spelling

in the main script file.

This means we don't need to run the script to obtain its

arguments, so we can add them to the database even if the

run has not started yet, e.g., because it is in the queued

run has not started yet, e.g., because it is in the queue

of a batch scheduling system.

Here's a simple :source:`example <examples/argument_parsing.py>`:

    "\n",

    "This notebook demonstrates how to use SQLTrack to replicate\n",

    "and improve on some features found in other tools, like the mlflow UI.\n",

    "You can find the original in notebook at `examples/notebook.ipynb`.\n",

    "You can find the original notebook at \n",

    "[examples/notebook.ipynb](https://gitlab.com/jfolz/sqltrack/-/blob/main/examples/notebook.ipynb).\n",

    "The data displayed here is generated by\n",

    "`examples/generate_experiment_data.py`.\n",

    "[examples/generate_experiment_data.py](https://gitlab.com/jfolz/sqltrack/-/blob/main/examples/generate_experiment_data.py).\n",

    "\n",

    "We will display lists of experiments and runs,\n",

    "show metrics and plot them,\n",

    "Let's start by setting up our environment.\n",

    "We use [itables](https://mwouts.github.io/itables) to display\n",

    "interactive tables and [Plotly](https://plotly.com/python/) for plots.\n",

    "The config for our SQLTrack `Client` is loaded from `./sqltrack.conf`."

    "The config for our SQLTrack\n",

    "[Client](generated/sqltrack.rst#sqltrack.Client)\n",

    "is loaded from `./sqltrack.conf`."

   ]

  },

  {

    "## Experiments\n",

    "\n",

    "First, we will show all experiments in the database.\n",

    "We use `query_dataframe` to run a query against the\n",

    "We use\n",

    "[query_dataframe](generated/sqltrack.pandas.rst#sqltrack.pandas.query_dataframe)\n",

    "to run a query against the\n",

    "database and pack all returned rows into a Pandas DataFrame.\n",

    "`format_dataframe` applies some HTML & CSS formatting\n",

    "[format_dataframe](generated/sqltrack.notebook.rst#sqltrack.notebook.format_dataframe)\n",

    "applies some HTML & CSS formatting\n",

    "to the columns.\n",

    "E.g., timestamps are converted to human-friendly\n",

    "relative time strings with the\n",

    "for the step that achieve maximum top-1 accuracy.\n",

    "For performance reasons, mlflow always display the latest metric\n",

    "value, which is not necessarily the best.\n",

    "We also see the Postgres syntax to interactiv with JSONB columns,\n",

    "We also see the PostgreSQL syntax to interactiv with JSONB columns,\n",

    "e.g. `tags ? 'marked'` to check if a key is present and\n",

    "`env->'GIT_COMMIT'` to extract a value.\n",

    "Check their\n",

    "for more details on all the different things you can do with\n",

    "JSONB columns.\n",

    "\n",

    "We again use `format_dataframe` for some nice formatting,\n",

    "this time of run state, progress, and tags.\n",

    "Additionally, we also tell it to format `val_top1` as a percentage."

    "We again use\n",

    "[format_dataframe](generated/sqltrack.notebook.rst#sqltrack.notebook.format_dataframe)\n",

    "for some nice formatting, but additionally tell it that we want\n",

    "`val_top1` formatted as a percentage."

   ]

  },

  {

   "name": "python",

   "nbconvert_exporter": "python",

   "pygments_lexer": "ipython3",

   "version": "3.10.6"

   "version": "3.10.6 (main, Nov 14 2022, 16:10:14) [GCC 11.3.0]"

  },

  "vscode": {

   "interpreter": {