Job Interview Bootcamp এবং ICPC Preparation Bootcamp Batch 9 এ ভর্তি চলছে
৮ মাসের কমপ্লিট প্রিপারেশন
Full Stack
AI/ML ENGINEERING BOOTCAMP
এই বুটক্যাম্পের ৩টা ফোকাস এরিয়া।
ক্যারিয়ারে একজন AI Engineer/ML Engineer/Data Engineer/Data Analyst হতে পারবেন।
একাডেমিক লাইফে Thesis/Research এ টপ-নচ পারফর্ম করতে পারবেন।
Research/Project এর মাধ্যমে Studey Abroad এর গাইডেন্স পাবেন।
Module - 01 : Python Programming
- Data Science overview
- AI vs ML vs Deep Learning
- Supervised vs Unsupervised Learning
- Regression, Classification, Clustering
- Real-world use cases
- Career paths for ML/AI domain (ML+Analytics vs. Academic vs. ML+SWE)
- Required tools
- Required platforms
- Additional materials on Linear Algebra and Statistics
- Setting up IDE (VS Code, PyCharm, Jupyter)
- Python syntax and basic operations
- Variables and data types
- Input/output operations
- Basic arithmetic and string operations
- Conditional statements
& Loops
- Control statements
- Functions & Arguments
- List
- Tuple
- Set
- Dictionary
- When and why to use each structure
- List comprehensions
Assignment - 01 : Python Problem Solving
Class - 04 : Advanced Python Basics & File Handling
- Lambda functions
- Map, Filter, Reduce functions
- List, Dictionary, Set comprehensions
- File handling (open, read, write, append)
- Working with CSV files
- JSON file operations
- Error handling with try-except
- String formatting (f-strings, format())
- Working with modules (import, from)
- Command line arguments
- Regular expressions basics (re module)
Assignment - 02 : Advanced Python Practice
Module - 02 : Data Manipulation & Analysis
Class - 05 : NumPy Fundamentals
- NumPy arrays vs Python lists
- Array creation and operations
- Indexing and slicing
- Mathematical operations
- Broadcasting
- Random number generation
- Array reshaping
Class - 06 : Pandas for Data Manipulation
- Series and DataFrame
- Reading/writing data (CSV, Excel, JSON)
- Data selection and filtering
- Handling missing values
- Data aggregation and groupby
- Merging and joining datasets
- Basic time series operations
Assignment - 03 : Data Manipulation with Pandas
Class - 07 : Data Cleaning & Preprocessing
- Understanding different data types
- Identifying and handling missing values (mean, median, mode, forward/backward fill)
- Outlier detection (IQR, Z-score)
- Outlier treatment (capping, removal)
- Data type conversion
- Duplicate removal
- String operations for text cleaning
Class - 08 : Exploratory Data Analysis (EDA) & Visualization
- EDA methodology
- Statistical summaries (describe, info)
- Distribution analysis
- Correlation analysis
- Matplotlib basics (line, bar, scatter plots)
- Seaborn for statistical plots
- Plotly for interactive visualizations
Assignment - 04 : Comprehensive EDA & Visualization
Module - 03 : Feature Engineering & ML Preparation
Class - 09 : Feature Engineering & Encoding
- Feature scaling (StandardScaler, MinMaxScaler, RobustScaler)
- Normalization techniques
- Encoding categorical variables (Label, One-Hot, Ordinal)
- Feature creation and extraction
- Binning and discretization
- Handling date-time features
Class - 10 : Train-Test Split & Data Leakage Prevention
- Importance of data splitting
- Train-test-validation split
- Stratified sampling
- Cross-validation (K-Fold, Stratified K-Fold)
- Data leakage concepts and prevention
- Handling imbalanced datasets (SMOTE, SMOTETomek, class weights)
- Pipeline creation for preprocessing
Assignment - 05 : Feature Engineering & Data Preparation
Module - 04 : Introduction to Machine Learning
Class - 11 : ML Fundamentals & First Model
- AI vs ML vs Deep Learning
- Supervised vs Unsupervised Learning
- Regression vs Classification
- ML workflow overview
- Introduction to scikit-learn
- Building first Linear Regression model
- Model evaluation basics
Class - 12 : Model Evaluation & Metrics
- Regression metrics (MAE, MSE, RMSE, R²)
- Classification metrics (Accuracy, Precision, Recall, F1)
- Confusion matrix
- ROC curve and AUC
- Overfitting vs Underfitting
- Bias-Variance tradeoff
- Learning curves
Assignment - 06 : First ML Model & Evaluation
Module - 05 : Regression Algorithms
Class - 13 : Linear & Polynomial Regression
- Simple Linear Regression theory
- Multiple Linear Regression
- Loss functions (MSE, MAE)
- Polynomial Regression
- When to use polynomial features
- Feature importance in regression
