# Setting up ML terminal

Again, very much like working with RNA-Seq, we need a Linux terminal which runs Python.  However, to train a ML model would need at least more than 10x of computing power or memory capacity than RNA-Seq.  That said, nothing is stopping one to set one up at home, but cloud computing service is more feasible - if you are not working on a personal project and you can justify your usage, then your company or clients will have to pay for it, on top of your skills.  But if you are working on a personal project, chances are you are going to use the rig heavily which justify the setup cost.  And if you are not going to use the machine so heavily, one always has the option to go cloud.  So to conclude if you are setting up your own, the setup will not deviate from [this](/rna-seq/4.-setting-up-terminal.md) that much, and if you have the resources to set up your own ML terminal, I would imagine I am only rambling for no good.

{% hint style="info" %}
To train an acoustic recognition model using only 66 samples with 66 vaildation track, a [2vCPU + 8GiB](https://instances.vantage.sh/aws/ec2/m5.xlarge) instance without GPU takes 3 hours.  [A GPU instance](https://instances.vantage.sh/aws/ec2/g4dn.xlarge) makes it in 4 minutes.
{% endhint %}

That said, this difference reminds me of my GTX950 sleeping in my garage, and my RNA-Seq terminal setting under my table - AWS is fast but it takes time to launch instance, it hangs and it is good for collaboration, meaning that lone wolf like me - oh man again I will go learn how to set up my own [machine learning rig](https://medium.com/@ashkan.abbasi/quick-guide-for-installing-python-tensorflow-and-pycharm-on-windows-ed99ddd9598). 

<figure><img src="/files/6UwREvcgnNcTSPq8WHPB" alt=""><figcaption><p>Shift + Enter</p></figcaption></figure>

<figure><img src="/files/7AMcPY38HRRb6lx50OB9" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/7iukwSfSgZFb5PVYm8nJ" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://nks.gitbook.io/rna-seq/9.-aws-cloud-and-machine-learning/setting-up-ml-terminal.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.