keras：expected density_1_input具有2维

Question

from keras import optimizers
from keras.models import load_model
from keras.preprocessing import image
import numpy as np
import scipy.misc
from keras.wrappers.scikit_learn import KerasClassifier
# dimensions of our images
img_width, img_height = 313, 220

# load the model we saved
model = load_model('hmodel.h5')
sgd = optimizers.SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy','mse'])

test_image= image.load_img('/Images/1.jpg',target_size = (img_width, img_height))
x= scipy.misc.imread('/Images/1.jpg').shape
print x
test_image = image.img_to_array(test_image)
test_image = np.expand_dims(test_image, axis = 0)
test_image = test_image.reshape(img_width, img_height,3)
result = model.predict(test_image)

print result

When I run this code i get this error:

/keras/engine/training.py", line 113, in _standardize_input_data 'with shape ' + str(data_shape)) ValueError: Error when checking : expected dense_1_input to have 2 dimensions, but got array with shape (313, 220, 3).

My first print displays: (313, 220, 3) .

How can I fix this error.

Answer 1

Your first layer Dense(150,kernel_initializer='normal', input_dim=36, activation='relu') expects an input with 2 dimensions: (*, 36) (with the first dimension corresponding to your batch size).

However your input x has actually 3 dimensions - 4 dimensions once properly batched: (*, 313, 220, 3) .

If you want a Dense layer accepting such an input, you could use the parameter input_shape=(313, 220, 3) instead of input_dim=36 .

---

Remark: You are not batching your image correctly.

test_image= image.load_img('/Images/1.jpg',target_size = (img_width, img_height))
test_image = image.img_to_array(test_image)       # shape = (313, 220, 3)
test_image = np.expand_dims(test_image, axis = 0) # shape = (1, 313, 220, 3)
# Remove this line below, as it would set back shape to (313, 220, 3)
# test_image = test_image.reshape(img_width, img_height,3)
result = model.predict(test_image)