Santiago Profile picture
Jun 2 10 tweets 5 min read Twitter logo Read on Twitter
AI will revolutionize the world in the next 3 to 5 years.

But we need more builders: people willing to work and learn solid AI/ML skills.

Here is how you can start:
Most people think that starting is hard. They are wrong.

These recommendations will get you through the first month.

1. Learn to use notebooks
2. Learn to deal with data
3. Learn data visualization
4. Learn basic algorithms
5. Build your first project

Here are a few resources:
But first, let me thank the sponsor for today's post:

Prompts Daily.

AI will not replace you. A person using AI will.

That's why almost 100k people read this newsletter to keep up with the latest AI insights, news, and tools.

Sign up here for free → neatprompts.com/subscribe Image
For this guide to be helpful, you need to know Python.

If you are comfortable writing Python, keep going. If you aren't, I'd suggest you start there.
1. Learn to use notebooks

You want to learn about notebooks: Jupyter or Google Colab are your friends.

Notebooks are a fantastic way to code, experiment, and communicate your results.

Here is a 30-minute tutorial on Jupyter Notebooks:
Image
2. Learn to deal with data

Pandas is a one-stop shop for this.

"10 minutes to pandas" is an excellent tutorial to get you started on the basics: pandas.pydata.org/pandas-docs/st….

Also, watch this video: . Image
3. Learn data visualization

It's critical to learn data visualization and how to showcase the work you are doing.

Kaggle's tutorial is a fantastic start: kaggle.com/learn/data-vis…

Also, watch this 6-minute tutorial on Seaborn: . Image
4. Learn basic algorithms

A few suggestions: Decision Trees, KNN, Linear Regression, and Neural Networks.

Kick it off with the Machine Learning Recipes from Google: youtube.com/results?search… Image
Before talking about your first project, it's time to go through an end-to-end tutorial that will put everything together for you.

Look at the "Intro to Machine Learning" tutorial.

It's a quick tutorial that will bring together all of the pieces:
kaggle.com/learn/intro-to… Image
5. Build your first project

The tutorial from the previous step ends with the Titanic exercise. You can find it on Kaggle: kaggle.com/c/titanic/data.

This is everything you need to get started and finish your first Machine Learning project! Image

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Santiago

Santiago Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @svpino

Jun 1
90% of AI applications won't make it.

Using Large Language Models is hard. Most developers rely on prompting techniques to improve the results of their models.

Sometimes you need more than that.

Here is how you can 10x the quality of language models: Image
Fine-tuning is key for complex tasks: You teach a model to solve specific problems by training it on your data.

But there are a few problems with fine-tuning:

• It's expensive
• Requires complex setup and management
• Hard to scale
• Requires a lot of experimentation
The team @monsterapis just released the ability to fine-tune these models:

• Llama 7B, 13B
• OPT 125m, 6B
• GPT J 6B
• Stable LM 3B, 7B
• GPT 2 XL

The first decentralized GPU platform to offer a no-code fine-tuning solution: monsterapi.ai.
Read 5 tweets
May 31
ChatGPT will not make you money.

You need actual skills that pay the bills and set you up for the next 10 years.

To become a builder, join the Machine Learning School for the June cohort.

Here is what you’ll get:
You will learn how to build a Machine Learning Production system from scratch.

It's a 9-hour live cohort. It's challenging.

You'll start with a problem and build a system like the attached image.

Cohort 3 starts Jun 19th.

Register here: svpino.gumroad.com/l/mlp. Image
And that's just the start!

When you join, you get immediate access to the following:

• Course: How to make money in Upwork.
• Course: Neural Networks From Scratch (Fall 2023)
• 350+ engineers willing to help
• Recording of every past session

But there's even more!
Read 6 tweets
May 30
There are ~130+ different Machine Learning algorithms.

This is overwhelming for anyone starting, yet experts do this almost effortlessly.

Here is their secret:
Sit with any professional and show them 10 problems.

In 10 seconds, they will tell you which algorithm to use, and their answer will be correct 99% of the time.

They do this effortlessly, just like you drive your car.

Practice turned into knowledge that became muscle memory.
You were taught at school that a good Machine Learning model is the one that performs the best.

Forget that immediately.

Model performance is necessary but not the only thing that matters.
Read 16 tweets
May 29
I thought art was impossible to automate.

I was wrong.

Today, anyone can turn their photo gallery into unlimited, amazing pictures in less than 10 minutes.

Here are a few lines of Python code to show you how: Image
Follow these 3 steps:

1. Open the notebook linked below
2. Go to tryleap.ai and get an API KEY
3. Run the code.

That's it. Here is the code:

colab.research.google.com/drive/18-hR4ur…
The code is dead simple:

• It uses the Realistic Vision model from @leap_api.
• Waits for the process to finish.
• Generates photos following a prompt.

Run this, and you'll have as many photos as your imagination will let you.
Read 5 tweets
May 26
7 ways I'm using ChatGPT to write code.

These won't make you 10x better. These won't change your life. If you ignore these, you'll do just fine.

With ChatGPT, you don't need better prompts. You need new ideas that will help you use the tool better.
In my experience, using ChatGPT for writing code is not about better prompts.

I don't know any secret prompts that will produce better results.

My recommendation:

Talk to the tool as you would talk to a person. Be clear and specific.
I deal with a ton of code, most of which I didn't write.

I stopped trying to "debug" code in my mind, and instead, I started using ChatGPT for that.

I ask, "Explain this snippet of code," and include the lines I want to understand.

ChatGPT always gives me a ton of details. Image
Read 11 tweets
May 25
Most "AI applications" will not survive 2023.

Most are just thin layers on top of OpenAI's API. Most were built on a weekend. Most are honestly not valuable.

But here is one of the exceptions that prove the rule.

3 years in the making!
@dazzeloid reached out and showed me their new product powered by AI, and now I wish every website had this.

You can "talk" to your site and find what you are looking for 10x easier!

I'm sure biased because I work with @CommandBar, but this is awesome!
Behind the scenes, you can configure the tool with no code:

• It will index your site
• You can connect to a CMS
• You can feed it information manually

The result is a personalized AI agent that changes based on the page the user is visiting.

Pretty cool stuff!
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(