09. Modern CNN Architectures: Residual Connections (ResNet), Squeeze-and-Excitation & ConvNeXt • Deep Learning & PyTorch Neural Networks Architecture • Rishabh Ki Kaksha

Deep Learning & PyTorch Neural Networks Architecture › 09. Modern CNN Architectures: Residual Connections (ResNet), Squeeze-and-Excitation & ConvNeXt

50 mins

← Prev Next →

Course Syllabus

Module 1: PyTorch Tensors, Autograd & Computational Graphs

01. PyTorch Tensor Architecture: Strides, Storage Buffers, Memory Layout (Contiguous vs Non-Contiguous) 02. Dynamic Computational Graphs & The PyTorch Autograd Engine (grad_fn, requires_grad, backward) 03. GPU Acceleration with CUDA: Tensor Device Placement (to("cuda")), CUDA Streams & Pinned Memory

Module 2: Multi-Layer Perceptrons (MLPs) & Custom Training Loops

04. Building Custom nn.Module Classes, Custom Parameters & Activation Functions (ReLU, GELU, Swish) 05. Custom Datasets & High-Performance DataLoaders (num_workers, collate_fn, pin_memory) 06. The Enterprise Training Loop: Forward Pass, Loss Computation, Zero Grad, Backward Pass & Optimizer Step 07. Neural Regularization Techniques: Dropout, Batch Normalization (BatchNorm1d) & Layer Normalization (LayerNorm)

Module 3: Convolutional Neural Networks (CNNs) & Computer Vision Architecture

08. Convolutional Operations: Kernels, Filters, Strides, Padding & Feature Map Dilation 09. Modern CNN Architectures: Residual Connections (ResNet), Squeeze-and-Excitation & ConvNeXt 10. Transfer Learning & Vision Fine-Tuning with Torchvision Pretrained Models

Module 4: Recurrent Architectures: RNNs, LSTMs & GRUs for Sequential Data

11. Sequential Modeling: Vanilla RNNs & The Vanishing / Exploding Gradient Problem 12. Long Short-Term Memory (LSTM) Networks: Forget Gate, Input Gate, Candidate State & Output Gate Math 13. Gated Recurrent Units (GRUs) & Bidirectional Sequence Processing in PyTorch

Module 5: Modern Transformer Architecture from Scratch in PyTorch

14. Scaled Dot-Product Attention: Query, Key, Value Matrices & Attention Softmax Calculation 15. Multi-Head Attention (MHA) & Causal Masking Implementation from Scratch in Pure PyTorch 16. Building a Complete GPT-Style Decoder-Only Transformer Language Model from Scratch

Module 6: PyTorch Distributed (DDP), Mixed Precision (AMP) & Model Checkpointing

17. Automatic Mixed Precision (torch.cuda.amp.autocast & GradScaler) for 2x Faster GPU Training 18. Distributed Data Parallel (DDP) Architecture: Multi-GPU Synchronization with torch.distributed 19. Robust Checkpointing: Saving State Dicts, Resuming Training & Early Stopping Callbacks 20. Model Export & Production Serialization: TorchScript (JIT Tracing), ONNX & TensorRT Compilation

Interactive Lesson Studio

# 09. Modern CNN Architectures: Residual Connections (ResNet), Squeeze-and-Excitation & ConvNeXt

> Deep Learning & PyTorch Notes by Rishabh - Topic 09

---

### 1. The Skip Connection Breakthrough (ResNet Residual Block 🌉)

He et al. ne discover kiya ki 50+ layer deep networks train karte waqt accuracy degrade hoti hai:
$$\mathcal{H}(x) = \mathcal{F}(x) + x \implies \text{Gradient: } \frac{\partial \mathcal{H}}{\partial x} = \frac{\partial \mathcal{F}}{\partial x} + 1$$
*The constant $+1$ gradient superhighway vanishing gradient problem ko 100% eradicate kar deta hai!*

```python
class ResidualBlock(nn.Module):
    def __init__(self, channels: int):
        super().__init__()
        self.conv1 = nn.Conv2d(channels, channels, 3, padding=1, bias=False)
        self.bn1 = nn.BatchNorm2d(channels)
        self.conv2 = nn.Conv2d(channels, channels, 3, padding=1, bias=False)
        self.bn2 = nn.BatchNorm2d(channels)
        self.relu = nn.ReLU(inplace=True)

def forward(self, x: torch.Tensor) -> torch.Tensor:
        residual = x
        out = self.relu(self.bn1(self.conv1(x)))
        out = self.bn2(self.conv2(out))
        out += residual # ⚡ Skip connection superhighway!
        return self.relu(out)
```

---

### 2. Master Comparison Matrix

---

### 3. Interview Ke 3 Sawal Is Topic Se

1. **Degradation Problem over-fitting se alag kaise hoti hai?** -> Degradation me deep model ka Training Error bhi shallow model se zyada bura hota hai (not overfitting, but optimization difficulty).
2. **Bottleneck Residual Block (ResNet-50/101/152) me $1\times 1 \rightarrow 3\times 3 \rightarrow 1\times 1$ structure kyu use hota hai?** -> $3\times 3$ convolution se pehle dimensions compress karna aur baad me expand karna computational FLOPs reduce karta hai.
3. **Global Average Pooling (GAP) Fully-Connected layers ko kyu replace karta hai?** -> Spatial $(H \times W)$ dimension ka average lekar 1D vector banata hai, 80% parameters reduce karta hai aur spatial overfitting rokkta hai.

---
**Next Topic:** 10. Transfer Learning & Vision Fine-Tuning with Torchvision Pretrained Models

Loading lesson content...

Python 3.12

Full IDE

Live Execution Terminal

Click "Execute Live" to run code.
                

Lesson Mastery AI Quiz

AI Doubt Solver

CERTIFICATE OF COMPLETION

Rishabh Gehlot

PHP 8+ & Laravel 11 Master Track