adding docs for some torch.* functions #392

soumith · 2017-01-02T08:16:29Z

also, removing all, any stateless methods

Still left:

…ethods

torch/docs.py

+Returns a new `Tensor` with the arccosine  of the elements of :attr:`input`.
+
+Args:
+    tensor (Tensor): the input `Tensor`


torch/docs.py

+acos(input, out=None) -> Tensor

-Computes the element-wise inverse cosine of a tensor.
+Returns a new `Tensor` with the arccosine  of the elements of :attr:`input`.


torch/docs.py

+::
+
+    >>> a = torch.randn(4)
+    >>> print(a)


torch/docs.py


 add_docstr(torch._C.add,
 """
+.. function:: add(tensor, value, out=None)


torch/docs.py

+addbmm(beta=1, mat, alpha=1, batch1, batch2, out=None) -> Tensor
+
+Performs a batch matrix-matrix product of matrices stored in :attr:`batch1` and :attr:`batch2`, 
+with a reduced add step (all matrix multiplications get accumulated in a single place). 


torch/docs.py

+
+If :attr:`mat1` is a `n x m` Tensor, :attr:`mat2` is a `m x p` Tensor, :attr:`out` and :attr:`mat` will be `n x p` Tensors.
+
+`alpha` and `beta` are scaling factors on `mat1@mat2` and `mat` respectively.


torch/docs.py

+:math:`out = (beta * M) + (alpha * mat1 * mat2)`
+
+Args:
+    beta (float, optional): multiplier for :attr:`mat`


torch/docs.py

+
+If :attr:`mat` is a `n x m` Tensor, :attr:`vec` is a 1D Tensor of size `m`, :attr:`out` and :attr:`tensor` will be 1D of size `n`.
+
+`alpha` and `beta` are scaling factors on `mat*vec` and `tensor` respectively.


torch/docs.py

+If :attr:`vec1` is a vector of size `n` and :attr:`vec2` is a vector of size `m`, then :attr:`mat` must be a matrix of size `n x m`
+
+Args:
+    beta (float, optional): multiplier for :attr:`mat`


torch/docs.py

 """
+baddbmm(beta=1, mat, alpha=1, batch1, batch2, out=None) -> Tensor
+
+Performs a batch matrix-matrix product of matrices stored in :attr:`batch1` and :attr:`batch2`. 


torch/docs.py

+Returns a new `Tensor` with the arccosine  of the elements of :attr:`input`.
+
+Args:
+    tensor (Tensor): the input `Tensor`


torch/docs.py

+
+Args:
+    input (Tensor): the input `Tensor`
+    value (Float): the number to be added to each element of :attr:`input`


…ch.* docs as they are not stateless

docs/source/torch.rst

 .. autofunction:: addmm
 .. autofunction:: addmv
 .. autofunction:: addr
-.. autofunction:: all


torch/docs.py

+acos(input, out=None) -> Tensor

-Computes the element-wise inverse cosine of a tensor.
+Returns a new `Tensor` with the arccosine  of the elements of :attr:`input`.


torch/docs.py

 Example:
-    >>> torch.acos(torch.FloatTensor([1, -1]))
-    FloatTensor([0.0000, 3.1416])
+::


torch/docs.py

+Example:
+::
+
+    >>> a = torch.randn(4)


Replaced `sum` with `mean` in line pytorch#392

fix typo in cheat sheet (pytorch#381)

Fix reduction heuristics so we don't recompile and we use the correct launch params. Co-authored-by: Kevin Stephano <kevin.stephano@gmail.com>

These scripts should probably point to the main repository where these are published. Will do a follow up PR to make it configurable later on. Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

…Variable (pytorch#392) * Implement UnspecializedPrimitiveVariable codegen * Make UnspecilizedPrimitiveVariable as GraphArg * Update make_call_generated_code * Update min/max builtin func * Support random.random * Remove unnecessary change * Fix lint * Refactor to support multiple random.random * Refactor out unspecialized numpy and python variables * Fix RandomValueSource guard * Support multiple random functions * Rebase to updated main * Refactor out random_values_var * Fix lint * Fix lint * Move random_values_var to output graph * Add need_unwrap to distinguish unspec from x.item() * Make global rand func unique * Fix lint * Add raw value propagation for unspec variables * Fix lint * Directly load type(raw_value) & update random func example value * Fix lint

soumith added 3 commits January 2, 2017 03:15

adding docs for some torch.* functions, removing all, any stateless m…

9158d52

…ethods

add doc for *mm* functions, *mv* functions and addcmul, addcdiv

78e2197

fix indentation issue for all examples, add doc for add

03fe236

apaszke reviewed Jan 2, 2017

View reviewed changes

soumith commented Jan 3, 2017

View reviewed changes

torch/docs.py Outdated

Returns a new `Tensor` with the arccosine of the elements of :attr:`input`.

Args:

tensor (Tensor): the input `Tensor`

This comment was marked as off-topic.

Sign in to view

fixed names and other cosmetics

47d5ef7

apaszke reviewed Jan 3, 2017

View reviewed changes

soumith added 4 commits January 2, 2017 19:37

Float -> float, Long -> long

0f352e3

implement more torch.* docs, remove zero, cauchy, log_normal from tor…

a38e3cd

…ch.* docs as they are not stateless

adding linspace, logspace, neg and range

af028c5

adding docs for more torch.* functions

a4d01ab

colesbury reviewed Jan 3, 2017

View reviewed changes

torch/docs.py Outdated

Example:

>>> torch.acos(torch.FloatTensor([1, -1]))

FloatTensor([0.0000, 3.1416])

::

This comment was marked as off-topic.

Sign in to view

This comment was marked as off-topic.

Sign in to view

colesbury reviewed Jan 3, 2017

View reviewed changes

torch/docs.py

Example:

::

>>> a = torch.randn(4)

This comment was marked as off-topic.

Sign in to view

This comment was marked as off-topic.

Sign in to view

This comment was marked as off-topic.

Sign in to view

soumith merged commit a461804 into master Jan 3, 2017

soumith deleted the moredocs branch January 3, 2017 23:29

simaiden added a commit to simaiden/pytorch that referenced this pull request Feb 26, 2020

Mistake in MSELoss documentation

792067f

Replaced `sum` with `mean` in line pytorch#392

simaiden mentioned this pull request Feb 26, 2020

Mistake in MSELoss documentation simaiden/pytorch#1

Closed

mrshenli pushed a commit to mrshenli/pytorch that referenced this pull request Apr 11, 2020

Merge pull request pytorch#392 from 9bow/fix_typo_in_cheat_sheet

d1b279f

fix typo in cheat sheet (pytorch#381)


		If :attr:`mat1` is a `n x m` Tensor, :attr:`mat2` is a `m x p` Tensor, :attr:`out` and :attr:`mat` will be `n x p` Tensors.

		`alpha` and `beta` are scaling factors on `mat1@mat2` and `mat` respectively.


		If :attr:`mat` is a `n x m` Tensor, :attr:`vec` is a 1D Tensor of size `m`, :attr:`out` and :attr:`tensor` will be 1D of size `n`.

		`alpha` and `beta` are scaling factors on `mat*vec` and `tensor` respectively.

adding docs for some torch.* functions #392

adding docs for some torch.* functions #392

Uh oh!

Conversation

soumith commented Jan 2, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

This comment was marked as off-topic.

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

soumith commented Jan 2, 2017 •

edited

Loading