I am working on a project in which I need to advance my skills in neural network programming using Python. I am currently attending Andrew Ng's courses on Neural Networks. (I reached the fourth course on Convolutional Neural Networks).
After completing this course, I plan to enhance my skills as a Pythonista. I am wondering, shall I start learning Tensorflow or PyTorch?
Pytorch is a smaller framework than Tensorflow, so may be easier to start off with. However, I have found that Pytorch documentation is sometime lacking, when compared to Tensorflow, so may require to be played around more. One thing I feel Tensorflow has over Pytorch is that it is more easily scalable and the models can be more production ready.
Neither one nor the other, I strongly recommend starting with Keras, which runs over Google TensorFlow!
I wish you good studies!
Both of them have their advantages as starting platforms to get into neural network programming. If you are in academia and you are getting started, Pytorch is your best choice. If you are in the industry where you need to deploy models in production, Tensorflow is your best choice.
Dear Rodrigo Clemente Thom De Souza
I appreciate your response. Thank you.
Dear Zahra Taheri
Thank you for your answer. That was really helpful.
for the academia is Pytorch is the best and other commerical choose Tensorflows. however from my personal experience , pytorch has less usergroup and tutorial available compared to tensrflow..
It depends on the objective of your study.
Pytorch has less features, easy to learn, it create dynamic graphs, better for passion projects and building rapid prototypes
Tensorflow is bit hard to learn, high level features, it create static graphs, better for passion projects and building rapid prototypes
Hi Alia Alkameli
you can find the following links for your reference
https://towardsdatascience.com/tensorflow-or-pytorch-146f5397278a
https://towardsdatascience.com/pytorch-vs-tensorflow-2021-d403504d7bc3
https://realpython.com/pytorch-vs-tensorflow/
1.If you’re already a Python programmer, you may find PyTorch a little easier to use than TensorFlow due to the different coding styles these frameworks encourage.
PyTorch will likely be more accessible to beginners. It will probably be a better fit if you’re interested in quick, hackier solutions or if you’re experimenting with machine learning for the first time. It’s also likely to be the better option if you prefer a more imperative, Pythonic coding style, rather than the declarative style that TensorFlow will require.
2.Tensorflow has a more steep learning curve than PyTorch. PyTorch is more pythonic and building ML models feels more intuitive. On the other hand, for using Tensorflow, you will have to learn a bit more about it’s working and so it becomes a bit more difficult to learn Tensorflow than PyTorch.
3.TensorFlow includes a visualization toolkit, TensorBoard, that enables the display of data dashboards. PyTorch supports a visualization toolkit of its own, Visdom, but it is not as fully featured as TensorBoard. PyTorch also integrates with TensorBoard.
4.Tensorflow has a much bigger community behind it than PyTorch. This means that it becomes easier to find resources to learn Tensorflow and also, to find solutions to your problems.
5.Both frameworks currently allow distributed computing both on the GPU and CPU, enabling faster processing of models. However, TensorFlow’s distributed computing platform does offer an added advantage over PyTorch’s. Google, TensorFlow’s parent company, released the Tensor Processing Unit (TPU), which processes faster than GPUs. It is much easier to run code on a TPU using TensorFlow than it is on PyTorch.
6.When deploying deep learning models for production, TensorFlow has an advantage over PyTorch. TensorFlow provides tools such as TensorFlow.js, TensorFlow Serving, and TensorFlow Lite that enable the easy deployment of models on browsers, the cloud, and IoT devices. This is the main reason why a lot of companies have chosen TensorFlow as their preferred framework of choice when it comes to production.
Finally, Tensorflow is much better for production models and scalability. It was built to be production ready. Whereas, PyTorch is easier to learn and lighter to work with, and hence, is relatively better for passion projects and building rapid prototypes.
Dear Milad Vazan .. Many thanks for your detailed response .. It's really useful and I highly appreciate it
Dear Alia Alkameli
TensorFlow provides a better way of visualizing data with its graphical approach. It also allows easy debugging of nodes with the help of TensorBoard. This reduces the effort of visiting the whole code and effectively resolves the neural network.
- Badges
- Science topic