Search topics…
Tutorials
100 Days of ML 100 Days of DL 100 Days of NLP MCP Trilogy FastAPI for ML PyTorch Deep Learning LangChain & GenAI LangGraph Agents Code Lab (LeetCode AI) Syllabus Path
Explore
All Tutorials Courses Free Resources 1-on-1 Mentoring Corporate Training AI for Beginners Generative AI Course Corporate AI Coaching
June 6 Offline Event →
CampusX · 100 Days of Deep Learning

Master Deep Learning
in 100 Days

Complete 100 Days of Deep Learning curriculum — Perceptrons, MLPs, CNNs, RNNs, LSTMs, Attention, and Transformers. Free forever.

A structured DL curriculum — from Perceptrons and MLPs through CNNs, RNNs, Attention mechanisms, and modern Transformer architectures.

100Days / Lessons
9Core Modules
150+Code Examples
∞Free Forever
Curriculum Overview

9 Modules to DL Mastery

Each module builds on the previous — click any module to dive into detailed notes with code, theory, and exercises.

Module 1
Days 1–10

DL Foundations & Perceptrons

Understand the biological inspiration behind neural nets and the mathematical model of a single neuron.

  • Biological vs. artificial neural models
  • Rosenblatt Perceptron — weights, bias, step activation
  • Perceptron learning rule and convergence
  • Non-linear XOR problem and its limits
  • Binary linear separation and decision boundaries
Start Module 1 →
Module 2
Days 11–22

MLPs & Backpropagation

Build multi-layer networks and learn how gradients flow backward through the network during training.

  • Multi-Layer Perceptrons — architecture and notation
  • Forward propagation — matrix equations
  • Activation functions — Sigmoid, Tanh, ReLU
  • Backpropagation — chain rule derivation
  • Gradient descent weight updates
Start Module 2 →
Module 3
Days 23–33

Gradients, Optimizers & Tuning

Diagnose and fix training instabilities, and find the right optimizer and hyperparameters for your model.

  • Batch vs. SGD vs. Mini-batch gradient descent
  • Vanishing and exploding gradients
  • Gradient clipping and weight initialization
  • Learning rate schedules and warmup
  • Keras Tuner hyperparameter search
Start Module 3 →
Module 4
Days 34–44

Regularization & Performance

Prevent overfitting and improve model generalization with modern regularization and normalization techniques.

  • L1 and L2 weight decay regularization
  • Dropout — random neuron deactivation
  • Batch Normalization — internal covariate shift
  • Data augmentation strategies
  • ReLU, LeakyReLU, ELU activation variants
Start Module 4 →
Module 5
Days 45–55

Deep Learning Optimizers

Master the full optimizer zoo — from basic momentum to Adam — and know when to use each.

  • Momentum and Nesterov Accelerated Gradient
  • AdaGrad — per-parameter learning rate decay
  • RMSProp — moving average of squared gradients
  • Adam — adaptive moment estimation
  • AdamW, Lion, and modern optimizer variants
Start Module 5 →
Module 6
Days 56–67

CNNs & Computer Vision

Apply convolutional neural networks to image classification, detection, and transfer learning tasks.

  • Convolution operation — filters, padding, stride
  • MaxPooling and spatial dimension reduction
  • LeNet-5, AlexNet, VGG architectures
  • Keras Functional API for complex models
  • Transfer learning — ImageNet pre-trained models
Start Module 6 →
Module 7
Days 68–78

RNNs, LSTMs & GRUs

Model sequential data — text, time series, speech — with recurrent architectures and gating mechanisms.

  • Vanilla RNN — sequential loops and hidden states
  • Backpropagation Through Time (BPTT)
  • LSTM — forget, input, and output gates
  • GRU — simplified gating mechanism
  • Bidirectional RNNs and stacked architectures
Start Module 7 →
Module 8
Days 79–88

Seq2Seq, Attention & LLMs

Build encoder-decoder systems with attention — the foundation of modern language models.

