MachineLearning_Algorithms
MACHINE LEARNING ROADMAP
This repo is collection of all Machine Learning Algorithms Implemented using Just Python, Numpy and Math.
Machine Learning Workflow
Problem → Data → Preprocessing → Vectorization → Model → Evaluation → Deployment
| Step No. | Step Name | What Happens Here |
|---|---|---|
| 1 | Problem Definition | Understand the problem. What are you trying to predict/classify? |
| 2 | Data Collection | Gather raw data (CSV files, databases, APIs, images, etc.). |
| 3 | Data Preprocessing | Clean data (handle missing values, remove duplicates, fix errors). |
| 4 | Data Exploration (EDA) | Analyze data: plots, correlations, distributions (to understand patterns). |
| 5 | Feature Engineering | Create new features, select important ones, transform inputs. |
| 6 | Data Splitting | Split into Training, Validation, and Test sets (e.g., 70%-20%-10%). |
| 7 | Model Selection | Choose a model type (e.g., Decision Tree, SVM, Neural Network). |
| 8 | Training the Model | Fit (train) the model on the training data. |
| 9 | Model Evaluation | Check how good it is (using validation data and metrics like accuracy, RMSE, F1-score, etc.). |
| 10 | Hyperparameter Tuning | Optimize settings to improve performance (Grid Search, Random Search). |
| 11 | Testing | Final check on the unseen test data. |
| 12 | Deployment | Deploy the model into production (web app, API, etc.). |
| 13 | Monitoring and Maintenance | Watch the model over time and retrain if performance drops. |
References
-
Linear Regression -> https://developers.google.com/machine-learning/crash-course/linear-regression
-
Logistic Regression -> https://developers.google.com/machine-learning/crash-course/logistic-regression