DequantizeLinear - 10 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Next section compares an older to a newer version of the same operator after both definition are converted into markdown text.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. DequantizeLinear10 → DequantizeLinear13 +7 -16
DequantizeLinear10 → DequantizeLinear13 RENAMED
@@ -1 +1 @@
1
- The linear dequantization operator. It consumes a quantized tensor, a scale, and a zero point to compute the full precision tensor.
1
+ The linear dequantization operator. It consumes a quantized tensor, a scale, a zero point to compute the full precision tensor.
2
- The dequantization formula is y = (x - x_zero_point) * x_scale. 'x_scale' and 'x_zero_point' must have same shape, and can be either a scalar
2
+ The dequantization formula is y = (x - x_zero_point) * x_scale. 'x_scale' and 'x_zero_point' are both scalars.
3
- for per-tensor / per layer quantization, or a 1-D tensor for per-axis quantization.
4
3
  'x_zero_point' and 'x' must have same type. 'x' and 'y' must have same shape. In the case of dequantizing int32,
5
4
  there's no zero point (zero point is supposed to be 0).
6
-
7
- **Attributes**
8
-
9
- * **axis**:
10
- (Optional) The axis of the dequantizing dimension of the input
11
- tensor. Ignored for per-tensor quantization. Negative value means
12
- counting dimensions from the back. Accepted range is [-r, r-1] where
13
- r = rank(input).
14
5
  **Inputs**
15
6
  Between 2 and 3 inputs.
16
7
  * **x** (heterogeneous) - **T**:
17
8
  N-D quantized input tensor to be de-quantized.
18
9
  * **x_scale** (heterogeneous) - **tensor(float)**:
19
- Scale for input 'x'. It can be a scalar, which means a per-
10
+ Scale for input 'x'. It's a scalar, which means a per-tensor/layer
20
- tensor/layer dequantization, or a 1-D tensor for per-axis
21
- dequantization.
11
+ quantization.
22
12
  * **x_zero_point** (optional, heterogeneous) - **T**:
23
- Zero point for input 'x'. Shape must match x_scale. It's optional.
13
+ Zero point for input 'x'. It's a scalar, which means a per-
14
+ tensor/layer quantization. It's optional. 0 is the default value
24
- Zero point is 0 when it's not specified.
15
+ when it's not specified.
25
16
  **Outputs**
26
17
  * **y** (heterogeneous) - **tensor(float)**:
27
18
  N-D full precision output tensor. It has same shape as input 'x'.
28
19
  **Type Constraints**
29
20
  * **T** in (
30
21
  tensor(int32),
31
22
  tensor(int8),
32
23
  tensor(uint8)
33
24
  ):
34
25
  Constrain 'x_zero_point' and 'x' to 8-bit/32-bit integer tensor.