|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/data-science-in-action-using-python/
课程评论：没有评论
课程名称：使用 Python 的数据科学实战 课程概述： 随着非结构化数据的爆炸性增长，我们在设计、开发和部署人工智能模型方面有了广阔的机会。尽管有很多课程教授数据科学，然而你需要一个逐步指导，来选择问题、探索数据、开发和部署模型，并通过用户反馈和学习来改进模型。本课程涵盖了许多大数据挑战，并修改了 CRISP-DM 方法论以应对大数据。本课程为 AI 模型的开发和部署提供了一种修改后的方法论，我们的修改方案已在多个真实大规模项目中得到验证。 课程中，我们将选择一个真实案例研究，提供在所选案例上进行数据科学项目设计和原型制作的实践经验，使你能够将结果运用到日常生活中。我们将数据科学家分为两类：点击者和编码者。点击者通过用户界面使用数据科学工具进行高层次的规范，而编码者则使用编程语言和库来编写数据科学相关的代码，Python 是最受欢迎的语言之一。本课程的目标是为你提供数据科学的入门编码体验。 课程开始时有两个关键活动： 1. 环境设置 - 逐步指导你准备沙箱环境以运行所有 Python 代码。 2. 数据科学方法论 - 回顾我们数据科学方法论中的关键步骤、任务和活动。 随后，课程将介绍我们七步数据科学方法论，并通过 Python 解释每一步，结合真实案例进行示范。这七个步骤包括： 1. 描述用例：阐释选择的数据科学用例。 2. 描述数据：描述数据来源，并使用 Python 解释数据集。 3. 准备数据集：使用 Python 准备数据集。 4. 开发模型：进行许多 AI 建模技术的实践，包括时间序列分析、分类、聚类、回归和预测，均使用 Python。 5. 评估模型：提供对 AI 模型结果进行评估的度量标准。 6. 部署模型：提供 AI 模型部署的流程。 7. 监控模型：提供对生产环境中的模型进行持续监控和评估的流程。 在本课程中，你将有机会设计一个用例，并使用 Python 作为主要语言进行实施。你需要下载所有数据集和示例 Python 代码，完成每个部分的作业，并按照提供的指导提交最终的笔记本。
With explosive growth of data in unstructured data, we have ample opportunities to design, develop and deploy AI models. While there are many courses which teach you Data Science, you need a step-by-step guide on how to select a problem, explore data, develop & deploy models, and improve the model using user feedback and learning. This course covers many big data challenges and modifies CRISP-DM to deal with big data. This course provides you a methodology for AI model development and deployment as modified by us to deal with AI and big data. Our modifications have been tried on many real-life large-scale projects. We will select a real case study for this data science project and will provide hands-on experience in Designing / prototyping a Data science engagement on the chosen case study. You will be able to use the results in your day-to-day life.We divide the data scientists into clickers and coders. Clickers are those data scientists who use a data science tool with a user interface to provide a high-level specification. Examples include SPSS Modeler, Excel and Alteryx. In each case you can add formula, but do not need to write code. The second set of data scientists are those who use a procedural language with libraries to write code for data science work. Python is the most popular language among data scientists. The objective of this course is to get you an introductory coding experience in data science. If you are interested in a clicker course, we offer a course using Dataiku. In addition, our data science methodology course is also designed for Business Analysts and Project Managers with limited development background.Course starts with two critical activities· Set up Environment - step by step instructions in preparing sandbox environment for executing all your python code· Data Science Methodology - to review key steps, tasks and activities associated with our data science methodologyAfter above section, This course introduces our 7 step data science methodology and use Python to explain each step using our real life use case example. These 7 steps include· Step 1: Describe Use Case to explain selected use case for data science work· Step 2: Describe Data to describe Data Sources and explain data sets using Python as a language.· Step 3: Prepare Datasets to Prepare Data Sets using Python· Step 4: Develop Model will provide hands-on exercises in applying many AI modeling techniques on data sets such as time series analysis, classification, clustering, regression, and forecasting. All these exercises will be using Python as a language.· Step 5: Evaluate Model will provide measurements to Evaluate your AI Model Results· Step 6: Deploy Model will provide process for deploying your AI models.· Step 7: Monitor model will provide process for continuous monitoring and evaluating your models in productionIn this course, we will give you an opportunity to design a use case and then work on its implementation using Python as your primary language. You should download all data sets and sample python code. Complete all assignment in each section of the course and submit your final notebook using instructions provided.
