> For the complete documentation index, see [llms.txt](https://titanml.gitbook.io/iris-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://titanml.gitbook.io/iris-documentation/getting-started/iris-commands.md).

# Iris commands

{% hint style="danger" %}
These docs are outdated! Please check out <https://docs.titanml.co> for the latest information on the TitanML platform.\
\
If there's anything that's not covered there, please contact us on our [discord](https://discord.com/invite/83RmHTjZgf).
{% endhint %}

[***Also see the API***](broken://pages/MHSKahtYKd1b81f4CIzr)

### Contents

* [iris ](#iris-post)[distil](#iris-post)
* [iris get](#iris-get)
* [iris status](#iris-status)
* [iris upload](#iris-upload)
* [iris download](#iris-download)
* [iris pull](#iris-pull-1)
* [iris makesafe](#iris-makesafe)

By default, these commands will send a request to the backend, then print a table with the response to the command line. If you would like to receive the response as a JSON object, simply add the tag `--json` after the command

### iris distil

Dispatches knowledge distillation jobs to the TitanML platform. You can look at all of the iris distil arguments by using this command: `iris distil --help.`

### iris finetune

Dispatches fine-tuning jobs to the TitanML platform.

For the above two commands, both `model` and `dataset` can be any of:&#x20;

* A HuggingFace model/dataset
* A UUID generated by `iris upload`
* The filepath of a model/dataset folder (which will then be uploaded as part of the job).

More explanation [here](/iris-documentation/titan-train-finetuning-service/using-iris-finetune/a-closer-look-at-iris-finetune-arguments.md), [here](/iris-documentation/titan-optimise-knowledge-distillation/using-iris-distil/a-closer-look-at-iris-distil-arguments.md) on how to launch a distillation or fine-tuning job.

The following arguments apply to both `iris distil` and `iris finetune`:

<table><thead><tr><th width="142">Command</th><th width="126">Shortcut</th><th width="161">Type</th><th>Description</th></tr></thead><tbody><tr><td>--model</td><td>-m</td><td>TEXT</td><td>The model to optimize. [default: None] [required] </td></tr><tr><td>--dataset</td><td>-d</td><td>TEXT</td><td>The dataset to optimize the model with. [default: None] [required]</td></tr><tr><td>--task</td><td>-t</td><td>[sequence_classification | question_answering | token_classification]</td><td>The task to optimize the model for.[default: None] [required]</td></tr><tr><td>--subset</td><td>-ss</td><td>TEXT</td><td>The subset of the dataset to use</td></tr><tr><td>--name</td><td>-n</td><td>TEXT</td><td>The name to use for this job. Visible in the                                                                            titan web interface.    </td></tr><tr><td>--file</td><td>-f</td><td>PATH</td><td>The .yaml file containing experiment parameters.</td></tr><tr><td>--short-run</td><td>-s</td><td><em>flag</em></td><td>Truncates the run after 1 batch and 1 epoch.                                                                            Will provide poor results, but useful to check that the model and dataset choices are valid.                                                      </td></tr><tr><td>--num-labels</td><td>-nl</td><td>INTEGER</td><td>Number of labels. Required for task sequence_classification [default: None]                                                                              </td></tr><tr><td>--text-fields</td><td>-tf</td><td>TEXT</td><td>Text fields. Required for task sequence_classification [default: None]   </td></tr><tr><td>--has-negative</td><td>-hn</td><td><em>flag</em></td><td>Has negative. Required for question_answering</td></tr><tr><td>--label-name</td><td>-ln</td><td>int:TEXT</td><td>The label names used for token classification</td></tr><tr><td>--help</td><td></td><td><em>flag</em></td><td>Show this message and exit.</td></tr></tbody></table>

You can use these additional arguments to specify hyperparameters when running a fine-tuning job:

| Command         | Shortcut | Type    | Description                                             |
| --------------- | -------- | ------- | ------------------------------------------------------- |
| --batch-size    | -bs      | INTEGER | The batch size to use for training. Default is 16.      |
| --learning-rate | -lr      | FLOAT   | The learning rate to use for training. Default is 2e-5. |
| --num-epochs    | -ne      | INTEGER | Number of epochs to finetune for. Default is 1.         |

### iris get

Gets objects from the TitanAPI. You can look at all of the Iris get commands by using this command: `iris get --help`

<figure><img src="/files/JUxMY2vHyiBgMFXkb44a" alt=""><figcaption></figcaption></figure>

### iris status

Retrieves the status of an experiment. You can look at all of the Iris status commands by using this command: `iris status --help`

<figure><img src="/files/Kn2ku24qsyK5wQjA4MOO" alt=""><figcaption></figcaption></figure>

### iris upload

Uploads local models and datasets by local filepath. You can look at all of the Iris upload commands by using `iris upload --help`.

<figure><img src="/files/jP92kWcwZMwdIjCcXarl" alt=""><figcaption></figcaption></figure>

### iris delete

Deletes a model or dataset (i.e. an artefact), or an entire experiment, from the TitanML store.

<figure><img src="/files/5uhuYE1S83TsD37u8oeH" alt=""><figcaption></figcaption></figure>

### iris download

Downloads a finished, Titan-optimised model directly onto your machine in the ONNX format.

<figure><img src="/files/XHNfSl0zNqNQ81vFarKv" alt=""><figcaption></figcaption></figure>

### iris pull

Downloads a finished, Titan-optimised model as a Docker image (usage is equivalent to `docker pull`).

<figure><img src="/files/Bpw5oSd2x0WnxKxn5uFY" alt=""><figcaption></figcaption></figure>

### iris infer

Runs inference by requesting the Triton Inference Server.

<figure><img src="/files/FIrsxIuvZu51Dq7ZWJwo" alt=""><figcaption></figcaption></figure>

### iris makesafe

Converts a model to the safetensors format, including those models which the `transformers` save method can't convert.

<figure><img src="/files/nZ6KxVwq4sJJj9kj3WRA" alt=""><figcaption></figcaption></figure>