Ortal Arel
8.3/10
(Our Score)
Product is rated as #265 in category Python
Learn the essential foundations of AI: the programming tools (Python, NumPy, PyTorch), the math (calculus and linear algebra), and the key techniques of neural networks (gradient descent and backpropagation).
Instructor Details
Votes: 0
Courses : 1
Specification: Learn AI Programming with Python
|
9 reviews for Learn AI Programming with Python
Show all
Most Helpful
Highest Rating
Lowest Rating
Add a review
| Price | $537 |
|---|---|
| Provider | |
| Duration | 120 hours |
| Year | 2020 |
| Level | Beginner |
| Language | English |
| Certificate | Yes |
Learn AI Programming with Python
$537.00
Elie K. –
This was a very good program, I learned a lot from it. Now I have a good understanding of AI network and how they are trained and used. and I’m ready to dive a bit deeper, and go to tthe next level All material presented was sufficient to complete the project. But with many personal interruptions, I felt that I had to do a treasure hunt to find the info to put together the final projects; we had bits and pieces of all the info, but with variation on implementation, and probably with some confusion on Pytorch usage (reading from files, or directory structure, or going from numpy to pytorch, etc). This was compounded by the inability to train torchvision VGG16 model with CUDA on my own NVidia 780 GTX TI with 3GB of ram (I was running out of memory, but I was able to do “prediction”). And the fact that I was getting disconnected every so often when running on Udacity server with CUDA (repeat, restart, etc), and not getting convergence, while trying to debug the code, and trying to internalize the transform and other factors, made it a bit difficult to complete some tasks in one sitting. I wish there were some hints, and more indication on what to look for; e.g. it would have been nice to about trying different lower value “LearnRate”, I would have saved a week. And there are many gotchas too: like saving the model parameters ( which ones), and if you save in CUDA, the model would not run on my desktop ( which has no CUDA). Also, it was nice to have classmate comments and solution column, but the web–format was painful to use (hard to read, and follow)& But I ’m very thankful for the few hints I got from other classmates.
Amresh V. –
It is a great course for getting started with AI. I would say I found it easy to cope up even when I have a day job. I have experience with Python and maths involved in the course. My reason behind taking this course was to develop an intuition in solving AI/DL problems and I am not disappointed. At the same time, I am not sure about someone who is completely new to the field. There is a lot to cover, python is a great programming language and numpy, pandas, seaborn are quite dense topics on their own. Then comes the maths. Great lessons by Grant Sanderson, really helpful visualizations for all the vector and matrix operations. In the end I feel the course managed to give me a good taste of what I was looking for: how to solve problems using AI (I come from a web development background). I feel confident about myself now. Cheers!
Matt P. –
Great program. It’s clear, meaningful and cuts out a lot of noise when compared to other training programs. Also, I’m studying at one of the most elite tech universities in the world, and I can say that these modules are a far better source of knowledge (better structures, better content and far more practical in terms of delivery methods). The only thing I’d recommend is to have the option to be able to download a digital PDF of all the content for future references. I’m very grateful for Udacity and it’s brilliant Nanodegrees!
Hyungoo J. –
My program is going well. This is my second Nanodegree program, thus I am familiar with the concept and the structure of it. That is why it is easier to plan my study and to learn what I am looking for. I think of keeping learning about AI related courses at the advanced level, thus try not to hurry up but to advance step by step every single day. Thanks for the wonderful course and the people who are there to help the students.
Aaron B. –
Time management was hard. I somehow completed the final project but i doubt i will remember the python libraries in coming weeks. Udacity should enforce, student not only learn but able to recall. I suggest it introduce a minimum check–in time of 5 hours/week where the system monitors student performing some programming tasks. This repetitive work will go long way towards recalling libraries even after nanodegree completion.
andres g. –
Overall a good experience. One would always want to have a somewhat quicker reply from mentors and community staff but I completely understand that at this scale it is a tricky job. Course content is awesome and focuses on the important parts. Great job there. Also, the project is structured in a way that allows us to think on our own but gives enough guidance to get there. I’m happy with it so far.
Zhanna B. –
I like this course because it gives knowledge both on basics of Python and on basics of machine learning. Some exersizes were challenging, so I needed more time to complete them. Study materials are informative and well structured. I also like the supportive services Udacity provides, like personal mentor, students forum, planner, study challenge #30DaysOfUdacity. Thanks to all this my studies were organized, regular and thus effective.
Hyungoo J. –
My program is going well. This is my second Nanodegree program, thus I am familiar with the concept and the structure of it. That is why it is easier to plan my study and to learn what I am looking for. I think of keeping learning about AI related courses at the advanced level, thus try not to hurry up but to advance step by step every single day. Thanks for the wonderful course and the people who are there to help the students.
Aaron B. –
Time management was hard. I somehow completed the final project but i doubt i will remember the python libraries in coming weeks. Udacity should enforce, student not only learn but able to recall. I suggest it introduce a minimum check in time of 5 hours/week where the system monitors student performing some programming tasks. This repetitive work will go long way towards recalling libraries even after nanodegree completion.