Project - 01 : House Price Prediction System
Class - 14 : Regularization Techniques
- Ridge Regression (L2 regularization)
- Lasso Regression (L1 regularization)
- ElasticNet (L1 + L2)
- When to use each technique
- Feature selection with Lasso
- Hyperparameter tuning for alpha
Class - 15 : KNN & Distance-Based Learning
- K-Nearest Neighbors algorithm
- Distance metrics (Euclidean, Manhattan)
- Choosing optimal K
- KNN for regression and classification
- Advantages and limitations
- Computational complexity
Assignment - 07 : Regression Models Comparison
Module - 06 : Classification Algorithms
Class - 16 : Logistic Regression
- Binary classification
- Sigmoid function
- Decision boundary
- Multiclass classification (One-vs-Rest, Softmax)
- Probability predictions
- Classification threshold tuning
Project - 02 : Disease Prediction System
Class - 17 : Support Vector Machines (SVM)
- Margin and hyperplane concepts
- Support vectors
- Kernel trick (Linear, RBF, Polynomial)
- C parameter (regularization)
- Gamma parameter
- SVM for binary and multiclass
Class - 18 : Naive Bayes & Text Classification
- Probability fundamentals
- Bayes theorem
- Naive Bayes assumptions
- Types (Gaussian, Multinomial, Bernoulli)
- Text preprocessing (lowercase, remove punctuation, stopwords)
- Text vectorization (CountVectorizer, TfidfVectorizer)
Project - 03 : Email Spam Classifier
Class - 19 : Decision Trees
- Decision tree structure
- Information gain and Gini impurity
- Tree building process
- Pruning techniques
- Feature importance
- Visualization of decision trees
- Hyperparameters (max_depth, min_samples_split)
Assignment - 08 : Classification Models Comparison
Module - 07 : Ensemble Learning
Class - 20 : Ensemble Methods - Bagging & Boosting
- Ensemble learning concept
- Bagging (Bootstrap Aggregating)
- Random Forest algorithm
- Boosting concept
- Gradient Boosting basics
- XGBoost, CatBoost, LightGBM
- Feature importance in ensemble methods
- When to use which ensemble method
Project - 04 : Customer Churn Prediction
Class - 21 : Hyperparameter Tuning & Model Selection
- Grid Search
- Random Search
- Bayesian Optimization concept
- Automated tuning with Optuna
- Cross-validation with tuning
- Model selection strategies
- Best practices for hyperparameter tuning
Assignment - 09 : Ensemble Methods & Hyperparameter Tuning
Module - 08 : Unsupervised Learning & Time Series
Class - 22 : Clustering Algorithms
- Unsupervised learning overview
- K-Means clustering algorithm
- Elbow method for optimal K
- Silhouette score
- Hierarchical clustering
- DBSCAN introduction
- Cluster evaluation
Project - 05 : Customer Segmentation System
Class - 23 : Time Series Forecasting
- Time series basics
- Components (trend, seasonality, residual)
- Stationary vs non-stationary data
- ARIMA model
- SARIMA for seasonal data
- Prophet by Facebook
- Evaluation metrics for time series
Project - 06 : Stock Market Price Forecasting
Module - 09 : Advanced NLP & Projects
Class - 24 : Advanced NLP & Sentiment Analysis
- Advanced text preprocessing
- N-grams
- Word embeddings concept
- Sentiment analysis techniques
- Building an end-to-end NLP pipeline
- Model deployment for text classification
Project - 07 : Fake News Detection System
Class - 25 : End-to-End ML Project Workflow
- Complete ML project lifecycle
- Problem definition and scoping
- Data collection strategies
- Comprehensive EDA
- Model selection criteria
- Evaluation and comparison
- Documentation best practices
- Presentation skills
Project - 08 : Amazon Product Review Sentiment Analysis
Module - 10 : Model Deployment
Class - 26 : Web Apps with Streamlit & Gradio
- Introduction to Streamlit
- Building interactive dashboards
- Model integration with Streamlit
- Gradio for quick ML interfaces
- User input handling
- File upload functionality
- Deployment on Streamlit Cloud/Hugging Face Spaces
- Best practices for UI/UX
Project - 09 : Multi-Model ML Dashboard
Class - 27 : FastAPI for ML Deployment
- What is FastAPI and why for ML
- REST API basics (GET, POST)
- Creating ML inference API
- Request/response schema with Pydantic
- Model serialization (pickle, joblib)
- Loading and using saved models
- Testing with Swagger UI
- Error handling and validation
- CORS configuration
Assignment - 10 : ML Model API Development
Module - 11 : Deep Learning with PyTorch
Class - 28 : Neural Networks Fundamentals
- What is Deep Learning?
