> ## Documentation Index
> Fetch the complete documentation index at: https://docs.gpuhub.com/llms.txt
> Use this file to discover all available pages before exploring further.
# TensorBoard
> TensorFlow is a software library for machine learning and artificial intelligence.
The instance on GPUhub comes pre-installed with the TensorBoard tool. You just need to save the TensorBoard event files to the /root/tf-logs/ directory (if you wish to use a different logs folder, see the section below on changing directories), or copy the event files saved in another path to this directory, and then use TensorBoard.
## Default Access Method
After powering on the instance, locate the access entry for AutoPanel:
## Switch TensorFlow Logs Directory
If you wish to use a different logs directory, you can manually start another instance of TensorBoard:
First, terminate the default TensorBoard process by executing the command:
```bash bash theme={null}
ps -ef | grep tensorboard | awk '{print $2}' | xargs kill -9
```
Then, run the following command in the terminal to start TensorBoard:
```bash bash theme={null}
tensorboard --port 6007 --logdir /path/to/your/tf-logs/directory
```
After starting, you can access TensorBoard again through AutoPanel.