LogSoftmax - 11 vs 13#

Next section compares an older to a newer version of the same operator after both definition are converted into markdown text. Green means an addition to the newer version, red means a deletion. Anything else is unchanged.

Files changed (1) hide show
  1. LogSoftmax11 → LogSoftmax13 +19 -10
LogSoftmax11 → LogSoftmax13 RENAMED
@@ -1 +1 @@
1
- The operator computes the log of softmax values for the given input:
1
+ The operator computes the logsoftmax (log of softmax) values for each layer in the batch
2
+ of the given input.
3
+ The input does not need to explicitly be a 2D vector; rather, it will be
2
- LogSoftmax(input, axis) = Log(Softmax(input, axis=axis))
4
+ coerced into one. For an arbitrary n-dimensional tensor
3
-
5
+ input in [a_0, a_1, ..., a_{k-1}, a_k, ..., a_{n-1}] and k is
4
- The "axis" attribute indicates the dimension along which LogSoftmax
6
+ the axis provided, then input will be coerced into a 2-dimensional tensor with
7
+ dimensions [a_0 * ... * a_{k-1}, a_k * ... * a_{n-1}]. For the default
8
+ case where axis=1, this means the input tensor will be coerced into a 2D tensor
9
+ of dimensions [a_0, a_1 * ... * a_{n-1}], where a_0 is often the batch size.
10
+ In this situation, we must have a_0 = N and a_1 * ... * a_{n-1} = D.
11
+ Each of these dimensions must be matched correctly, or else the operator
5
- will be performed. The output tensor has the same shape
12
+ will throw errors. The output tensor has the same shape
6
- and contains the LogSoftmax values of the corresponding input.
13
+ and contains the logsoftmax values of the corresponding input.
7
14
  **Attributes**
8
15
  * **axis**:
9
- Describes the dimension LogSoftmax will be performed on. Negative
16
+ Describes the axis of the inputs when coerced to 2D; defaults to one
17
+ because the 0th axis most likely describes the batch_size. Negative
10
18
  value means counting dimensions from the back. Accepted range is
11
19
  [-r, r-1] where r = rank(input).
12
20
  **Inputs**
13
21
  * **input** (heterogeneous) - **T**:
14
- The input tensor of rank >= axis.
22
+ The input tensor that's coerced into a 2D matrix of size (NxD) as
23
+ described above.
15
24
  **Outputs**
16
25
  * **output** (heterogeneous) - **T**:
17
- The output values with the same shape as the input tensor.
26
+ The output values with the same shape as input tensor (the original
27
+ size without coercion).
18
28
  **Type Constraints**
19
29
  * **T** in (
20
- tensor(bfloat16),
21
30
  tensor(double),
22
31
  tensor(float),
23
32
  tensor(float16)
24
33
  ):
25
34
  Constrain input and output types to float tensors.