> For the complete documentation index, see [llms.txt](https://webcanvas.gitbook.io/webcanvas-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://webcanvas.gitbook.io/webcanvas-docs/3.-evaluation.md).

# 3. Evaluation

1. ### Create a New Channel

<figure><img src="/files/oCJAeEdnTKkR87DhJyd3" alt=""><figcaption><p>Create a New Channel for your Proposed Challenge</p></figcaption></figure>

Since the default channel is designed to be a draft channel and it's private, you should create a new channel on imean builder, and then move all the tasks you want to include in the challenge to that channel.

<figure><img src="/files/xLZ5qFzyd5Khl9Zl4wfI" alt=""><figcaption><p>Move the Tasks to the Created Channel</p></figcaption></figure>

2. ### Create a Challenge bonded with that Channel

Then you can create a challenge on WebCanvas platform and connect it with the channel in step 1.

<figure><img src="/files/asufDNPnjBoiYXnmQF9A" alt=""><figcaption><p>Create a Challenge and Connect it with the Channel</p></figcaption></figure>

3. ### Export the Data and Run the Evaluation
   1. Check out our GitHub repository for the complete user story on running online evaluation after you set up a challenge. <https://github.com/iMeanAI/WebCanvas>
