There are so many points which I liked aboutTensorFlow i.e. It is fast and it's scalability on the larger dataset, Witht the help of TensorFlow I am able to write customisable evaluation functions.

Tensorflow is an excellent library for implementing linear algebra equations and algorithms. It also has Keras as its inbuilt module, a perfect module for deep learning and implementing neural network models. I use it majorly for implementing and training deep learning models. It provides high customizability for defining our loss functions, activation functions, etc.

Python based and API is intuitive. Keras is great and uses the Tensorflow library. I used scikit-learn prior and it was so much harder to understand and require way more code to get the same things done. The user-friendly interface is honestly the best part of Tensorflow/Keras.