I recently got my TensorFlow Developer Certificate. Here are some tips I learned from my latest experience with the exam.

Preparation: Get familiar with Python coding and TensorFlow (of course!). Take some courses if needed. There are plenty of online courses available these days. For example, I found TensorFlow in Practice helpful to my preparation. Also, make sure you read the candidate handbook before taking the exam.

PyCharm setup: You’ll take the exam in PyCharm. So it’ll be a good idea to play around a bit with this IDE. I recommend setting up a practice project according to the instructions in candidate handbook, so that you’ll have an idea what the environment will look like.

Build your collection of reference code: The exam covers a variety of TensorFlow use cases: Computer Vision, NLP, Time Series. It’ll be a good idea to take notes of typical models and code snippets, and build a collection of reference code. They can be very useful when you want to quickly check something or if you have doubts about the code you are writing.

Get handy! In addition to taking notes, make sure you exercise at least once on each topic. Examine some real-world data by yourself, write Python program that actually runs. This will give you good sense of how you would process your data and what you’ll need to do to build a workable model. For example, I wrote programs that covered these topics:

• Basic DNN
• CNN (binary classification, multi-class classification)
• CNN (transfer learning)
• NLP: classification
• NLP: text generation
• Timer series models: DNN, RNN, LSTM

Do some “shape analysis”. Make sure you understand the shapes involved in your model. I find this a useful way to get more insight into your model. You should at least be able to tell things like: what shape(s) are expected at input/output of your model, and why. For example, why an LSTM model outputs 2-D prediction, while another one outputs 3-D prediction? If you get a shape mismatch error, how would you solve it?

These are my tips for preparing the TensorFlow Developer Certificate exam. I hope they can be useful to you too. Thanks for reading, and enjoy TensorFlow!