如何在tensorflow/keras中使用预定的内核列表初始化Conv2D层？

Question

我想使用Conv2D层来跨越输入图像并运行三个2x2内核

这不是tensorflow的目的，但我真的希望使用tensorflow作为后端引擎来高效运行内核，并在不同的设备GPU和/或CPU之间分配工作负载

我尝试了如下代码。但它似乎不太管用

import tensorflow as tf

class InitConvKernels(tf.keras.initializers.Initializer):

  def __init__(self, num_kernels, kernel_tensor):
    self.kernel_list= kernel_tensor
    self.index = -1
    self.num_kernels = num_kernels

  def __call__(self, shape, dtype=None):
    index += 1 
    assert(self.index <= self.num_kernels) # doesn't affect anything
    tf.print(shape) # doesn't work
    return self.kernel_list[index]

  def get_config(self):
    return {'kernel_list': self.kernel_list, 'num_kernels': self.num_kernels}

我正在调用自定义初始值设定项，但返回的层为空：

kernel_list = tf.constant([[[-1, -1],  [-1, -1]], [[1, 1],   [1, 1]],  [[-1, 1],  [1, -1]],])
layer = layers.Conv2D(
    filters=3,
    kernel_size=2,
    kernel_initializer=InitConvKernels(3,kernel_list),
    bias_initializer=initializers.Zeros()
)

layer.variables为空（[]） layer.layer.get_weights()也是空的（[]）

我的目标是评估输入图像上kernel_list中三个内核的卷积，并汇总所有结果

Answer 1

from PIL import Image
import requests
from io import BytesIO
import numpy as np
from keras.models import Sequential
from keras.layers import Conv2D


response = requests.get('https://upload.wikimedia.org/wikipedia/commons/thumb/0/02/Stack_Overflow_logo.svg/1280px-Stack_Overflow_logo.svg.png')
image = Image.open(BytesIO(response.content))

正在加载image from url。

构建一个运行kernel的模型（运行更多内核，使kernel_init成为生成器，并在初始化Conv2D时随时调整过滤器的数量）

def kernel_init(shape, dtype=None, partition_info=None):
    kernel = np.zeros(shape)
    kernel[:,:,0,0] = np.array([[1,0,1],[-1,0,-1],[1,0,1]])
    return kernel

#Build Keras model
model = Sequential()
model.add(Conv2D(1, [3,3], kernel_initializer=kernel_init, 
                 input_shape=(251,1280,4), padding="valid"))
model.build()

# To apply existing filter, we use predict with no training
out = model.predict(image)

以及可视化输出：

import matplotlib.pyplot as plt
plt.matshow(out[0,:,:,0])

编辑： 值得一提的是OpenAI's Triton，它可以帮助使用更高级的语言和框架（如pytorch）来运行高效的GPU代码：

Python-like programming language which enables researchers with no CUDA experience to write highly efficient GPU code—most of the time on par with what an expert would be able to produce.