用于 conv2d 和手动加载图像的 Keras input_shape

问题描述

我正在从许多 384x286 黑白图像手动创建我的数据集.

I am manually creating my dataset from a number of 384x286 b/w images.

我加载这样的图像:

x = []
for f in files:
        img = Image.open(f)
        img.load()
        data = np.asarray(img, dtype="int32")
        x.append(data)
x = np.array(x)

这导致 x 是一个数组 (num_samples, 286, 384)

this results in x being an array (num_samples, 286, 384)

print(x.shape) => (100, 286, 384)

阅读 keras 文档并检查我的后端，我应该向卷积步骤提供由(行、列、通道)组成的 input_shape

reading the keras documentation, and checking my backend, i should provide to the convolution step an input_shape composed by ( rows, cols, channels )

因为我不知道样本大小，所以我希望作为输入大小传递，类似于

since i don't arbitrarily know the sample size, i would have expected to pass as an input size, something similar to

( None, 286, 384, 1 )

模型构建如下:

model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=input_shape))
# other steps...

作为 input_shape (286, 384, 1) 传递导致:

passing as input_shape (286, 384, 1) causes:

检查输入时出错:预期 conv2d_1_input 有 4 个维度，但得到形状为 (85, 286, 384) 的数组

传递 as_input_shape (None, 286, 384, 1 ) 导致:

passing as_input_shape (None, 286, 384, 1 ) causes:

输入 0 与层 conv2d_1 不兼容:预期 ndim=4，发现 ndim=5

我做错了什么?我该如何重塑输入数组?

what am i doing wrong ? how do i have to reshape the input array?

推荐答案

将 input_shape 设置为 (286,384,1).现在模型需要 4 个维度的输入.这意味着您必须使用 .reshape(n_images, 286, 384, 1) 重塑您的图像.现在，您已在不更改数据的情况下添加了额外的维度，并且您的模型已准备好运行.基本上，您需要将数据重塑为 (n_images、x_shape、y_shape、channels).

Set the input_shape to (286,384,1). Now the model expects an input with 4 dimensions. This means that you have to reshape your image with .reshape(n_images, 286, 384, 1). Now you have added an extra dimension without changing the data and your model is ready to run. Basically, you need to reshape your data to (n_images, x_shape, y_shape, channels).

很酷的一点是，您还可以使用 RGB 图像作为输入.只需将 channels 更改为 3.

The cool thing is that you also can use an RGB-image as input. Just change channels to 3.

也检查这个答案:Keras 输入解释:input_shape、units、batch_size、dim、等

示例

import numpy as np
from keras.models import Sequential
from keras.layers.convolutional import Convolution2D
from keras.layers.core import Flatten, Dense, Activation
from keras.utils import np_utils

#Create model
model = Sequential()
model.add(Convolution2D(32, kernel_size=(3, 3), activation='relu', input_shape=(286,384,1)))
model.add(Flatten())
model.add(Dense(2))
model.add(Activation('softmax'))

model.compile(loss='binary_crossentropy',
                  optimizer='adam',
                  metrics=['accuracy'])

#Create random data
n_images=100
data = np.random.randint(0,2,n_images*286*384)
labels = np.random.randint(0,2,n_images)
labels = np_utils.to_categorical(list(labels))

#add dimension to images
data = data.reshape(n_images,286,384,1)

#Fit model
model.fit(data, labels, verbose=1)

这篇关于用于 conv2d 和手动加载图像的 Keras input_shape的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！