In this 90-min long project-based course you will learn how to use Tensorflow to construct neural network models. Specifically, we will design, execute, and evaluate a neural network model to help a retail company with their marketing campaign by classifying images of clothing items into 10 different categories. Throughout this course, you will learn how to use Tensorflow to build and analyze neural neural networks that can perform multi-label classification for applications in image recognition. You will also be able to identify and adapt the main components of neural networks as well as evaluate the performance of different models and implement measures to improve their accuracy. At the end of the project, you will be able to design and implement convolutional neural networks helping a retail store with their targeted ad campaign, and the models can be easily adapted for self-driving cars, computer-assisted medical diagnosis, etc.
CNNs with TensorFlow: Basics of Machine Learning
CNNs with TensorFlow: Basics of Machine Learning
访问权限由 New York State Department of Labor 提供
您将学到什么
Adapt the main components of neural networks: inputs, layers, weights, and activation functions according to the specific application.
Use TensorFlow and Keras to design, implement, and adapt convolutional neural networks for image recognition tasks.
Evaluate neural network models and measure their accuracy, modify the parameters of the model if needed to improve its accuracy.
您将练习的技能
要了解的详细信息
添加到您的领英档案
仅桌面可用
了解顶级公司的员工如何掌握热门技能
在 2 小时内学习、练习并应用岗位必备技能
- 接受行业专家的培训
- 获得解决实训工作任务的实践经验
- 使用最新的工具和技术来建立信心
关于此指导项目
分步进行学习
在与您的工作区一起在分屏中播放的视频中，您的授课教师将指导您完成每个步骤：
-
Understand the main components of neural networks in machine learning
-
Train your first neural network for image classification
-
Improve neural network accuracy through hidden layers and different optimizers
-
Practice Activity: Fine tune a neural network and improve its accuracy
-
Visualize training data and performance of the model
-
Create a convolutional neural network with Conv2D and MaxPooling2D
-
Reduce overfitting with BatchNormalization, Dropout, and L2 regularization
-
Practice Activity: Create alternative neural network models to reduce overfitting
-
CIFAR-10 Classification Challenge
推荐体验
Basic familiarity with Python. In particular, importing libraries, defining variables, arrays, functions, and classes, and creating plots.
9个项目图片
位教师
提供方
学习方式
基于技能的实践学习
通过完成与工作相关的任务来练习新技能。
专家指导
使用独特的并排界面，按照预先录制的专家视频操作。
无需下载或安装
在预配置的云工作空间中访问所需的工具和资源。
仅在台式计算机上可用
此指导项目专为具有可靠互联网连接的笔记本电脑或台式计算机而设计，而不是移动设备。
人们为什么选择 Coursera 来帮助自己实现职业发展
Felipe M.
Jennifer J.
Larry W.
