# PyTorch Functions and Tensor Operations

5 Likes

Hi everyone,
first of all thanks @aakashns for the time and effort to create this course!

For those interested in understanding a little bit more some of the most famous torch functions, here is my first assignment!
The notebook covers the basics of `view`, `reshape`, `permute`, `flatten` and `cat`, why they are useful and what are the main differences!

Any feedback would be greatly appreciated, here is the link!

2 Likes

Hi folks,
I am so glad to be in this course, my best thanks to @aakashns for so much time and effort spent, yet offering this course for free. I am wowed!
folks, I just completed my first assignment in the Pytorch module and I explained five interesting torch functions, they include:

1. chunk
2. transpose
3. rand
5. inverse
I would be glad to receive your feedback, suggestions and even corrections too, enjoy the rest of the course!. Zero to Gans!
4 Likes

nice work! @jacopo-repossi, truly, the functions view, reshape, permute and even flatten are quite too similar in a way, you were able to safely guide me through the difference, good job and well done.

1 Like

Hi Guys,

1 Like

Thank you to @aakashns for the interesting course and to jovian for this cool cloud infrastructure!
Good to see other guys sharing their assignments here, I find them useful and I hope to complete my own asapâ€¦

Edit
Here it is now!

Hey guys,
Here is my notebook link to our first assignment

Thanks to @aakashns for creating such a great opportunity and hats off to all his efforts.

Hi everyone,
thanks to @aakashns for this very interesting course.
Here is my notebook about the first lesson.

Impressed by the job of you all, very motivating that I am not alone to learn.

Assignment 1 completed !!!

What will be the deadline for the first assignment on PyTorch Functions and Tensor Operations?
Bcoz Iâ€™ll be a little bit busy in these 3 days. So let me know soon.

Hey everyone! Hope youâ€™re all safe. Hereâ€™s the link to my first pytorch assignment - https://jovian.ai/lavsdio2019/01-tensor-operations.

Thanks @aakashns and Jovian team for your efforts!

can someone pls explain about the below error

@aakashns

1 Like

can someone tell about torch.sparse_coo_tensor function. i already read the explanation in the given reference but unable to understand.
@aakashns

Hello everyone, and thanks to @aakashns for this great course.
Here is my first assignment on tensor-operations - https://jovian.ai/kunalsb/01-tensor-operations

Hello, here is my notebook on our first assignment based on â€śPyTorch Functionsâ€ť - https://jovian.ai/sathi-satb/01-tensor-operations

1 Like

Hello everyone, here is my assignment about functions that can take Boolean tensors as input.

Hi all,
Here is a list of Pytorch functions on tensors

Hi everyone,

Thanks to @@aakashns and jovian team to provide such opportunity to enhance our knowledge.
I just completed my first assignment in the Pytorch module and I explained five interesting torch functions, they include:

• PyTorch fmod() function
• pytorch sin() function
• Pytorch torch.eye() function
• Pytorch cat() function

here is the link of my notebookhttps://jovian.ai/parveenrohilla06/assignment-1-pytorch-and-its-functions:

Deep Learning with Pytorch Assignment 1

Your `g` is a multi-dimensional tensor, which will thus throw this error.
During lesson 1, the `mse` loss was a scalar, not a multi-dimensional tensor, thatâ€™s why the `loss.backward()` was giving no problem in the computations.

If you need to calculate the gradiend for a multi-dimensional tensor, you could pass a `gradient` with the same shape as `g` :

``````g.backward(torch.ones_like(g))