Python scikit-learn vs Tensor-flow

Question

For those familiar with both scikit-learn and tensor, I will like to know which is better for Machine Learning (ML) especially in the area of Robotics and Automation. Thank you.

Answer 1

Well Scikit Learn is like the Swiss Army knife of the Machine Learning world, its easy to get up and running with, just stick to the 4 principles: import, instantiate, fit, predict. Keras, which is another powerful library was built upon Tensorflow, and I also like Tensorflow because it gives more granular control allowing you to take advantage of hardware GPUs, algorithm optimization, etc.
I like both, they both have their place. Generally speaking, if you can make it in SciKit Learn, you can probably make it better with Tensorflow.

Answer 2

That sounds great. I have already completed the Machine Learning in sololearn. But I am wondering if I still need to learn tensorflow. It sounds like I should add the tensorflow.

Answer 3

Tensorflow is a deep learning framework which is used for large scale projects, typically large Neural Networks, performing complex parallel computations...
I say do it, watch a few tutorials, maybe hop over to Kaggle.com and play with it over there. One problem with Tensorflow is it can become a resource hog, this might be why it is not available here on Sololearn