Hands-on Deep Learning V2

Hands-on deep learning

• Chapter 1：Preview (3 minutes and 13 seconds)

• Chapter 2：Course Schedule (8 minutes and 8 seconds)

• Chapter 3：Introduction to Deep Learning (13 minutes and 34 seconds)

• Chapter 4：Install (16 minutes and 6 seconds)

• Chapter 5：Data manipulation + data preprocessing (4 minutes and 41 seconds)

• Chapter 6：Linear Algebra (9 minutes and 3 seconds)

• Chapter 7：Matrix calculations (12 minutes and 38 seconds)

• Chapter 8：Automatic differentiation (11 minutes and 40 seconds)

• Chapter 9：Linear regression + basic optimization algorithm (14 minutes and 57 seconds)

• Chapter 10：Softmax regression + loss function + image classification dataset (10 minutes and 26 seconds)

• Chapter 11：Multilayer Perceptron + Code Implementation (13 minutes and 52 seconds)

• Chapter 12：Model Selection + Overfitting and Underfitting (18 minutes and 36 seconds)

• Chapter 13：Weight decay (13 minutes and 2 seconds)

• Chapter 14：Discard method (11 minutes and 59 seconds)

• Chapter 15：Numerical stability + model initialization and activation functions (15 minutes and 20 seconds)

• Chapter 16：Practice: Kaggle House Price Prediction + Course Competition: California House Price Prediction for 2020 (14 minutes and 30 seconds)

• Chapter 17：PyTorch Neural Network Basics (13 minutes and 27 seconds)

• Chapter 18：Using and Purchasing GPUs (13 minutes and 34 seconds)

• Chapter 19：Summary of the House Price Prediction Competition (16 minutes and 38 seconds)

• Chapter 20：Convolutional Layer (14 minutes and 59 seconds)

• Chapter 21：Padding and stride in convolutional layers (14 minutes and 7 seconds)

• Chapter 22：The technology behind AutoGluon (3 minutes and 47 seconds)

• Chapter 23：Multiple-input and multiple-output channels in convolutional layers (19 minutes and 33 seconds)

• Chapter 24：Pooling Layer (10 minutes and 20 seconds)

• Chapter 25：Classic convolutional neural network LeNet (8 minutes and 30 seconds)

• Chapter 26：AlexNet, a deep convolutional neural network (35 minutes and 15 seconds)

• Chapter 27：Network VGG using blocks (9 minutes and 3 seconds)

• Chapter 28：Network in Network (10 minutes and 54 seconds)

• Chapter 29：GoogLeNet - Inception V3 network with parallel connections (26 minutes and 58 seconds)

• Chapter 30：Batch Normalization (17 minutes and 58 seconds)

• Chapter 31：Residual Network ResNet (14 minutes and 9 seconds)

• Chapter 32：Why can ResNet train a 1000-layer model? (8 minutes and 21 seconds)

• Chapter 33：Part II Contest: Image Classification (6 minutes and 36 seconds)

• Chapter 34：Deep Learning Hardware: CPU and GPU (39 minutes and 22 seconds)

• Chapter 35：Deep Learning Hardware: TPUs and Beyond (29 minutes and 18 seconds)

• Chapter 36：Single machine with multiple cards in parallel (6 minutes and 58 seconds)

• Chapter 37：Multi-GPU training implementation (21 minutes and 40 seconds)

• Chapter 38：Distributed training (21 minutes and 44 seconds)

• Chapter 39：Data augmentation (12 minutes and 48 seconds)

• Chapter 40：Fine-tuning (13 minutes and 52 seconds)

• Chapter 41：The results of the second competition leaf classification (5 minutes and 30 seconds)

• Chapter 42：Practical Kaggle Competition: Image Classification (CIFAR-10) (27 minutes and 56 seconds)

• Chapter 43：Practical Kaggle Competition: Dog Breed Identification (ImageNet Dogs) (19 minutes and 8 seconds)

• Chapter 44：Object Detection and Datasets (11 minutes and 35 seconds)

• Chapter 45：Anchor Box (18 minutes and 6 seconds)

• Chapter 46：Technical Summary of Leaf Classification Competition (26 minutes and 28 seconds)

• Chapter 47：Object detection algorithms: R-CNN, SSD, YOLO (42 minutes and 4 seconds)

• Chapter 48：SSD Implementation (9 minutes and 56 seconds)

• Chapter 49：Semantic Segmentation and Datasets (7 minutes and 11 seconds)

• Chapter 50：Transposed Convolution (9 minutes and 20 seconds)

• Chapter 51：Transposed convolution is a type of convolution (9 minutes and 38 seconds)

• Chapter 52：Fully connected convolutional neural network FCN (4 minutes and 22 seconds)

• Chapter 53：Style Transfer (4 minutes and 57 seconds)

• Chapter 54：Course Competition: Denim Outfit Testing (11 minutes and 3 seconds)

• Chapter 55：Sequence Model (20 minutes and 3 seconds)

• Chapter 56：Text preprocessing (18 minutes and 3 seconds)

• Chapter 57：Language Model (16 minutes and 17 seconds)

• Chapter 58：Recurrent Neural Network (RNN) (25 minutes and 15 seconds)

• Chapter 59：Implementation of Recurrent Neural Network (RNN) (48 minutes and 34 seconds)

• Chapter 60：Gated Recurrent Unit (GRU) (15 minutes and 22 seconds)

• Chapter 61：Long Short-Term Memory (LTSM) (9 minutes and 38 seconds)

• Chapter 62：Deep Recurrent Neural Networks (3 minutes and 22 seconds)

• Chapter 63：Bidirectional recurrent neural network (7 minutes and 24 seconds)

• Chapter 64：Machine Translation Datasets (10 minutes and 15 seconds)

• Chapter 65：Encoder-Decoder Architecture (3 minutes and 32 seconds)

• Chapter 66：Sequence to Sequence Learning (seq2seq) (12 minutes and 34 seconds)

• Chapter 67：Beam Search (11 minutes and 43 seconds)

• Chapter 68：Attention Mechanism (13 minutes and 57 seconds)

• Chapter 69：Attention score (10 minutes and 44 seconds)

• Chapter 70：Seq2seq using attention mechanism (10 minutes and 14 seconds)

• Chapter 71：Self-Attention (27 minutes and 51 seconds)

• Chapter 72：Transformer (21 minutes and 3 seconds)

• Chapter 73：BERT Pre-training (21 minutes and 50 seconds)

• Chapter 74：BERT fine-tuning (9 minutes and 44 seconds)

• Chapter 75：Summary of Object Detection Competition (23 minutes and 57 seconds)

• Chapter 76：Optimization Algorithm (34 minutes and 3 seconds)

• Chapter 77：Course summary and further learning (22 minutes and 36 seconds)