docs/source/starting_ui.bck

# Starting the UI

The AutoTrain UI can be started in multiple ways depending on your needs.
We offer UI on Hugging Face Spaces, Colab and locally!

## Hugging Face Spaces

To start the UI on Hugging Face Spaces, you can simply click on the following link:

[![Deploy on Spaces](https://huggingface.co/datasets/huggingface/badges/resolve/main/deploy-on-spaces-md.svg)](https://huggingface.co/login?next=/spaces/autotrain-projects/autotrain-advanced?duplicate=true)

Please make sure you keep the space private and attach appropriate hardware to the space.
You can also read more about AutoTrain on the homepage and follow the link there to start your own training instance on
Hugging Face Spaces. [Click here](https://huggingface.co/autotrain) to visit the homepage.

## Colab

To start the UI on Colab, you can simply click on the following link:

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/huggingface/autotrain-advanced/blob/main/colabs/AutoTrain.ipynb)

Please note, to run the app on Colab, you will need an ngrok token. You can get one by signing up for free on [ngrok](https://ngrok.com/).
This is because Colab does not allow exposing ports to the internet directly.


## Locally

To run the autotrain app locally, install autotrain-advanced python package:

```bash
$ pip install autotrain-advanced
```

and then run the following command:

```bash
$ export HF_TOKEN=your_hugging_face_write_token
$ autotrain app --host 127.0.0.1 --port 8000
```

This will start the app on `http://127.0.0.1:8000`.

AutoTrain doesn't install pytorch, torchaudio, torchvision, or any other dependencies. You will need to install them separately.
It is thus recommended to use conda environment:


```bash
$ conda create -n autotrain python=3.10
$ conda activate autotrain

$ pip install autotrain-advanced

$ conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
$ conda install -c "nvidia/label/cuda-12.1.0" cuda-nvcc
$ conda install xformers -c xformers

$ python -m nltk.downloader punkt
$ pip install flash-attn --no-build-isolation
$ pip install deepspeed

$ export HF_TOKEN=your_hugging_face_write_token
$ autotrain app --host 127.0.0.1 --port 8000
```

In case of any issues, please report on the [GitHub issues](https://github.com/huggingface/autotrain-advanced/).