How to prepare for Tensorflow Exam

•

How to prepare for Tensorflow Exam 🎓 A lot of people asked how to learn TensorFlow and prepare for the certification exam, Here is how you can do it ↓

• Tensorflow is a free and open-source framework by Google that allows creating and deploying machine learning/deep learning models. • Tensorflow certification is an official exam that you can take and have a trophy, although it's not necessary for learning.

• The major topics concerned with the certification are: ‣ Tensorflow basics ‣ Training models with TF 2.X ‣ Convolutional Networks in TF2 ‣ Natural Language Processing in TF2 ‣ Time Series Forecasting in TF2

• If you have been a Tensorflow user for a while (like I am) and are very familiar but just want the certification, you don't need a lot. • Just go through this course by Laurence Moroney on Coursera once, some practice and you should be fine. coursera.org/professional-certificates/tensorflow-in-practice

• If you are not a regular TensorFlow user but have some knowledge about it then I'd suggest you go through the brilliant course by Daniel Bourke. • It covers all the mentioned topics above and more, it's good for learning TensorFlow overall, not just for certification.

• You can find the video tutorial here: udemy.com/course/tensorflow-developer-certificate-machine-learning-zero-to-mastery/

• There's also a free online book that goes with the course. You can refer to that and assess if you need the videos or not. dev.mrdbourke.com/tensorflow-deep-learning/11_passing_the_tensorflow_developer_certification_exam/ • Also, you can still check out the very first course mentioned by Laurence Moroney if you have completed the previous one.

• If you are lacking sufficient knowledge in deep learning, I wouldn't suggest going for the certificate right away. • Get some detailed knowledge and intuition about each topic first, a lot of practice, and then aim for the Tensorflow certification.

Some information regarding the exam. • The cost for the exam is $100 (pricey! I know right!), but you can opt for a scholarship if you require some support. tensorflow.org/extras/cert/TF_Education_Stipend.pdf

• The price you pay is just for one attempt, no do-overs. • If your connection breaks, the system shuts down or aliens invade your city destroying everything, you're screwed and might need to pay again for the next try.

• The total time for the exam is 5 hours. You can obviously finish early. On a low-end system, training the model can be quite hard and lengthy. • Use Google Colab to train your models quickly, download them and place them in the question's directory where they can be found.

• There will be five questions on the exam and they won't be equally marked. • Questions are weighted according to their difficulty. Solving 4 easy and losing the 5th one can cost result in failing the exam.

• The exam is conducted inside the PyCharm IDE, so you should know your way around PyCharm. • You don't need a lot but just being able to set up a virtual environment and run python scripts is good enough. • Still, if you need tutorials visit here: jetbrains.com/pycharm/learn/

• All the mentioned facts are referred from the official site which has a lot more detail on whatever you need. tensorflow.org/certificate

• Once you complete certification, you can see yourself in the network of all the developers who have completed the certification here (It's pretty cool, honestly!). developers.google.com/certification/directory/tensorflow

• Finally. Don't overthink about certifications. They are cool to have not need to have. • Learning should be the first and foremost, certificates act only as a milestone and give some level of satisfaction.

Thanks for reading and good luck!