Overwrite existing outputsīy default, jupyter-runner skip the run when output file(s) already exists. The process create output file s3://bucket/results/notebook.html based on a notebook stored from S3.įiles are downloaded to a local temporary only available to the current user and removed at the end or in case of exceptions. Use S3 inputs/outputs jupyter-runner -output-directory=s3://bucket/results/ s3://bucket/notebooks/notebook.ipynb ``results directory is created if it does not pre-exist. The process create output file results/notebook.html`. Change output directory jupyter-runner -output-directory results notebook.ipynb jupyter-runner -parameter-file=my_parameter_file notebookA.ipynb notebookB.ipynbīy default, the process creates output files notebookA_1.html, notebookA_2.html, notebookB_1.html, notebookB_2.html in current directory. When there are N sets of parameters running on M notebooks, there will be NxM distinct output files. Jupyter-runner can combine multiple set of parameters on multiple notebooks. Run multiple notebooks with multiple sets of parameters Then run jupyter-runner specifying the path to my_parameter_file just created: jupyter-runner -parameter-file=my_parameter_file notebook.ipynbīy default, the process creates output files notebook_1.html and notebook_2.html in current directory. VAR1=VAL5 VAR2=VAL18 VAR3='VAL42 with space' Run notebook with multiple sets of parametersĬreate a file with multiple set of parameters, one set of parameters per line.Įxample file containing 2 sets of 3 parameters: VAR1=VAL1 VAR2=VAL2 VAR3=VA元 The notebook variables passed by the user can impact the rendering of the output. get ( 'ENV_VAR', 'a_default_value' ) # usage of env_var in your codeīy default, the process creates output file notebook.html in current directory. In python notebook, variables can be retrieved using os.environ: import os env_var = os. ENV_VAR=xxx jupyter-runner notebook.ipynb Use environment variables on command-line. Run multiple notebooks jupyter-runner notebookA.ipynb notebookB.ipynbīy default, the process creates output files notebookA.html and notebookB.html in current directory. Tutorial Run a simple notebook jupyter-runner notebook.ipynbīy default, the process creates output file notebook.html in current directory. allow-errors Allow errors during notebook execution. timeout= Cell execution timeout in seconds. overwrite Overwrite output files if they already exist. workers= Maximum number of parallel execution parameter-file=** Optional parameters files containing one parameter instance by line, setting the environment. Installation pip install jupyter-runner Usage jupyter-runner. Notebook execution can happen in parallel with a fixed number of workers. Jupyter runners allows to run multiple notebooks over multiple sets of parameters. Welcome to Jupyter-Runner’s documentation
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |