[Term Entry] PyTorch Sigmoid Example #8105
Conversation
|
|
Updated the title to '.sigmoid()' and clarified the description of the sigmoid function.
mamtawardhani
left a comment
mamtawardhani
left a comment
There was a problem hiding this comment.
Hey @aedpluto, I've requested some changes in this entry, could you please check and tag me once those are done? Thank you! 😄
| @@ -0,0 +1,45 @@ | |||
| --- | |||
| Title: '.sigmoid()' | |||
| Description: 'The sigmoid function is an S-shaped curve typically used in binary classification problems' | |||
There was a problem hiding this comment.
Description should always start with a verb, and this description is slightly incorrect. A correct example description could be: "Applies the sigmoid activation to each element of a tensor, mapping values to a range between 0 and 1."
| A **`.sigmoid()`** function is an S-shaped curve which maps any real-valued input to a bounded output, typically between 0 and 1. Sigmoid functions are regularly used as activation functions in non-linear classification problems, for example in neural networks, where the probability of a binary outcome is required. | ||
|
|
||
| Although there are a range of sigmoid functions, in the field of AI the sigmoid function is usually synonymous with the logistic function, which is bound between 0 and 1. |
There was a problem hiding this comment.
Can this be made better? It is modtly correct, but it needs a small adjustment for precision and tone, especially since this is about PyTorch’s .sigmoid() and not sigmoid functions in general. Example:
"The .sigmoid() function applies the sigmoid (logistic) function to each element of a tensor, producing an S-shaped curve that maps any real-valued input to a value between 0 and 1.
In machine learning, sigmoid is commonly used as an activation function in binary classification tasks, where outputs represent probabilities."
| The formula for the sigmoid function is given by: | ||
| $$σ(x) = 1 / (1 + e^(-x))$$ | ||
|
|
||
| ## Examples |
There was a problem hiding this comment.
| ## Examples | |
| ## Example |
| x = torch.linspace(-10, 10, steps=400) | ||
| y = torch.sigmoid(x) | ||
|
|
||
| plt.plot(x.numpy(), y.numpy() |
There was a problem hiding this comment.
The parenthesis are not closing and that will give an error, please fix that
| plt.xlabel("x") | ||
| plt.ylabel("σ(x)") | ||
| plt.show() | ||
| ``` |
There was a problem hiding this comment.
Please attach the image of the output here, and I also suggest to add one more example that shows practical uses, example Applying .sigmoid() to a Tensor
There was a problem hiding this comment.
Here's the image of the output.
mamtawardhani
added
python
Many thanks - I've now updated the file with the changes. I just wanted to double check whether I tag you in a comment like this when I reupload the file or do I tag you somewhere else? (I'm very new to collaborating on GitHub so wanted to double check) |
|
@mamtawardhani I've tried to rebase onto the latest branch and pushed the updated commit. Please let me know if anything else is needed. |
3 participants
Description
Issue Solved
Type of Change
Checklist
mainbranch.Issues Solvedsection.