- Neuron and Perceptron model
- Artificial Neural Networks (ANN) architecture
- Forward propagation
- Activation functions (Sigmoid, Tanh, ReLU, Leaky ReLU, Softmax)
- Loss functions (MSE, Binary Cross-Entropy, Categorical Cross-Entropy)
- Backpropagation concept (intuition)
- Gradient Descent and learning rate
- Optimizers (SGD, Momentum, Adam, RMSprop)
- Batch, Mini-batch, and Stochastic gradient descent
- Overfitting in neural networks
- Regularization techniques (Dropout, L1/L2)
Hands-on:
- Build perceptron from scratch using NumPy
- Visualize activation functions
- Understand gradient descent with simple examples
Class - 29 : PyTorch, Keras Basics & Building Neural Networks
- Introduction to PyTorch and Keras
- Why PyTorch for Deep Learning?
- Tensors and tensor operations
- PyTorch vs NumPy
- Autograd and automatic differentiation
- Building neural networks with nn.Module
- Defining layers (Linear, Activation)
- Dataset and DataLoader
- Training loop structure
- Evaluation loop
- Saving and loading models
- GPU acceleration basics
Project - 10 : Image Classification with ANN
Class - 30 : Convolutional Neural Networks (CNN)
- Limitations of ANN for images
- Introduction to CNN
- Convolution operation
- Filters and feature maps
- Pooling layers (Max pooling, Average pooling)
- CNN architecture (Conv → Pool → FC)
- Famous architectures overview (LeNet, AlexNet, VGG)
- Transfer Learning concept
- Data augmentation techniques
- Building a CNN with PyTorch
Project - 11 : Advanced Image Classification with CNN
Assignment - 11 : CNN Implementation & Comparison
Class - 31 : Recurrent Neural Networks (RNN) & Sequence Models
- Sequential data and its challenges
- Introduction to RNN
- RNN architecture and working
- Vanishing gradient problem
- Long Short-Term Memory (LSTM)
- Gated Recurrent Unit (GRU)
- Bidirectional RNN
- Sequence-to-sequence models
- Applications: Text generation, time series, translation
- Building RNN/LSTM with PyTorch
Project - 12 : Text Generation & Sentiment Analysis with LSTM
Option A - Text Generation
Option B - Multilingual Sentiment Analysis
Assignment - 12 : Deep Learning Projects & Documentation
Class - 32 : Transformers & Attention Mechanisms
- Limitations of RNNs and motivation for Transformers
- Self-attention mechanism
i) Query, Key, Value concepts
ii) Scaled dot-product attention
iii) Multi-head attention
- Transformer architecture
i) Encoder-decoder structure
ii) Positional encoding
iii) Feed-forward networks
iv) Layer normalization and residual connections
- Pre-trained Transformer models
i) BERT (Bidirectional Encoder Representations from Transformers)
ii) GPT (Generative Pre-trained Transformer)
iii) T5 (Text-to-Text Transfer Transformer)
- Transfer learning and fine-tuning
- Hugging Face Transformers library
- Applications: Text classification, NER, question answering, summarization
Vision Transformers (ViT) - brief introduction
Project - 13 : Transformer-based NLP Application
Option A - Text Classification with BERT
Option B - Question Answering System
Assignment - 13 : Transformer Experimentation
Module - 12 : Thesis, Publication, Documentation & Career Guideline
Class - 33 : LaTeX, Thesis Writing & Career Guidelines - 1
- Introduction to LaTeX
- Why LaTeX for academic writing
- Setting up an Overleaf account
- Document structure (sections, subsections)
- Mathematical equations formatting
- Tables and figures
- Citations and bibliography (BibTeX)
- Cross-referencing
- Research paper template
- Thesis formatting guidelines
- Conference/Journal Publication Guideline
Hands-on:
- Create a research paper template
- Write mathematical equations
- Add tables and figures
- Manage references
- Export to PDF
Class - 34 : LaTeX, Thesis Writing & Career Guidelines - 2
- ML/AI career paths:
i) ML + Analytics: Data Scientist, ML Analyst
ii) Academic Research: Research Scientist, PhD path
iii) ML + Software Engineering: ML Engineer, AI Engineer
- Building professional resume for ML roles
- GitHub portfolio optimization
- LinkedIn profile optimization
- Kaggle competitions strategy
- Interview preparation (technical + behavioral)
- Networking in ML community
- Continuous learning resources
- Freelancing opportunities
- Building personal brand
Assignment - 14 : Documentation & Career Preparation
FINAL CAPSTONE PROJECT
এই বুটক্যাম্পে যারা পড়াবেন
