About Me

• Name: Vito Ghifari
• NIM: 16520255

TensorFlow: A Google Developer Product

First things first, what is TensorFlow?

TensorFlow is an open source library from Google that mainly used to create machine learning models. This library is easy to learn, both by beginner and expert in machine learning field.

What is TensorFlow used for?

Actually, TensorFlow is used in many fields. The model can be used for research, development, or even for production. You can use TensorFlow on the web, on mobile devices (by TensorFlow lite), and many more. Most Google products use TensorFlow for machine learning purposes. Technically speaking, TensorFlow is used for image and text classification, language translation, regression, etc.

Why use TensorFlow?

• As we read from the explanation above, TensorFlow can be used in a wide range of fields.
• The machine learning model is easy to build, especially by using Keras, a higher-level API.
• Powerful experimentation for research.

What are the advantages and disadvantages of using TensorFlow?

✅ Advantages:

• Open-source.
• Good community support.
• GPU and TPU (Tensor Processing Unit) support. The model can be trained much faster by utilizing GPU or TPU than just using CPU.
• The performance of the model can be visualized by using TensorBoard.
• Can be used for data preprocessing.

⛔ Disadvantages:

• Long learning curve. TensorFlow is difficult for those who have no prior experience in programming and deep learning.
• Can be hard to debug.
• Implementation for complex architecture is difficult.

Are there any alternatives rather than using TensorFlow?

Definitely!

For Python, there are several alternatives for deep learning frameworks that I've known:

• Theano
• PyTorch
• Caffe
• Scikit-learn
• MXNet

Show me an example of TensorFlow project!

Sure. You can go here to check about my repository about TensorFlow implementation in Python.