Hi,
I have just completed Course 2. While i understand all of the topics discussed in the lectures and I understand the broader idea in the labs I feel as if there is a lot of tensorflow functions being used to develop the scripts and I look at it and think “wow I really wouldn’t have any idea how to build this” as I have no previous exposure to TF.
I think once I finish this specialisation I will then completed the Tensorflow Specialisation offered on Coursera as well. However I feel like I am almost missing out on part of the learning because of this. Does anyone else feel this way?
1 Like
Took some courses in this specializations. Like you said, there are ways to do the same without TF functions. For instance in the data cleaning and pipeline stage, we can use pandas and sk learn functions.
@vsnupoudel, yes I think that the beauty of the process is that it is as much art as it is science and that art can come down to what resources you use to do to do your preprocessing etc.
However I really do think that a few examples in training by completing the same labs with different methods would have been greatly useful.
1 Like
Hi Balaji,
I have seen this one and was interested. The reason I chose this specialisation is that I am an engineer looking to apply these practices in the workplace.
In your experience would you recommend the deep learning specialisation for addressing the shortcomings I mentioned in the initial post? Or would a straight tensorflow course be more worthwhile.
If you’re serious about deep learning, deep learning specialization the way to go.
If your priority is to quickly ramp up on tensorflow (there’s less emphasis on theory and more on using the framework), see DeepLearning.AI TensorFlow Developer
Would completing the latter give me the understanding of the framework to better comprehend the MLOps course?
You should read the syllabus and decide for yourself.