免费现场课程：使用PyTorch进行深度学习

Are you interested in learning about Deep Learning? We are hosting a free 6-week live course on our YouTube channel, starting Saturday, May 23rd at 8:30 AM PST.

您对学习深度学习感兴趣吗？ 我们将于5月23日(星期六)太平洋标准时间上午8:30开始在我们的YouTube频道上举办为期6周的免费实时课程 。

Passively watching a video is often not enough to learn a software concept. You need to be able to ask questions and build real projects. That is exactly what you will be able to do in the course “Deep Learning with PyTorch: Zero to GANs”.

被动地观看视频通常不足以学习软件概念。 您需要能够提出问题并建立真实的项目。 这就是您在“使用PyTorch进行深度学习：从GAN归零”课程中可以做到的。

This is an online course intended to provide a coding-first introduction to deep learning using the PyTorch framework. The course takes a hands-on coding-focused approach and will be taught using live interactive Jupyter notebooks, allowing students to follow along and experiment.

这是一门在线课程，旨在为使用PyTorch框架的深度学习提供编码优先的入门知识。 该课程采用动手编程为重点的方法，并将使用实时交互式Jupyter笔记本电脑进行授课，使学生可以继续学习并进行实验。

This course is taught by Aakash N S. He is the co-founder and CEO of Jovian.ml, a project management and collaboration platform for machine learning.

该课程由Aakash NS教授。 他是Jovian.ml(机器学习的项目管理和协作平台)的联合创始人兼首席执行官。

Theoretical concepts will be explained in simple terms using code. Students will receive weekly assignments, work on a project with real-world datasets and participate in a private data science competition to test their skills. Upon successful completion of the course, students will receive a certificate of completion.

理论概念将使用代码以简单的术语进行解释。 学生将接受每周的作业，使用真实数据集进行项目研究，并参加私人数据科学竞赛以测试其技能。 成功完成课程后，学生将获得结业证书。

This is a beginner-friendly course, and no prior knowledge of data science, machine learning or deep learning is assumed. It is preferable to have some background in the following areas:

这是一门适合初学者的课程，不假定您具备数据科学，机器学习或深度学习的先验知识。 最好具有以下方面的背景：

• Programming knowledge, preferably in Python
  编程知识，最好是Python
• Basics of linear algebra (vectors, matrices, dot products)
  线性代数的基础(向量，矩阵，点积)
• Basics of calculus (differentiation, geometric interpretation of derivative)
  微积分的基础(微分，导数的几何解释)

教学大纲 (Syllabus)

The course is divided into 6 modules, and will be taught over 6 weeks via video lectures and interactive Jupyter notebooks. Each lecture will be around 2 hours long.

该课程分为6个模块，将通过视频讲座和交互式Jupyter笔记本电脑进行为期6周的教学。 每个讲座将持续2个小时左右。

单元1：PyTorch基础知识-张量和渐变 (Module 1: PyTorch Basics - Tensors & Gradients)

• Introduction to Jupyter notebooks & Data Science in Python
  Jupyter笔记本和Python数据科学简介
• Creating vectors, matrices & Tensors in PyTorch
  在PyTorch中创建向量，矩阵和张量
• Tensor operations and gradient computations
  张量运算和梯度计算
• Interoperability of PyTorch with Numpy
  PyTorch与Numpy的互操作性

单元2：线性回归和梯度下降 (Module 2: Linear Regression & Gradient Descent)

• Linear Regression from scratch using Tensor operations
  使用Tensor操作从头开始进行线性回归
• Weights, biases and the mean squared error loss function
  权重，偏差和均方误差损失函数
• Gradient descent and model training with PyTorch Autograd
  使用PyTorch Autograd进行梯度下降和模型训练
• Linear Regression using PyTorch built-ins (nn.Linear, nn.functional etc.)
  使用PyTorch内置的线性回归(nn.Linear，nn.functional等)

单元3：用于图像分类的逻辑回归 (Module 3: Logistic Regression for Image Classification)

• Working with images from the MNIST dataset
  使用MNIST数据集中的图像
• Training and validation dataset creation
  训练和验证数据集的创建
• Softmax function and categorical cross entropy loss
  Softmax函数和分类交叉熵损失
• Model training, evaluation and sample predictions
  模型训练，评估和样本预测

