[2026] Tensorflow 2: Deep Learning & Artificial Intelligence (Udemy.com)

I am a robotics researcher who initially applied for this course because of the computer vision segment, wasn't exactly keen on the NLP or recommender systems. But it's really great that the lectures have placed deep learning in context to the real world while delving into the technicalities. At least now I have a greater understanding of the significance and implications of deep learning in the real world.

Too slow pace, a lot of dragging, e.g. there is no need to read out formulas. Everyone can read them, and who cant is probably not the target group of the course. It is a terrible idea to split code between two repos. Let them be python files or notebook files, but they should all reside in github in one repo. The material about recommender systems is completely useless. Seems like a copy-paste from a 15 years old source (from the days of the famous Netflix competition). Small and medium size businesses do not have enough ratings data to predict ratings, and large retailers (e.g. Amazon, Netflix) have in-house AI teams and dont need this course. A modern recommender system should be based on some wide and deep network or agents using customers' and products' side features. Sections 20-22 do not belong here. They are a blog post material.

This course is just a Frankenstein-ed consolidation of the various subjects from other courses in the Lazy Programmer catalogue but without the heavily detailed mathematical derivations. If you've taken those courses, you'll find you've already written most of the code or maybe even all of it. For the courses I haven't taken (e.g., reinforcement learning), I cannot speak as to how much of the theory I'm missing. Bottom line: if you've taken all or most of Lazy Programmer's other courses at this point, you might as well just skip this one. If you've taken very few or none of those courses, you'll find this to be very useful and comprehensive.

This is amazing course. I have enrolled into several such courses. Content wise this is the best one. For any new learners: There are few to none interactive sessions. So you might get bored. Make sure you are paying attention. Also make sure you are paying sharp attention to what he is saying. All the equations roll of instructor's tongue. So there is chance you might miss it. Make sure, if needed, go through lessons twice.

This course offers phenomenal value for its cost. It carries out its objective of providing a "bird's eye" view of Tensorflow and machine learning topics quite well. I graduated from CS school in 2020 and I would have loved to have known about this course when I took introductory AI and ML classes, as it would have complemented university lectures in a clear and concise way, demistifying a considerable portion of the math behind ML. The instructor can sound a bit grumpy in the videos sometimes, but that's part of their view of AI/ML as a "non-magical" (but still awesome) field of study that requires individual effort, which I completely agree with.

I have already taken 4 to 5 ML courses on Supervised and Unsupervised models, Plotly visualization, Python Programming, basic Tensorflow 1.0 courses. All those courses mostly focused on coding skills rather than more detailed explanation. I think this course gave me a very nice background on mathematical side, code along capabilities and indirectly I had to do my own google research to understand from DataScience.com, medium.com etc. The mathematical part is not fully covered with derivations, but i could understand in googling and Khan Academy web sites with full derivations using Algebra and Calculus. I do believe that i do need to attend some sort of Statistics course to get more details. I think if I can keep the understanding in fingertips like Normal/Random distribution, what is Gaussian/Bernouli distribution etc. all the time on my notes, it would be very helpful. After taking all these course, I still lag the confidence to apply the models. I do believe that I do need take some data sets from web and develop model predictions which might boost my confidence level. I truly want to become a ML programmer and I am in the making process, but still wondering how to become a good ML programmer.

As a total beginner in the area of deep learning, this course has opened my mind in a more constructive way. Lectures were well structured with parallel accent on both theory and code related with the use of TensorFlow library. Solid math foundation is important for tracking and learning while taking this course. Expectations are above everything that could be imagined. This course will definitely force me to continue with further learning about the use of neural networks and ways how models in deep learning are being implemented. Advises that author gave are very precious and every professional should stick with them if he wants to make progress in his area. Huge recommendation!

Hi, Lazy Programmer! It was a great pleasure taking this course. Thank you very much. The next course will be A/B testing. Now, couple of words about this one. Reinforcement Learning was a real pain because of huge amount of theory. I wish you had dedicated a bunch of lessons to different kind of autoencoders. I really miss this topic. But the whole course is very balanced in terms of theory/practice ratio. Excellent!

I am simply impressed with how much depth can be put into a single course. I have now completed four courses by this guy and I am continuously impressed with the quality of the lectures. He is very good at explaining complex concepts precisely, without relying too much on intuition or metaphors, but instead go into the mathematics. Compared to the earlier courses the coding part of this course is of much higher quality. I learned a lot directly applicable to my work as a data scientist. Thank you.

This is the finest course on TensorFlow you can ever get. It's a real course that covers up the complex math and the practical stuff in TensorFlow. It is a very well designed course, covers up all topics of Deep Learning with different data sets and code that we don't get elsewhere. LazyProgrammer is a true programmer and he is very authentic about the knowledge. I literally loved it. Thank you so much LazyProgrammer for this.