This commit will introduce gaussian filler in tflchef
And also recipe of readme will be modified
Signed-off-by: Seok NamKoong <sk.namkoong@samsung.com>
## Tutorial: How to use?
This example explains how to generate a tensorflow lite model with a single Conv2D operation
-with constant (1.1f) bias with _tflchef_.
+with a kernel filled with random values generated according to normal (or gaussian) distribution (mean = 0.0f / stddev = 1.0f) and bias with constant values (1.1f) with _tflchef_.
The first step is to write a recipe!
Type the following command, and then you may get ``sample.recipe``:
name: "ker"
type: FLOAT32
shape { dim: 1 dim: 1 dim: 1 dim: 2 }
+ filler {
+ tag: "gaussian"
+ arg: "0.0"
+ arg: "1.0"
+ }
}
operand {
name: "bias"
name: "ker"
type: FLOAT32
shape { dim: 1 dim: 1 dim: 1 dim: 2 }
+ filler {
+ tag: "gaussian"
+ arg: "0.0"
+ arg: "1.0"
+ }
}
operand {
name: "bias"