单元4：前馈神经网络和GPU (Module 4: Feedforward Neural Networks & GPUs)

• Working with cloud GPU platforms like Kaggle & Colab
  与Kaggle＆Colab等云GPU平台一起使用
• Creating a multilayer neural network using nn.Module
  使用nn.Module创建多层神经网络
• Activation function, non-linearity and universal approximation theorem
  激活函数，非线性和通用逼近定理
• Moving with datasets and models to the GPU for faster training
  将数据集和模型移至GPU以进行更快的训练

单元5a：使用卷积神经网络进行图像分类 (Module 5a: Image Classification using Convolutional Neural Networks)

• Working with the 3-channel RGB images from the CIFAR10 dataset
  处理来自CIFAR10数据集的3通道RGB图像
• Introduction to Convolutions, kernels & features maps
  卷积，内核和特征图简介
• Underfitting, overfitting and techniques to improve model performance
  拟合不足，拟合过度和技术以提高模型性能

单元5b：数据扩充，正则化和残留网络 (Module 5b: Data Augmentation, Regularization and Residual Networks)

• Improving the dataset using data normalization and data augmentation
  使用数据规范化和数据扩充来改善数据集
• Improving the model using residual connections and batch normalization
  使用残差连接和批处理规范化改进模型
• Improving the training loop using learning rate annealing, weight decay and gradient clip
  使用学习速率退火，权重衰减和梯度夹改善训练循环
• Training a state of the art image classifier from scratch in 10 minutes
  从零开始训练最先进的图像分类器

第6单元：使用生成对抗网络(GAN)生成图像 (Module 6: Image Generation using Generative Adversarial Networks (GANs))

• Introduction to generative modeling and application of GANs
  GAN的生成建模和应用简介
• Creating generator and discriminator neural networks
  创建生成器和鉴别器神经网络
• Generating and evaluating fake images of handwritten digits
  生成和评估手写数字的伪造图像
• Training the generator and discriminator in tandem and visualizing results
  串联训练生成器和判别器并可视化结果

练习与作业 (Exercises & Assignments)

每周作业 (Weekly Assignments)

• Week 1: Linear Regression
  第一周：线性回归
• Week 2: Image Classification
  第2周：图片分类
• Week 3: Feedforward neural networks
  第三周：前馈神经网络

课程项目 (Course Project )

For the course project, students will create an image classification model using Convolutional neural networks, on a real-world dataset of their choice. The project will allow students to experiment with different types of models and regularization techniques. Students will also present their work at the end of the course and publish a blog post describing their approach and results.

对于本课程项目，学生将使用卷积神经网络在他们选择的真实数据集上创建图像分类模型。 该项目将允许学生尝试不同类型的模型和正则化技术。 学生还将在课程结束时介绍他们的工作，并发布一篇博客文章，介绍他们的方法和结果。

Kaggle课堂比赛 (Kaggle In-Class Competition)

Students will participate in a private data science competition hosted on the Kaggle platform. The competition will run for 3 weeks, allowing students to apply & improve their skills in a competitive environment. Students will gain exposure to working with cloud GPU platforms.

学生将参加在Kaggle平台上举办的私人数据科学竞赛。 比赛将持续3周，让学生在竞争激烈的环境中申请并提高自己的技能。 学生将接触到使用云GPU平台的知识。

结业证书 (Certificate of Completion)

Students who attend at least 5 out of 6 video lectures and make valid submissions for all assignments will be eligible to receive a Certificate of Completion by Jovian.ml. Selected projects will also be receive a Best Project Award based on evaluation criteria determined by the instructors.

参加至少6个视频讲座中的5个并为所有作业提交有效作品的学生将有资格获得Jovian.ml的结业证书。 选定的项目还将根据讲师确定的评估标准获得“最佳项目奖”。

注册 (Sign up)

You can sign up for the course here: https://bit.ly/pytorchcourse.

您可以在这里注册该课程： https : //bit.ly/pytorchcourse 。

Whether or not you sign up, you can watch the course on the freeCodeCamp.org YouTube channel.

无论您是否注册，都可以在freeCodeCamp.org YouTube频道上观看该课程。

翻译自: https://www.freecodecamp.org/news/free-deep-learning-with-pytorch-live-course/