|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/pytorch-deep-learning-hero/
课程评论：没有评论
**课程名称：** PyTorch深度学习 **课程概述：** 本课程是“PyTorch深度学习从零到英雄”系列的一部分，专为已掌握Python基础语法并寻求提升编程技能至商业应用水平的学习者设计。课程将带领您使用PyTorch和Hugging Face，通过Google Colab进行动手实践，从简单的感知器到BERT迁移学习和Google T5模型，全面掌握深度神经网络的实现。每个章节都包含一个挑战性的编码任务，以巩固所学知识。 **课程内容亮点：** * **基础概念：** 介绍Google Colaboratory，讲解神经元、感知器、数据归一化、激活函数、损失函数以及梯度下降等核心概念，并演示如何优雅地利用PyTorch实现梯度下降。 * **多层感知机（MLP）：** 学习MLP的原理，掌握One-Hot编码、数据准备、模型定义、训练与评估，并通过“LOVE Letters”分类任务实践MLP。 * **卷积神经网络（CNN）：** 深入理解CNN的工作原理，学习如何准备Fashion MNIST数据集，定义CNN模型，进行训练、评估和推理。 * **循环神经网络（RNN）：** 学习RNN的原理，掌握数据准备、模型定义、训练和推理。 * **Transformer模型：** * **BERT：** 学习BERT模型，通过“Sesame Street”和IMDB数据集应用BERT进行数据准备、模型定义、训练、评估和预测，并完成BERT最终项目。 * **T5：** 学习T5模型，同样通过应用T5进行数据准备、模型定义、训练、评估和预测，并完成T5最终项目。 **学习目标：** 掌握深度学习的基础理论和实践技巧，能够独立使用PyTorch和Hugging Face实现和应用各种深度学习模型，包括但不限于MLP、CNN、RNN以及先进的Transformer模型。 **学习方法：** 理论与实践相结合，通过Google Colab提供的实际操作环境，完成每个章节的编码任务，实现从零到精通。 **课程格言：** 秉持Pythonic的编程哲学，追求代码的优雅、清晰和可读性，鼓励学习者通过实践解决问题，不断精进。
Pytorch & Hugginface Deep Learning Course(Colab Hands-On) Welcome to Pytorch Deep Learning From Zero To Hero Series.If you have already mastered the basic syntax of python and don't know what to do next, this course will be a rocket booster to skyrocket your programming skill to a business applicable level.In this course, you will be able to master implementing deep neural network from the very beginning(simple perceptron) to BERT transfer learning/Google's T5 by using pytorch and huggingface yourself by colab. Each Section will have one assignment for you to think and code yourself. The Agenda is below. Agenda:IntroductionGoogle ColaboratoryNeuronPerceptronMake Your Perceptron TrainableNormalize DataActivation FunctionLoss FunctionGradient DescentElegant Pytorch Gradient DescentFinal ProjectFinal Project ExplainedMulti Layer Perceptron(MLP)One Hot EncodingPrepare data for MLPDefine MLPTrain & Evaluate MLPFinal Project for MLPFCNN ExplainedFCNN LOVE Letters Classification using MLPFinal Project For FCNNCNN ExplainedCNN Prepare data(Fashion MNIST) CNN Define Model CNN Train & Evaluate ModelCNNInferenceFinal Project For CNNRNN ExplainedRNN Prepare dataRNN Define ModelRNN Train ModelRNN InferenceBERT Sesame StreetBERT Prepare Data IMDBBERT Model definitionBERT Model TrainingBERT Model EvaluationBERT Model PredictionBERT Final ProjectT5 Prepare Data T5 Model definitionT5 Model TrainingT5 Model EvaluationT5 Model PredictionT5 Final ProjectLet's start our journey together.Beautiful is better than ugly.Explicit is better than implicit.Simple is better than complex.Complex is better than complicated.Flat is better than nested.Sparse is better than dense.Readability counts.Special cases aren't special enough to break the rules.Although practicality beats purity.Errors should never pass silently.Unless explicitly silenced.In the face of ambiguity, refuse the temptation to guess.There should be one- and preferably only one -obvious way to do it.Although that way may not be obvious at first unless you're Dutch.Now is better than never.Although never is often better than *right* now.If the implementation is hard to explain, it's a bad idea.If the implementation is easy to explain, it may be a good idea.Namespaces are one honking great idea - let's do more of those!
