The number next to the code cell will change to an asterisk to indicate that the code is executing. Type some code in the notebook, then hold the Shift key and hit the Return key to execute it. To start a Jupyter notebook, select the icon for the Python or Julia version you want to use. The pane on the left is the /work directory of the account you ran the job with. Once JupyterLab is loaded, you will see something like this: Select 'Keep Waiting' unless it has been more than 5 minutes. You may also see an error screen like this. This may take several seconds, so please be patient. Eventually, an animated Jupyter icon should show up. Click the 'Connect to Jupyter' button.Īfter you click the button, you will probably see a white screen. When the job is ready, it will say 'Running'. Please be patient as this process can take a few minutes. Once the job has been assigned to a node, it will say 'Starting'. ![]() ![]() Queue times are the same as for any other pronto job. When you are finished click the 'Launch' button. The only version of Python available will be the one installed in that container. Note: If you select the GPU partition, your job will run within the ml-gpu container. Getting Started ¶įirst, ensure you have followed the directions in the Open OnDemand article to get access to Open OnDemand.Ĭlick the JupyterLab tile on the main OnDemand dashboard.įill out the job submission form to specify what resources you want JupyterLab to have available. Our JupyterLab currently supports Python and Julia. JupyterLab is a web-based interactive development environment.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |