I want to test the accuracy of my CNN model for the test-images. Following is the code for converting Ground-truth images in mha format to png format.
def save_labels(fns):
'''
INPUT list 'fns': filepaths to all labels
'''
progress.currval = 0
for label_idx in progress(xrange(len(fns))):
slices = io.imread(fns[label_idx], plugin = 'simpleitk')
for slice_idx in xrange(len(slices)):
'''
commented code in order to reshape the image slices. I tried reshaping but it did not work
strip=slices[slice_idx].reshape(1200,240)
if np.max(strip)!=0:
strip /= np.max(strip)
if np.min(strip)<=-1:
strip/=abs(np.min(strip))
'''
io.imsave('Labels2/{}_{}L.png'.format(label_idx, slice_idx), slices[slice_idx])
This code is producing 240 X 240 images in png format. However most of them are low contrast or completely blackened. Moving on, Now I pass these images to function for calculating knowing the class of labelled image.
def predict_image(self, test_img, show=False):
'''
predicts classes of input image
INPUT (1) str 'test_image': filepath to image to predict on
(2) bool 'show': True to show the results of prediction, False to return prediction
OUTPUT (1) if show == False: array of predicted pixel classes for the center 208 x 208 pixels
(2) if show == True: displays segmentation results
'''
imgs = io.imread(test_img,plugin='simpleitk').astype('float').reshape(5,240,240)
plist = []
# create patches from an entire slice
for img in imgs[:-1]:
if np.max(img) != 0:
img /= np.max(img)
p = extract_patches_2d(img, (33,33))
plist.append(p)
patches = np.array(zip(np.array(plist[0]), np.array(plist[1]), np.array(plist[2]), np.array(plist[3])))
# predict classes of each pixel based on model
full_pred = keras.utils.np_utils.probas_to_classes(self.model_comp.predict(patches))
fp1 = full_pred.reshape(208,208)
if show:
io.imshow(fp1)
plt.show
else:
return fp1
I am getting ValueError: cannot reshape array of size 172800 into shape (5,240,240). I changed 5 to 3 so that 3X240X240=172800. But then there is new problem then ValueError: Error when checking : expected convolution2d_input_1 to have 4 dimensions, but got array with shape (43264, 33, 33).
My model looks like this:
single = Sequential()
single.add(Convolution2D(self.n_filters[0], self.k_dims[0], self.k_dims[0], border_mode='valid', W_regularizer=l1l2(l1=self.w_reg, l2=self.w_reg), input_shape=(self.n_chan,33,33)))
single.add(Activation(self.activation))
single.add(BatchNormalization(mode=0, axis=1))
single.add(MaxPooling2D(pool_size=(2,2), strides=(1,1)))
single.add(Dropout(0.5))
single.add(Convolution2D(self.n_filters[1], self.k_dims[1], self.k_dims[1], activation=self.activation, border_mode='valid', W_regularizer=l1l2(l1=self.w_reg, l2=self.w_reg)))
single.add(BatchNormalization(mode=0, axis=1))
single.add(MaxPooling2D(pool_size=(2,2), strides=(1,1)))
single.add(Dropout(0.5))
single.add(Convolution2D(self.n_filters[2], self.k_dims[2], self.k_dims[2], activation=self.activation, border_mode='valid', W_regularizer=l1l2(l1=self.w_reg, l2=self.w_reg)))
single.add(BatchNormalization(mode=0, axis=1))
single.add(MaxPooling2D(pool_size=(2,2), strides=(1,1)))
single.add(Dropout(0.5))
single.add(Convolution2D(self.n_filters[3], self.k_dims[3], self.k_dims[3], activation=self.activation, border_mode='valid', W_regularizer=l1l2(l1=self.w_reg, l2=self.w_reg)))
single.add(Dropout(0.25))
single.add(Flatten())
single.add(Dense(5))
single.add(Activation('softmax'))
sgd = SGD(lr=0.001, decay=0.01, momentum=0.9)
single.compile(loss='categorical_crossentropy', optimizer='sgd')
print 'Done.'
return single
I am using keras 1.2.2. Please refer here and here( is it due to this change in full_predict in above code) for my previous post for background information. Please refer this for knowing why these specific sizes like 33,33.
You should check the shape of the patches array. This should have 4 dimensions (nrBatches, nrChannels, Width, Height). According to your error message there are only 3 dimensions. Therefore it seems like you merged your channel dimension with your batch dimension.
Related
I have a data generator that produces batches of input data (X) and targets (Y), and also a mask (batch_mask) to be applied to the model output (the same mask applies to all the datapoint in the batch; there are different masks for different batches and the data generator takes care of doing this).
As a result, the first dimension of batch_mask could have shape 1 or batch_size (by repeating the same mask along the first dimension batch_size times). I was expecting Keras to let me use either, and I wanted to simply create masks having a shape of 1 on the first dimension.
However, when I tried this, I got the error:
ValueError: Data cardinality is ambiguous:
x sizes: 128, 1
y sizes: 128
Make sure all arrays contain the same number of samples.
Why won't Keras broadcast along the first dimension? It seems like this should not be complicated.
Here's some minimal example code to observe this behavior
import tensorflow.keras as tfk
import numpy as np
#######################
# 1. model definition #
#######################
# model parameters
nfeatures_in = 6
target_size = 8
# model inputs
input = tfk.layers.Input(nfeatures_in)
input_mask = tfk.layers.Input(target_size)
# model graph
out = tfk.layers.Dense(target_size)(input)
out_masked = tfk.layers.Multiply()((out,input_mask)) # multiply all model outputs in the batch by the same mask
model = tfk.Model(inputs=(input, input_mask), outputs=out_masked)
##########################
# 2. dummy data creation #
##########################
batch_size = 32
# create masks the batch
zeros_vector = np.zeros((1,target_size)) # "batch_size"==1
zeros_vector[0,:6] = 1
batch_mask = zeros_vector
# dummy data creation
X = np.random.randn(batch_size, 6)
Y = np.random.randn(batch_size, target_size)*batch_mask # the target is masked by design in each batch
############################
# 3. compile model and fit #
############################
model.compile(optimizer="Adam", loss="mse")
model.fit((X, batch_mask),Y, batch_size=batch_size)
I know I could make this work by either:
repeating the mask to make the first dimension of batch_mask be the size of the first dimension of X (instead of 1).
using pure tensorflow (but I feel like broadcasting along the batch dimension should not be a problem for Keras).
How can I make this work with Keras?
Thank you!
You can create an IdentityLayer which receives as an external input parameter the batch_mask and returns it as a tensor.
class IdentityLayer(tfk.layers.Layer):
def __init__(self, my_mask, **kwargs):
super(IdentityLayer, self).__init__()
self.my_mask = my_mask
def call(self, _):
my_mask = tf.convert_to_tensor(self.my_mask, dtype=tf.float32)
return my_mask
def get_config(self):
config = super().get_config()
config.update({
"my_mask": self.my_mask,
})
return config
The usage of IdentityLayer in a model is straightforward:
# model inputs
input = tfk.layers.Input(nfeatures_in)
input_mask = IdentityLayer(batch_mask)(input)
# model graph
out = tfk.layers.Dense(target_size)(input)
out_masked = tfk.layers.Multiply()((out,input_mask))
model = tfk.Model(inputs=input, outputs=out_masked)
Where batch_mask is a numpy array created as you reported:
zeros_vector = np.zeros((1,target_size)) # "batch_size"==1
zeros_vector[0,:6] = 1
batch_mask = zeros_vector
The solution is to (properly) use a DataGenerator.
See the gist with the working code: https://gist.github.com/iranroman/2aaecf5b5621051df6b1b6b5394e5ef3
Thank you #Marco Cerliani for the discussion that led to figuring out the solution.
I want to make a prediction using knn and I have following lines of code:
def knn(trainImages, trainLabels, testImages, testLabels):
max = 0
for i in range(len(trainImages)):
if len(trainImages[i]) > max:
max = len(trainImages[i])
for i in range(len(trainImages)):
aux = np.array(trainImages[i])
aux.resize(max)
trainImages[i] = aux
max = 0
for i in range(len(testImages)):
if len(testImages[i]) > max:
max = len(testImages[i])
for i in range(len(testImages)):
aux = np.array(testImages[i])
aux.resize(max)
testImages[i] = aux
scaler = StandardScaler()
scaler.fit(list(trainImages))
trainImages = scaler.transform(list(trainImages))
testImages = scaler.transform(list(testImages))
classifier = KNeighborsClassifier(n_neighbors=5)
classifier.fit(trainImages, trainLabels)
pred = classifier.predict(testImages)
print(classification_report(testLabels, pred))
I got the error at testImages = scaler.transform(list(testImages)). I understand that its a difference between arrays number. How can I solve it?
scaler in scikit-learn expects input shape as (n_samples, n_features).
