This is the beginning of my journey of learning machine learning. I found this tutorial playlist by NeuralNine on YouTube, and decided it was worth following it through.
The files attached here are what was made as part of this, and I've tried to annotate where I could to explain what I understood from the tutorial.
These are the different files linked to the tutorials:
| Tutorial | File |
|---|---|
| #2 | linear_regression.py |
| #3 | k_nearest_neighbors_classification.py |
| #4 | support_vector_machines.py |
| #5 | decision_trees_and_random_forest.py |
| #6 | k_means_clustering.py |
| #8 | handwritten_digit_recognition.py |
I learnt multiple machine learning algorithms which are in the files above:
- Linear Regression
- K-Nearest Neighbor Classification
- Support Vector Machines Classification
- Decision Trees Classification
- Random Forest Classification
- K-Means Clustering
I can't explain their implementations, but I understand how they work and how to use them
The final part of this tutorial was making a more useful application, which was the handwritten digit recognition. There are 10 PNG files attached which were my own handwritten digits, and I used these to test if the model worked. On one occasion, the model correctly identified all of the digits.
Although it was a more surface level introduction, I have understood some things such as the algorithms but more importantly when I tried to jump in before, I got stuck with things such as feed forward, neurons, and loss functions. Now that I understand these, I can look deeper into machine learning and eventually make something more advanced.
Thanks for reading!
by Humzah Awais