be accessed without logging in.Once you have an Activeloop account, you can create tokens in the Deep Lake App (
Organization Details -> API Tokens) and login in the CLI using the token:!activeloop login --token <your_token>
If you are not logged in through the CLI, you may also pass the token to python commands that require authentication:
ds = deeplake.load(deeplake_path, token = 'xyz')