If your second dimension in train and test set is not equal, then not only in sklearn it is incorrect and cause to raise error, but also in theory it does not make sense. n_features dimension of test and train set should be equal, but first dimension can be different, since it show number of samples and you can have any number of samples in train and test sets.
When you execute scaler.transform(test) it expects test have the same feature numbers as where you executed scaler.fit(train). So, all your images should be in the same size.
For example, if you have 100 images, train_images shape should be something like (90,224,224,3) and test_images shape should be like (10,224,224,3) (only first dimension is different).
So, try to resize your images like this:
import cv2
resized_image = cv2.resize(image, (224,224)) #don't include channel dimension
First off, why am I using Keras? I'm trying to stay as high level as possible, which doesn't mean I'm scared of low-level Tensorflow; I just want to see how far I can go while keeping my code as simple and readable as possible.
I need my Keras model (custom-built using the Keras functional API) to read the left image from a stereo pair and minimize a loss function that needs to access both the right and left images. I want to store the data in a tf.data.Dataset.
What I tried:
Reading the dataset as (left image, right image), i.e. as tensors with shape ((W, H, 3), (W, H, 3)), then use function closure: define a keras_loss(left_images) that returns a loss(y_true, y_pred), with y_true being a tf.Tensor that holds the right image. The problem with this approach is that left_images is a tf.data.Dataset and Tensorflow complains (rightly so) that I'm trying to operate on a dataset instead of a tensor.
Reading the dataset as (left image, (left image, right image)), which should make y_true a tf.Tensor with shape ((W, H, 3), (W, H, 3)) that holds both the right and left images. The problem with this approach is that it...does not work and raises the following error:
ValueError: Error when checking model target: the list of Numpy arrays
that you are passing to your model is not the size the model expected.
Expected to see 1 array(s), for inputs ['tf_op_layer_resize/ResizeBilinear']
but instead got the following list of 2 arrays: [<tf.Tensor 'args_1:0'
shape=(None, 512, 256, 3) dtype=float32>, <tf.Tensor 'args_2:0'
shape=(None, 512, 256, 3) dtype=float32>]...
So, is there anything I did not consider? I read the documentation and found nothing about what gets considered as y_pred and what as y_true, nor about how to convert a dataset into a tensor smartly and without loading it all in memory.
My model is designed as such:
def my_model(input_shape):
width = input_shape[0]
height = input_shape[1]
inputs = tf.keras.Input(shape=input_shape)
# < a few more layers >
outputs = tf.image.resize(tf.nn.sigmoid(tf.slice(disp6, [0, 0, 0, 0], [-1, -1, -1, 2])), tf.Variable([width, height]))
model = tf.keras.Model(inputs=inputs, outputs=outputs)
return model
And my dataset is built as such (in case 2, while in case 1 only the function read_stereo_pair_from_line() changes):
def read_img_from_file(file_name):
img = tf.io.read_file(file_name)
# convert the compressed string to a 3D uint8 tensor
img = tf.image.decode_png(img, channels=3)
# Use `convert_image_dtype` to convert to floats in the [0,1] range.
img = tf.image.convert_image_dtype(img, tf.float32)
# resize the image to the desired size.
return tf.image.resize(img, [args.input_width, args.input_height])
def read_stereo_pair_from_line(line):
split_line = tf.strings.split(line, ' ')
return read_img_from_file(split_line[0]), (read_img_from_file(split_line[0]), read_img_from_file(split_line[1]))
# Dataset loading
list_ds = tf.data.TextLineDataset('test/files.txt')
images_ds = list_ds.map(lambda x: read_stereo_pair_from_line(x))
images_ds = images_ds.batch(1)
Solved. I just needed to read the dataset as (left image, [left image, right image]) instead of (left image, (left image, right image)) i.e. make the second item a list and not a tuple. I can then access the images as input_r = y_true[:, 1, :, :] and input_l = y_true[:, 0, :, :]
After training my model, I tried to plot graph of the softmax output, but it resulted in the runtime error mentioned in the title.
