Chevron Left
返回到 Structuring Machine Learning Projects

学生对 deeplearning.ai 提供的 Structuring Machine Learning Projects 的评价和反馈

4.8
35,191 个评分
3,679 条评论

课程概述

You will learn how to build a successful machine learning project. If you aspire to be a technical leader in AI, and know how to set direction for your team's work, this course will show you how. Much of this content has never been taught elsewhere, and is drawn from my experience building and shipping many deep learning products. This course also has two "flight simulators" that let you practice decision-making as a machine learning project leader. This provides "industry experience" that you might otherwise get only after years of ML work experience. After 2 weeks, you will: - Understand how to diagnose errors in a machine learning system, and - Be able to prioritize the most promising directions for reducing error - Understand complex ML settings, such as mismatched training/test sets, and comparing to and/or surpassing human-level performance - Know how to apply end-to-end learning, transfer learning, and multi-task learning I've seen teams waste months or years through not understanding the principles taught in this course. I hope this two week course will save you months of time. This is a standalone course, and you can take this so long as you have basic machine learning knowledge. This is the third course in the Deep Learning Specialization....

热门审阅

AM

Nov 23, 2017

I learned so many things in this module. I learned that how to do error analysys and different kind of the learning techniques. Thanks Professor Andrew Ng to provide such a valuable and updated stuff.

DC

Mar 08, 2018

Going beyond the technical details, this part of the course goes into the high level view on how to direct your efforts in a ML project. Really enjoyable and useful. Thanks for making this available!

筛选依据:

151 - Structuring Machine Learning Projects 的 175 个评论(共 3,648 个)

创建者 Mark Z

Jun 11, 2019

I've decided to take this course after seeing its feedback from other people and the comment which got me was the following: "This course is could be summarized as a machine learning master giving useful advice". I think it perfectly describes the course's content. This course is definitely worth investing time into.

创建者 Dunitt M

Feb 10, 2019

Excelente curso, muy recomendado para quienes tienen una idea de Deep Learning pero con frecuencia se encuentran en situación que no saben cómo afrontar o cuál camino intentar primero. El conjunto de habilidades impartidas aquí no te harán un mejor programador, pero te ahorraran muchas horas de esfuerzo innecesario.

创建者 Gaurav K

Sep 07, 2017

Amazing tips shared for structuring machine learning projects, which were ignored in most of the other ML books. Building a model is one thing, but tuning it to make it work better in the real world is more important which this course focuses.

Thanks Prof. Andrew Ng for the consistent support of spreading knowledge!

创建者 Yuezhe L

Nov 20, 2018

This is a very helpful class. I have been working on machine learning projects for years. This course provides methods to systematically trouble shoot problems in a machine learning project. Despite all the samples are using neural networks, the methodology can be applied to improve other machine learning projects.

创建者 Bernard O

Oct 25, 2018

Excellent course on managing through the thick of bias/variance tradeoffs. Been doing a lot just based on things I have picked up through experience, but this course puts a the quantitative rigor and discipline behind the art. The sections on transfer and end to end deep learning were eye opening sections for me.

创建者 Gema P

Feb 25, 2018

This course is strategically very important so congrats on making it

I would add a programming assignment including transfer learning or multi-task learning implementation due to the multiple cases of use that are today in the industry.

Thanks again for making this Wonderfull material available to the community ^^

创建者 BAZIL F

Dec 29, 2019

Very useful course for understanding nuances of AI and different useful techniques in strategizing the approaches. Extremely useful in architecting, designing and delivery of the complex solutions involving AI (even as a sub-component). Prof. Andrew Ng is always a pleasure and honor to learn from. Thank You Sir!

创建者 Harvey Q

Sep 04, 2017

Really inspiring course, and UNIQUE. No other class, I think, provide these suggestions on the big question "what's next?" in ML projects. The videos are a bit weirdly sequenced. But they provide very systematic ways of project starting, data splitting, model evaluating, problem finding and tuning. Great course!

创建者 Pedro B M

Feb 28, 2019

This a course on key practices one should have when developing a ML project. Once again Andrew Ng is very pedagogical, teaching sometimes complex concepts in a easy to understand and practical way. I particularly liked the case studies, where the learned concepts had to be put into practice for decision taking.

