Programmer Group

8. Sequence labeling Task08 text classification This study refers to Datawhale open source learning: https://github.com/datawhalechina/learn-nlp-with-transformers The content is generally derived from the original text and adjusted in combination with their own learning ideas. Personal summary: first, the structure of sequence annotation task ...

Cost function: quadratic, cross entropy, log likelihood (course: refining numbers into gold) quadratic cost function C = 1 2 n ...

In this article, we will explore how to build a simple network model with triple loss. It is widely used in face verification, face recognition and signature verification. Before entering the code, let's learn what triple loss is and how to implement it in PyTorch. Triple loss Triple loss function is a widely used loss function at present. It ...

brief introduction Super resolution refers to the process of enlarging or improving image details. When increasing the size of the image, additional pixels need to be interpolated in some way. Traditional image processing techniques can not get good results because they do not take the surrounding environment as the background when zooming in. ...

6, Text classification source Datewhle29 issue__ NLP transformer: Erenup (more notes), Peking University, principalZhang Fan, Datawhale, Tianjin University, Chapter 4Zhang Xian, Harbin Institute of technology, Chapter 2Li luoqiu, Zhejiang University, Chapter 3CAI Jie, Peking University, Chapter 4hlzhang, McGill University, Chapter 4T ...

Make a little progress every day!   What is a structure? A structure is a data set composed of the same or different data Before that, learn some little knowledge 2.1 only structure variables allocate addresses, and the definition of structure does not allocate space. 2.2 the definition of each member in the structure is the s ...

Modern neural network This week, I mainly learned several classic modern neural networks: AlexNet, VGG, NiN, geoglenet and ResNet. These networks can be regarded as a process of continuous refinement in the development of neural networks: Starting from the LeNet network, AlexNet is generated by changing the activation function and using DropOu ...

         After selecting the architecture and setting the super parameters, we enter the training stage. Our goal is to find the parameter value that minimizes the loss function. After training, we need to use these parameters to make future predictions. Sometimes we need to extract parameters to reuse th ...

1 Preface The four algorithms are similar, and have the following similarities and differences 2 similarities and differences Take STR1 = "ABCDEF", STR2 = "zabcdze" as an example Similarities: 1. Get a target on a string; 2. The core of the algorithm is the idea of dynamic programming. difference: 1. The targets are d ...

1. Get to know PyTorch 1.1 tensor 1. Import pytorch package import torch 2. Create an empty 5x3 tensor x = torch.empty(5, 3) print(x) 3. Create a randomly initialized 5x3 tensor x = torch.rand(5, 3) print(x) 4. Create a 5x3 0 tensor of type long x = torch.zeros(5, 3, dtype=torch.long) print(x) 5. Create tensors directly from the a ...