What are the main ideas and approaches of machine learning you have applied when creating a model from scratch?

When creating a machine learning model from scratch, the following main ideas and approaches are usually applied:

• Data collection and preparation: cleaning, normalization, splitting into training and testing sets.
• Model selection: determining the type of model (linear regression, decision trees, neural networks, etc.) depending on the task.
• Model training: optimizing model parameters on training data using algorithms (gradient descent, least squares method, etc.).
• Quality assessment: using metrics (accuracy, recall, F1, MSE) on test data to check generalization ability.
• Hyperparameter tuning: selecting model parameters to improve results.

Example in Python using sklearn:

from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error

# Assume X and y are data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
model = LinearRegression()
model.fit(X_train, y_train)
predictions = model.predict(X_test)
mse = mean_squared_error(y_test, predictions)
print(f"MSE: {mse}")