  • Encoder-decoder translation architecture
  • Information bottleneck problem
  • Bahdanau soft-attention mechanism
  • Attention scores and alignment visualization
  • Evolution from RNNs to Transformers to ChatGPT
Start Module 8 →
Module 9
Days 89–100

Transformer Architectures

Understand the architecture that powers GPT, BERT, and every modern LLM — from scratch.

  • Scaled dot-product self-attention
  • Query, Key, Value projection matrices
  • Multi-Head Attention — parallel attention heads
  • Positional encodings and residual Add & Norm
  • Encoder-only (BERT) vs. decoder-only (GPT) variants
Start Module 9 →
Quick Navigation

Jump to Any Topic

All 100 DL topics mapped below — click to navigate directly.

What is DL? Biological Neurons Perceptron Model Step Activation Perceptron Learning Rule XOR Problem Linear Separability Decision Boundaries DL vs ML Keras & TensorFlow Setup MLP Architecture Forward Propagation Matrix Notation Sigmoid Activation Tanh Activation ReLU Activation Loss Functions Backpropagation Chain Rule Gradient Descent MLP in Keras MLP Project Batch vs SGD Mini-batch Training Vanishing Gradients Exploding Gradients Gradient Clipping Weight Initialization Learning Rate Schedules Keras Tuner Training Curves Hyperparameter Tuning Gradient Flow Project Overfitting in DL L2 Regularization L1 Regularization Dropout Batch Normalization Layer Normalization Data Augmentation ELU & LeakyReLU Early Stopping Regularization Project Model Comparison Momentum SGD Nesterov Momentum AdaGrad RMSProp Adam Optimizer AdamW Optimizer Comparison Warmup Schedules Optimizer Project Convolution Operation Filters & Feature Maps Padding & Stride MaxPooling LeNet-5 AlexNet VGG Network ResNet Skip Connections Keras Functional API Transfer Learning Fine-tuning Strategies CNN Project — Image Classifier Sequential Data Vanilla RNN Hidden States BPTT LSTM Gates LSTM Cell State GRU Bidirectional RNN Stacked LSTMs Time Series Forecasting Text Generation RNN Seq2Seq Architecture Encoder-Decoder Bottleneck Problem Bahdanau Attention Attention Scores Alignment Visualization Neural Machine Translation LLM Evolution History From RNNs to ChatGPT Transformer Overview Self-Attention Query Key Value Scaled Dot-Product Multi-Head Attention Positional Encoding Add & Norm Layers Feed-Forward Sublayer BERT Architecture GPT Architecture Fine-tuning BERT Hugging Face Transformers Transformer from Scratch Capstone Project Final Review 🎓

Ready to Start Your Deep Learning Journey?

Begin with Module 1 — no prior DL knowledge required. Solid ML fundamentals will help.

Start Day 1 → 100 Days of NLP All Generative AI Tutorials →

Frequently Asked Questions

Is this deep learning course free?

Yes. GenAIWallah's 100 Days of Deep Learning course is completely free — no signup, no paywall. It covers Perceptrons, MLPs, CNNs, RNNs, LSTMs, Attention, and Transformers in Hindi and English.

Do I need to know machine learning before deep learning?

It is recommended but not strictly required. A solid understanding of ML fundamentals (linear regression, gradient descent, loss functions) will help you learn deep learning faster. GenAIWallah recommends completing the 100 Days of Machine Learning course first before starting the Deep Learning track.

What deep learning topics are covered?

GenAIWallah's 100 Days of Deep Learning covers: Perceptrons and MLPs, backpropagation and gradient descent, regularization techniques, optimizers (Adam, RMSprop), CNNs for image recognition, RNNs and LSTMs for sequences, Attention mechanisms, and the full Transformer architecture from scratch. All explained in Hindi and English.

Best deep learning course in Hindi?

GenAIWallah's 100 Days of Deep Learning is India's best free deep learning course in Hindi and English. Created by Harsh Dhariwal (IIT Kanpur), it covers all major deep learning architectures from scratch — CNNs, RNNs, LSTMs, Attention, and Transformers — with practical code examples.