|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/deep-convolutional-generative-adversarial-networks-dcgan/
课程评论：没有评论
**课程名称：** 深度卷积生成对抗网络 (DCGAN) **课程概述：** 本课程将深入探讨生成对抗网络 (GANs) 和深度卷积生成对抗网络 (DCGANs)，这些是当前计算机科学领域最热门和最引人入胜的技术之一。您将学习到如何通过对抗性过程同时训练两个模型：一个生成器 (Generator)，它会学习生成逼真的图像；以及一个判别器 (Discriminator)，它会学习区分真实图像和生成图像。 **学习目标：** 完成本课程后，您将掌握 Python 编程基础，并深入理解生成对抗网络 (GANs) 和深度卷积生成对抗网络 (DCGANs) 的核心概念。 **课程内容与实践：** 课程将提供详细的步骤指导，涵盖以下关键环节： * **导入 TensorFlow 及其他库** * **加载和准备数据集** * **构建模型（生成器和判别器）** * **定义损失函数和优化器（生成器损失、判别器损失）** * **定义训练循环** * **训练模型** * **分析模型输出** **建议先修知识：** * **Python 编程：** 课程中会提供必要的复习。 * **梯度下降** * **神经网络基础知识**
Generative Adversarial Networks (GANs) & Deep Convolutional Generative Adversarial Networks (DCGAN) are one of the most interesting and trending ideas in computer science today. Two models are trained simultaneously by an adversarial process. A generator , learns to create images that look real, while a discriminator learns to tell real images apart from fakes.At the end of the Course you will understand the basics of Python Programming and the basics ofGenerative Adversarial Networks (GANs) & Deep Convolutional Generative Adversarial Networks (DCGAN).The course will have step by step guidance Import TensorFlow and other librariesLoad and prepare the datasetCreate the models (Generator & Discriminator)Define the loss and optimizers (Generator loss , Discriminator loss)Define the training loopTrain the modelAnalyze the output Suggested Prerequisites:Python coding: some revision is provided during this courseGradient descentBasic knowledge of neural networks
