Awesome

Awesome

Thanks!

elegant

Amazing

great!

Great!

nice

Good

good

nice

Good

good

<3

Ok

GJ

s

Course material: 5 stars (although it could be more rigorous, this is part of an into to dln with Keras). The course dropped the requirement for code submission which I thought was a bad idea - code submission should be required. Tools: 3 stars - these are standard tools but honestly the tools are still pretty bad (by which I mean you have to use them a bit to get used to them - I have always objected to this is software development and in code I wrote conventional was not relied upon as a requirement).

The lectures were great. And I liked that there were still examples for us to work through like the previous courses in the specialization. That being said there were frequently concepts that seemed to be introduced in the examples that were never before mentioned and thus seemed out of place as they were not necessary to complete the assignment. It might be helpful to include short introductory statements to some of these so that we can better learn when/why some of these concepts are used.

I think I might say this for every course of this specialisation:

Great content all around!

It has some great colab examples explaining how to put these models into action on TensorFlow, which I'm know I'm going to revisit time and again.

There's only one thing that I think it might not be quite so good: the evaluation of the course. There isn't one, apart from the quizes. A bit more evaluation steps, as per in Andrew's Deep Learning Specialisation, would require more commitment from students.

This a great course on it own, it contains the fundamentals for natural language processing, from the encodings, embeddings and all the process involved before you can actually use the sequences into recurrent neural networks. I was hoping to do more exercises and with a higher difficulty than the ones defined here that are more focussed on the fundamentals. I mean these were good, the pre-processing is always good but I would like more design/program more models.

The explanations in the videos are good. And you get a fast intro into NLP with Tensorflow (Keras) with good, working code examples. However, due to the shortness of the course, it lacks quite some depth. The biggest disadvantage in my view is that often the programming exercises are not graded. This course is intended to give you practical skills. Then, the programming needs to be graded and cannot be optional.

Again this course was pretty fast (I'm starting to feel like all four courses together are about the length of one standard course). One downside is there are not graded programming exercises like in the Convolutional Neural Networks course. I learned how to use the text processing tools in the keras API. Also was cool to see how effective the stacked LSTM models are.

Very detailed step by step tutorials of using Tensorflow with lots of effort to make things as easy to understand as possible. The use cases also quite interesting. A little lack of theory comparing to other courses by deeplearning.ai. Quizzes are quite undeveloped. But that is understandable, because the main goal of the course to introduce Tensoflow.

Gives you a nice overall understanding of what NLP is. There are notebooks to play with concepts. However, this course dials down on the practical aspect (and the theoretical one) even more than the previous course. I think the students will benefit more if more ground is covered on the theoretical aspect of RNNs, LSTMs, and GRUs. Nice course overall.