Artificial Intelligence & Machine Learning
12 week Course Curriculum
Here’s a 3month Artificial Intelligence & Machine Learning curriculum designed like your previous webdev/Blockchain plan, with weekly breakdowns, monthly tests, assignments, and projects aligned to current industry standards (Python, ML/DL, MLOps basics, and practical tooling).
Assumptions :
4–5 hours per day, 5 day weeks → 12 weeks ≈ 3 months.
Tools :
Python, Jupyter, scikitlearn, pandas, NumPy, PyTorch/TensorFlow (choice), Streamlit/Flask for deployment.
Outcome :
Covers current standards: Python + PyTorch/TensorFlow, scikitlearn, and basic deployment/streaming or web APIs
Overall Assessment Plan :
Weekly assignments: 2–3 Python notebooks (GitHub).
Monthly tests: 1hour MCQ + practical coding (Colab/Google Drive).
Monthly projects: Endtoend ML apps (peerreviewed, documented)
Month 1 – Foundations of ML & Classical Algorithms (Weeks 1–4)
Weeks & Topics | Daily Work (4-5 hrs) | Assignments | Milestones |
|---|---|---|---|
week 1 : Python for ML & Data basics | NumPy arrays, pandas DataFrames, Matplotlib/seaborn, basic statistics. | 1) Clean a CSV dataset; 2) Compute basic stats/visualizations. | Short quiz on Python + stats |
week 2 : Supervised learning intro | Regression vs classification, train/test split, metrics (MSE, accuracy, confusion matrix). | 1) Simple linear regression from scratch; 2) Scikitlearn regression on a dataset. | Benchmark sheet with multiple models |
week 3 : Classical ML algorithms | Linear/logistic regression, kNN, decision trees, random forests, SVM (concept). | 1) Classification on Iris/UCI datasets; 2) Feature importance analysis. | Model comparison table. |
week 4 : Model evaluation & preprocessing | Crossvalidation, hyperparameter tuning (GridSearch/RandomSearch), scaling, encoding, Monthly Test 1 (Python + basic ML). | 1) Pipeline for titanicstyle dataset; 2) Tune hyperparameters. | Project 1: “Predictive Dashboard” (Jupyter notebook + Streamlit/Flask app to upload data, train a model, and show predictions/metrics). |
Month 2 – Deep Learning & CV/NLP (Weeks 5–8)
Weeks & Topics | Daily Task (4-5 hrs) | Assignment | Milestone |
|---|---|---|---|
week 5 : Neural networks basics | Perceptron, feedforward networks, activation functions, loss, gradient descent, backpropagation (concept). | 1) Simple MLP from scratch (NumPy); 2) PyTorch/TensorFlow version. | Loss/accuracy curve plots |
week 6 : Deep learning frameworks | PyTorch or TensorFlow basics (tensors, datasets, DataLoader, training loop) | 1) MNIST/CIFAR10 classifier; 2) Custom loss function | Tested notebook on GitHub |
week 7 : Computer vision basics | CNNs, image augmentation, transfer learning (ResNet/VGG), object detection concepts. | 1) Image classifier on a custom dataset; 2) Finetune a pretrained model. | Classification report (precision, recall). |
week 8 : NLP fundamentals | Text preprocessing, embeddings (Word2Vec, GloVe), sequence models (RNN/LSTM), transformers concept, Monthly Test 2 (deep learning + CV/NLP). | 1) Sentiment classifier; 2) Named entity recog (simple) or questionanswering demo. | Project 2: “Vision + NLP App” (e.g., image captioning or texttoimage classifier) using a pretrained model + simple API/UI). |
Month 3 – Advanced ML, MLOps & Capstone (Weeks 9–12)
Weeks & Topics | Daily task (4-5 hrs) | Assignment | Milestones |
|---|---|---|---|
week 09 : Unsupervised & ensemble methods | Clustering (Kmeans, DBSCAN), PCA, ensemble methods (AdaBoost, XGBoost), autoencoders concept | 1) Customer segmentation; 2) Dimensionalityreduction visualizations. | Clustering evaluation report. |
week 10 : Time series & recommender systems | ARIMA/SARIMA, LSTM for time series; collaborative/contentbased filtering concepts | 1) Predict stock/close prices (concept); 2) Simple movie recommender. | Errorcomparison table. |
week 11 : MLOps basics | Model versioning (MLflow), logging, simple deployment (Docker optional), monitoring, Final Test 1 (advanced ML + MLOps). | 1) Package a model with pickle/ joblib; 2) Write a REST API (Flask/FastAPI). | Model registry exercise. |
week 12 : Capstone & deployment | Endtoend project from data collection to deployment, documentation, presentation, Final Test 2 (capstone review). | 1) Integrate reports, tests, docs; 2) Deploy model API or dashboard. | Project 3: Industrialstyle ML product (e.g., “Churn Prediction System” or “Medical Diagnostic Assistant”) with data pipeline, trained model, API/UI, and MLOps trace; peerreviewed + pitchstyle demo. |