Here is the following code snippet:
%matplotlib inline
%config InlineBackend.figure_format = 'retina'
import helper
# Test out your network!
dataiter = iter(testloader)
images, labels = dataiter.next()
img = images[1]
# TODO: Calculate the class probabilities (softmax) for img
ps = torch.exp(model(img))
# Plot the image and probabilities
helper.view_classify(img, ps, version='Fashion')
The problem is with this part (I guess).
img = images[1]
# TODO: Calculate the class probabilities (softmax) for img
ps = torch.exp(model(img))
Problem: image you are loading is of dimension 28x28, however, the first index in input to the model is generally batch size. Since there is 1 image only, so you have to make the first dimension to be of size 1. To do that do img = img.view( (-1,) + img.shape) or img=img.unsqueeze(dim=0). Also, it seems that the first layer weight is 784 x 128. i.e the image should be converted to vector and fed to model. For that we do img=img.view(1, -1).
So, in total, you need to do
img = images[1]
img = img.unsqueeze(dim=0)
img=img.view(1, -1)
# TODO: Calculate the class probabilities (softmax) for img
ps = torch.exp(model(img))
or you can just use one command instead of two (unsqueeze is unnecessary)
img = images[1]
img=img.view(1, -1)
I am using Transfer learning for recognizing objects. I used trained VGG16 model as the base model and added my classifier on top of it using Keras. I then trained the model on my data, the model works well. I want to see the feature generated by the intermediate layers of the model for the given data. I used the following code for this purpose:
def ModeloutputAtthisLayer(model, layernme, imgnme, width, height):
layer_name = layernme
intermediate_layer_model = Model(inputs=model.input,
outputs=model.get_layer(layer_name).output)
img = image.load_img(imgnme, target_size=(width, height))
imageArray = image.img_to_array(img)
image_batch = np.expand_dims(imageArray, axis=0)
processed_image = preprocess_input(image_batch.copy())
intermediate_output = intermediate_layer_model.predict(processed_image)
print("outshape of ", layernme, "is ", intermediate_output.shape)
In the code, I used np.expand_dims to add one extra dimension for the batch as the input matrix to the network should be of the form (batchsize, height, width, channels). This code works fine. The shape of the feature vector is 1, 224, 224, 64.
Now I wish to display this as image, for this I understand there is an additional dimension added as batch so I should remove it. Following this I used the following lines of the code:
imge = np.squeeze(intermediate_output, axis=0)
plt.imshow(imge)
However it throws an error:
"Invalid dimensions for image data"
I wonder how can I display the extracted feature vector as an image. Any suggestion please.
Your feature shape is (1,224,224,64), you cannot directly plot a 64 channel image. What you can do is plot the individual channels independently like following
imge = np.squeeze(intermediate_output, axis=0)
filters = imge.shape[2]
plt.figure(1, figsize=(32, 32)) # plot image of size (32x32)
n_columns = 8
n_rows = math.ceil(filters / n_columns) + 1
for i in range(filters):
plt.subplot(n_rows, n_columns, i+1)
plt.title('Filter ' + str(i))
plt.imshow(imge[:,:,i], interpolation="nearest", cmap="gray")
This will plot 64 images in 8 rows and 8 columns.
A possible way to go consists in combining the 64 channels into a single-channel image through a weighted sum like this:
weighted_imge = np.sum(imge*weights, axis=-1)
where weights is an array with 64 weighting coefficients.
If you wish to give all the channels the same weight you could simply compute the average:
weighted_imge = np.mean(imge, axis=-1)
Demo
import numpy as np
import matplotlib.pyplot as plt
intermediate_output = np.random.randint(size=(1, 224, 224, 64),
low=0, high=2**8, dtype=np.uint8)
imge = np.squeeze(intermediate_output, axis=0)
weights = np.random.random(size=(imge.shape[-1],))
weighted_imge = np.sum(imge*weights, axis=-1)
plt.imshow(weighted_imge)
plt.colorbar()
In [33]: intermediate_output.shape
Out[33]: (1, 224, 224, 64)
In [34]: imge.shape
Out[34]: (224, 224, 64)
In [35]: weights.shape
Out[35]: (64,)
In [36]: weighted_imge.shape
Out[36]: (224, 224)