How I Learned Machine Learning from DeepSeek

Hello! In this article I'm talking about my journey in the AI domain, from being someone who didn't know the difference between AI and machine learning, to becoming an AI expert in my university. I even guided people who had more years of experience in AI more than me, but through self-learning, I was able to surpass them.

So the story started when I was sitting in the cafe shop, scrolling through LinkedIn, And I saw a video of person using computer vision in his work. he created a model that could recognize sign language gestures with his hands and convert them into words. At that time, I didn't know what this technology was, but I just knew it was AI, and I couldn't let it pass like that. So I opened the  new LLM in that period DeepSeek and I wrote with my imperfect English:

"i want to lern ai coding. can you give me first simple code with python and i well indrestend the code and level up step by step"

 

I didn't want to betray my own principles, because since I started in IT, my social circle of teachers and friends have always seen me as pro in the problem solving , the self-learning and coding. These perspectives from close people push me to always give my best.

well, I followed DeepSeek step by step from understanding the libraries and the sub fields of AI  like ML, DL, NLP and CV to understanding the code, and finally writing my own code. Totally in this conversation  I talked with DeepSeek for about 5 or 6 days, let's say a week. That was enough for me to understand the supervise models and create one of California Housing Price Prediction. 

I was thrilled by this model. Imagine, when my classmates were having problems with basics of the matplotlib in the python course, I did my own model from scratch with full consciousness. Fully aware of what I'm doing, from downloading the data from kaggle to opening it with pandas to training the model and updating the hyperparameters .

I think I enough talked about the story. Now let's talk about the project technically. 

👉 You can find the full project here: https://github.com/Ad2m1109/Machine-Lerning

This GitHub repository contains many files, but the most important for me are:

    -StepByStep.ipynb: This was my “friend” notebook. I used it to save my progress step by step. Because this I named it like that. So I wrote each step individually, by reading it line by line and looking for each piece of information, I was able to grasp it.

    -Summary.md: I also wrote this file to summarize each step from the notebook.

    -gui_app.py: This file presents my final model, gradient_boosting_model.pkl.

In it I used Tkinter for the graphical interface and matplotlib for data visualization.

Well, I'm writing this article to share my experience, because I'm proud of myself. Also to increase my knowledge, so let's talk together about one of AI's sub-fields Machine learning(ML).

This is what I learned from that experience. So what did I do ?

Machine Learning(ML) is a field of artificial intelligence that allows computers to learn from data, identify patterns, and make decisions or predictions without being explicitly programmed.

This training built for algorithm such as :

-Supervised Learning Algorithms: the model learns from labeled human data

-Unsupervised Learning Algorithms: the model learns using its experience and find patters by itself 

I invite you to search about them or maybe I will write article about each field of AI and example of each time of algorithm.

This is a simple diagram I hope it can explain what did I say:   

   

 

 

By the way, thank you to Google for Nano Banana, the best image generator I've used. 

I want also to thank each one read this article. I'm Adem Youssfi, a big passionate about the AI world, always searching to level up my knowledge and share it with others.