Rasin Tsukuba Blog

Crystal Nets as Graphs

Introduction to graph theory and its application to crystal nets

Origin: Crystal Nets as Graphs Crystal Nets as Graphs Molecular topology is a graph Atom (vertices) joined by bonds (edges). Crystals e.g. diamond have topology specified by an infinite periodi...

Posted by Rasin on August 11, 2020

3D and Deep Learning

三维数据深度学习三维数据 3D数据与2D数据最大的区别即在于数据的表现形式。众所周知，2D数据可以表示为一个二维矩阵，但3D数据通常有许多种表现形式：深度图、点云、网格、体积网格(volumetric grids)。 3D数据的表达形式通常由应用驱动，例如在计算机图形学中做渲染和建模通常我们选择网格化数据，而将空间进行三维划分我们一般使用体积网格等，而在3D场景理解时，我们一般使...

Posted by Rasin on August 7, 2020

Transfer Learning Tricks

迁移学习任务场景把预训练的CNN模型当做特征提取器得到已经在大数据集上训练好的模型，去掉最后一层全连接层，然后将剩下的全部网络结构当做一个特征提取器，原来的网络最后一层的输出就是你的特征，然后将该特征输入一个SVM分类器或者softmax分类器就可以快速实现你自己的分类任务。 finetune model 这也是最常用的，因为一般我们并不会简单的把模型像1一样当做...

Posted by Rasin on August 6, 2020

Persistent-Homology-based Machine Learning and its Application

A Survey

Persistent-Homology-based Machine Learning and its Applications – A Survey Abstract A suitable feature representation that can both preserve the data intrinsic information and reduce data complex...

Posted by Rasin on August 4, 2020

Nlp With Pytorch 3 Classifying Sentiment Of Restaurant Reviews

«««< HEAD layout: post title: NLP with PyTorch 3 Classifying Sentiment of Restaurant Reviews subtitle: date: 2020-07-23 author: Rasin header-img: img/nlp-31.jpg catalog: true tags: Deep Lear...

Posted by Rasin in Tsukuba on July 23, 2020

NLP with PyTorch 3 Fundational Components of Neural Network

Chapter 3. Foundational Components of Neural Networks Prologue 本章通过介绍构建神经网络的基本思想，如激活函数、损失函数、优化器和监督训练设置，为后面的章节奠定了基础。我们从感知器开始，这是一个将不同概念联系在一起的一个单元的神经网络。感知器本身是更复杂的神经网络的组成部分。 Perceptron: The Simplest...

Posted by Rasin on July 22, 2020

NLP with PyTorch 2 Quick Tour of Traditional NLP

Chapter 2.A Quick Tour of Traditional NLP Conceptions 自然语言处理(NLP)和计算语言学(CL)是人类语言计算研究的两个领域。NLP旨在开发解决涉及语言的实际问题的方法，如信息提取、自动语音识别、机器翻译、情绪分析、问答和总结。另一方面，CL使用计算方法来理解人类语言的特性。 Corpora, Tokens, and Types ...

Posted by Rasin on July 22, 2020

NLP with PyTorch 1 Basics

Introduction and basic conceptions

Natural Language Processing with PyTorch 目标发展对监督学习范式的清晰理解，理解术语，并发展一个概念框架来处理未来章节的学习任务学习如何为学习任务的输入编码理解什么是计算图掌握PyTorch的基本知识 The Supervised Learning Paradigm 简单的监督学习，是指将Target(被预测的内容)...

Posted by Rasin on July 21, 2020

Deep Learning in Bioinformatics

Introduction, Application, and Perspective in Big Data Era

Deep learning in bioinformatics: introduction, application, and perspective in big data era Abstract In this review, we provide both the exoteric introduction of deep learning, and concrete examp...

Posted by Rasin on July 10, 2020

Graph Embedding on Biomedical Networks

Paper reading and summary

Graph Embedding on Biomedical Networks: methods, applications and evaluations Abstract Graph Embedding learning that aims to automatically learn low-dimensional node representations. We select 1...

Posted by Rasin on July 9, 2020

Rasin in Tsukuba