Quickstart
The following steps require an active account. If you don't already have one, create an account.
Install the python package
The python package is the core client to connect with the lab. The only prerequisite for installing pyhectiqlab
is Python itself. Then, install the client with
pip install pyhectiqlab
You can validate its been installed with
python -c "import pyhectiqlab; print(pyhectiqlab.__version__)"
Login in the python client
Open a console and type
hectiqlab add-profile
Your username and password will be asked. If you don't already have an account, create an account. A secret API key will be generated and stored in ~/.hectiqlab/credentials
. You are now authentificated and ready to start.
Create a run
Th next step requires an existing project. If you haven't created a project or join an existing one, then you must create a project here. If you don't know whetheryou have a project or not, you can consult your projects here.
- Open a python script or jupyter notebook.
- Copy the snippet below and change
project
with your project path (format{owner}/{name}
).
from pyhectiqlab.run import Run
# Set project with the project path.
run = Run(name="My first run", project=project)
If a run with this name doesn't already exist, it will be created. If a run with this name already exists, it will connect to the existing run and the data will be pushed in the existing run. However, if you attach to an existing run that you are not the original author (you didn't created it), your run
will be in a read-only mode.
You can print the run status which will display the run url
print(run)
"""
<Run 59>
project : my_project
author : John Smith
mode : read-write
url : https://lab.hectiq.ai/my_project/runs/my-first-run
"""
Push data
Once the run is created, you can play with all the awesome features such as datasets, models, metrics, apps, and many more. Try modifying the run status and logging your pyhectiqlab
version:
run.pending()
import pyhectiqlab
run.add_package_version(globals())