學生對 deeplearning.ai 提供的 Sequence Models 的評價和反饋

I like the course. It's beneficial and clear. Also, the concept is clear.

for more improvement

I would suggest that for jupyter implementation :

I hope you put 2 versions of the code

thus, the student can have a choice to work on a famous frame

1- using Tensorflow (TF)

2- using PyTorch

This course was really awsome,learning has been fun in all the 4 courses, the number of new things learnt in this course was remarkable.Even the mot complicated things were taught in such a way that it never seemed tough.Doing assignments really helped to make concepts even more clear.

Fantastic course! Presents both the theory and practical uses in a straightforward manner that is easy to grasp. Programming assignments are a mix of NumPy and Keras API, with the former being more illustrative of the inner workings of RNNs and the latter being more practically useful.

Excellent! Amazing! Such good quality of lecture and assignments. Thank you Andrew and team for giving me such a good overview of what i can use this for. I feel as though this series dramatically lowered the barriers to entry for me to get started on any ML project i decide to. Thanks

No better introductory material. I suggest doing NLP specialization by deeplearning.ai after this though I have still not tried it out myself yet but hoping to do it some time. Thank you Andrew! I got emotional in your last video of the course. You are such an example for educators <3

This was by far the Best Course and Specialization that I have done. Thank You Coursera and Thank You Sir Andrew NG . You have made me confident and able in the Field of Deep Learning. I am grateful to you Sir. I will try my best to use this knowledge as a superpower in the right way.

In my opinion, this module was a little more complex than the previous ones, however, it has been an excellent course to deepen my knowledge in recurrent neural networks and close this specialization program in deep learning. Thank you very much, Dr. Andrew Ng and deeplearningai!!!!

Great course with interesting exercises. One can really see the amount of effort being put into creating the assignment material, so they are at a suitable level and with a lot to take away, and the students have a good understanding of details through practice. Really appreciated.

the programing assignment ins WEEK 1 was a bit ambiguous in nature. helped me improve my debugging skills.

Also a huge thank you to MENTOR Mr. GEOFF for the instant support to all my queries. His way of providing HINTS lead me to finally complete the course. a BIG thank you to him.

More easier to understand than the ConvNets course!

Week 1 and 2 took me a little time to get through. Week 3 is easy.

For better understanding, don't forget to download the notebooks and practice on your own local Jupyter notebook while using the assignment notebook as a reference.

This is a great course. It provides a good introduction to RNNs and how they are used in sequence modeling. Introduction to GRU, LSTMs, BiRNNs, attention model is great way to learn these in depth. The exercises are designed to make you familiar with the internals of these layers.

At the beginning this is very counter-intuitive. But later on when I am on the final assignment, I finally realized that we are not focusing on gradient descent, but architecture building and training set assembling! When everything start to make sense, it is really intriguing.

This course made me have great understanding around the Sequence Models such as GRU, LSTM, Attention, etc. I had a lot of fun while completing programming exercises such as Trigger word detection. As always, it was one of the best class ever. Thanks Professor Andrew Ng and all-!

Very good course. In quite short time you get understanding of a lot of principles and intuitions. The pace is good, explanations are consistent and clear, top-down approach from generic to specific, from simple to complex, very good instructional videos and interesting projects

Remarkably lucid exposition of complex learning architectures with directly applicable programming exercises. Highest recommendation. Thanks to the deep learning.ai staff for putting this entire specialization together and sharing their abundantly clear mastery of the subjects.

Some of the lectures were not quite up to par with professor Ng's standard. Some of the programming assignments were hard to follow and missing some details. Nevertheless, I came away with good understanding of sequence models and RNN. I can't thank enough. 5 stars from me.

This is very logical and especially the addition of probability between the words to improve predictions. It will be interesting to compare language , which language is easier to predict and why and study backward- how human creates them. We might learn something unexpected.

The lectures were super informative. It's almost unreal how easily he explains such difficult concepts, such they look a child's play. The coding assignments are incredibly informative and super interesting. I am very thankful for this specialization that it taught me so much

Before starting the course, I wanted to have a strong knowledge of the basics of Natural Language Processing as I wanted to specialize in this domain. I am thoroughly satisfied at the end of this course. This course has given me the confidence to dive deeper into this domain.

Course of lectures are excellent, but please fix the following problem

week 1 Programming Assignments

Improvise a Jazz Solo with an LSTM Network - v3

Dimensionality in djmodel()

https://www.coursera.org/learn/nlp-sequence-models/discussions/weeks/1/threads/NAoSHgf0Eei8aw6tWi-efA

I'll remain forever indebted to Andrew and his team for preparing this rigorous course. I can imagine the effort put in designing video lectures, going through research papers, crafting out these well-knit coding exercises. He has democratized AI and education in true sense.

I appreciate all the hard work and effort Andrew and his team puts in all his material.

I had hard time with most of Keras homework’s, I think it's hard to get the overall logic of a framework without an extensive explanation.

Besides that, the topics discussed are amazing.

A great introduction to Recurrent Neural Network models with lots of examples (text generation, music generation, sentiment analysis, word embedding, speech recognition, attention-based machine translations etc.). Thanks a lot to Andrew and the team for this awesome course.

Andrew, again explained complicated structures very clearly. The first week might be a bit overwhelming and you might get lost with huge information overflow, but trust me, in week2 and week3 you will see how the dots are connected. Thank you Andrew for this amazing course.

Thanks for making this course! I have been through all the courses in this specialization, and they are really excellent! Andrew has a great way of explaining things simply. It was easier for me to look at a couple of research papers after having gone through the lectures.