创建者 Niyas M

Oct 29, 2017

What a great session! Full of practical advice and strategies to help you iterate fast. Prof. Andrew draws on his years of hands-on experience at top companies to put together the best practices for structuring your machine learning projects. This has been the most valuable course in this series for me so far!

创建者 Zebin C

May 18, 2019

In the course, I learned how to divide train set, dev set, and test set, and how to solve the problem of different distributions of train set and test set. Impressive is the transfer learning. Transfer learning is a very effective way to help me provide a completely different approach to solving new problems.

创建者 Abhilash V

Sep 11, 2017

This is a good course to get a feel of real projects and insights on how to go about executing them.I got some good tips to approach a deeplearning project.I don't know if this is too short of a course but I would trust Andrew Ng if he thinks this is fine to get a sense of deep learning projects.

Thank you.

创建者 Fahad S

Sep 06, 2018

The content is very unique and extremely insightful in how to structure a machine learning project. As a machine learning practitioner, I can personally vouch for the usefulness of the suggestions made by Andrew NG. Had I known all of this before, it would have saved me a lot of time on numerous projects.

创建者 Tushar M

Mar 17, 2018

This is the best ML course I have taken so far. A lot of ideas around train/dev/test sets, bias variance trade-off and difference of data distributions between train and dev sets snapped into place for me. I am sure it will take me a while to internalize this content but I feel like I have found the path.

创建者 Edward D

Oct 12, 2017

Brings a lot of useful insight of how to tune the model more from the data point instead of the model or algorithms. This could be super helpful in solving real world problems. Also the two case study homework helped me a lot to get a better understanding of what Andrew meant in his lecture. Great course.

创建者 Ayush K

Jan 19, 2020

Amazing course where Andrew NG shares his advice on how to work with datasets of different distributions etc. Coming from such an experienced practitioner is so helpful.

The Quizes are really helpful as they deal with case study and really make you feel like you're in the spotlight

Loved this course!!!

创建者 Zoheb A

Feb 05, 2019

The two quizzes of this course were unique. Never came upon such a quiz in any other online course. Along with the videos and supplementary pdfs, this course was quite unique and important in every aspect. I will use the approach I learnt here on my next ML projects. Thanks to Andrew Ng and the team.

创建者 João F

May 25, 2019

Very good course. Professor Ng explains very well why some strategies are better than others and how a deep learning practitioner or team can save a huge amount of working hours by following the instructions taught in this course. There are also useful, in-depth discussions in the forum. Thank you!

创建者 Lien C

Apr 04, 2019

Great practical insights of how to start a ML project, how to improve/optimize the system, how to identify and troubleshoot common problems in deep learning. The course provides comprehensive high level guidelines for anyone who uses machine learning, even without having any programming experience!

创建者 Dariusz J

Jul 19, 2019

The course has practical content. When took in the Deep Learning Specialization I noticed that some parts of the material were already known from previous courses. Indeed, in previuos courses the repeated aspectes are presented from a different angle, but probably there is an area for limiting it.

创建者 Jialin Y

Apr 22, 2018

It's like understanding deep learning: a team leader's perspective. Andrew may be the first instructor to give this kind of course. Based on his experience in building practical and large scale machine learning system in Google and Baidu, the course content is highly inspiring and worth listening.

创建者 Ged R

Oct 03, 2017

As an Ops person by nature, i like to see methodology and structure along with systematic approaches to results - be they solutions or problem solving. This course adds to that area, by providing best practices and ideas, it forms the basis from which these challenges can be addressed. Very good.

创建者 Mihai L

Jan 28, 2018

This course had no programming assignments. Yet I found it amazing. It truly gives you insight into how to engineer your projects to account for real world conditions.

Liked the flight simulator analogy to this course. Accelerated learning is really the great benefit of following Andrew's advice.

创建者 Gabriel L

Aug 25, 2017

I've done a Master degree in IA and the things covered in this course have never been addressed by any of my professors. Now I've been working in a Machine Learning team for the past two years now, and I believed these lessons would have been of great value, and would have saved me a lot of time!

创建者 Julio E H E

Jun 16, 2019

This course is very helpful to learn best-practices and problem-solving strategies that can help improve our deep learning algorithms. While I think the ultimate way of learning is through practice, here you can at least get a list of things to try in the future as you work on these algorithms.