I want to test the trained model with a single image(rgb). But I am encountering an error. I used cat and dog images while training the model.Also, while creating the model, I got the first layers from resnet50. I created the last layers myself. Before exporting the dataset to the model, I did some preliminary work and converted the classes to 0-1 values.(with encoder cat:0,dog:1) Now I want to test this model with a dog image. I expect it to return 0 or 1, but I have a problem.
my code blok:
*from keras.models import load_model
from keras.preprocessing import image
import numpy as np
from PIL import Image
import numpy as np
from skimage import transform
# dimensions of our images -----
img_width, img_height = 750, 422
# load the model we saved
model = load_model('.../Resnet50_Save_model.hdf5')
test_image = image.load_img('.../5c02ed550f25442260cff6ab.jpg', target_size=(img_width, img_height))
test_image = image.img_to_array(test_image)
test_image = np.expand_dims(test_image, axis=0)
test_image = Image.open('.../5c02ed550f25442260cff6ab.jpg')
test_image = test_image.resize((750,422))
test_image = test_image / 255.0
test_image = test_image.reshape(224,224)
result = model.predict(test_image, batch_size=1)
print(result)*
Error:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-89-cd9abec3a0ce> in <module>()
23 # test_image = test_image.reshape(224,224)
24
---> 25 result = model.predict(test_image, batch_size=1)
26 print(result)
9 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
975 except Exception as e: # pylint:disable=broad-except
976 if hasattr(e, "ag_error_metadata"):
--> 977 raise e.ag_error_metadata.to_exception(e)
978 else:
979 raise
ValueError: in user code:
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1478 predict_function *
return step_function(self, iterator)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1468 step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1259 run
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2730 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:3417 _call_for_each_replica
return fn(*args, **kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1461 run_step **
outputs = model.predict_step(data)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1434 predict_step
return self(x, training=False)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:998 __call__
input_spec.assert_input_compatibility(self.input_spec, inputs, self.name)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/input_spec.py:274 assert_input_compatibility
', found shape=' + display_shape(x.shape))
ValueError: Input 0 is incompatible with layer sequential: expected shape=(None, 224, 224, 3), found shape=(1, 750, 3)
- List item
The error is telling you that the shape of your input (1, 750, 3)
doesn't match the expected shape by your model (None, 224, 224, 3).
I recommend you resize your image to 224 x 224 first, then normalize it using the division by 255. After that expand the dimensions so it becomes (1, 224, 224, 3) and try again.
There are two image loading operations, you can select one of them.
from keras.models import load_model
from keras.preprocessing import image
# load the model we saved
test_image = image.load_img('.../5c02ed550f25442260cff6ab.jpg', target_size=(224, 224))
test_image = image.img_to_array(test_image) / 255.0
model = load_model('.../Resnet50_Save_model.hdf5')
result = model.predict(test_image, batch_size=1)
print(result)
Related
I am trying to train a 1D-CNN on tabular data to then use LRP on it (as it has been done here). I am stuck at the model.fit() part of the implementation. Up to that point everything seems to work.
Here is the error I get:
model_history = model.fit(X_train_smote, y_train_smote, batch_size=100, epochs=100, validation_split = 0.2)
Epoch 1/100
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
Input In [60], in <cell line: 1>()
----> 1 model_history = model.fit(X_train_smote, y_train_smote, batch_size=100, epochs=100, validation_split = 0.2)
File D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\utils\traceback_utils.py:67, in filter_traceback.<locals>.error_handler(*args, **kwargs)
65 except Exception as e: # pylint: disable=broad-except
66 filtered_tb = _process_traceback_frames(e.__traceback__)
---> 67 raise e.with_traceback(filtered_tb) from None
68 finally:
69 del filtered_tb
File ~\AppData\Local\Temp\__autograph_generated_filelswqetod.py:15, in outer_factory.<locals>.inner_factory.<locals>.tf__train_function(iterator)
13 try:
14 do_return = True
---> 15 retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
16 except:
17 do_return = False
ValueError: in user code:
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\training.py", line 1051, in train_function *
return step_function(self, iterator)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\training.py", line 1040, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\training.py", line 1030, in run_step **
outputs = model.train_step(data)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\training.py", line 890, in train_step
loss = self.compute_loss(x, y, y_pred, sample_weight)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\training.py", line 948, in compute_loss
return self.compiled_loss(
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\engine\compile_utils.py", line 201, in __call__
loss_value = loss_obj(y_t, y_p, sample_weight=sw)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\losses.py", line 139, in __call__
losses = call_fn(y_true, y_pred)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\losses.py", line 243, in call **
return ag_fn(y_true, y_pred, **self._fn_kwargs)
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\losses.py", line 1787, in categorical_crossentropy
return backend.categorical_crossentropy(
File "D:\Programme\Anaconda\envs\LRP_innvestigate\lib\site-packages\keras\backend.py", line 5119, in categorical_crossentropy
target.shape.assert_is_compatible_with(output.shape)
ValueError: Shapes (None,) and (None, 6, 2) are incompatible
Yet, for the sake of clarity here is the rest of my current implementation (adapted from a Kaggle notebook). The model structure is based on the feedback of my previous question here on SO.
import matplotlib.pyplot as plt
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns
from matplotlib.colors import ListedColormap
from sklearn.metrics import plot_confusion_matrix
from scipy.stats import norm, boxcox
from sklearn.metrics import confusion_matrix, classification_report, accuracy_score
from collections import Counter
from scipy import stats
import tensorflow as tf
import warnings
warnings.simplefilter(action='ignore', category=Warning)
dataset = pd.read_csv('F:/Programmieren/this_one/data/Churn_Modelling.csv')
# split into variables and target
X = dataset.iloc[:, 3:-1].values
y = dataset.iloc[:, -1].values
# here gender is encoded
from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
X[:, 2] = le.fit_transform(X[:, 2])
# one hot encoding the country (as explained in Ullah et al.)
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import OneHotEncoder
ct = ColumnTransformer(transformers=[('encoder', OneHotEncoder(), [1])], remainder='passthrough')
X = np.array(ct.fit_transform(X))
# split the data
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size= .2)
# upsample minority class with SMOTE
import imblearn
from imblearn.over_sampling import SMOTENC
#get original class distribution
counter = Counter(y)
print(counter)
#SMOTENC is used instead of SMOTE because there are multiple categorical features in the dataset
oversample = SMOTENC(categorical_features=[0, 1, 2, 4, 9, 10])
X_train_smote, y_train_smote = oversample.fit_resample(X_train, y_train)
#get new class distribution
counter = Counter(y_train_smote)
print(counter)
# normalize values to range 0-1 as explained in Ullah et al.
from sklearn.preprocessing import MinMaxScaler
mms = MinMaxScaler()
X_train_smote = mms.fit_transform(X_train_smote)
X_test = mms.transform(X_test)
#record-wise normalization for relative value comparison as stated by one of the authors I was in contact with
from sklearn.preprocessing import normalize
X_train_smote = normalize(X_train_smote, axis=1, norm='l1')
X_test = normalize(X_test, axis=1, norm='l1')
#reshape data for CNN
sample_size = X_train_smote.shape[0] # number of samples in train set
time_steps = X_train_smote.shape[1] # number of features in train set
input_dimension = 1
train_data_reshaped = X_train_smote.reshape(sample_size,time_steps,input_dimension)
print("After reshape train data set shape:\n", train_data_reshaped.shape)
print("1 Sample shape:\n",train_data_reshaped[0].shape)
print("An example sample:\n", train_data_reshaped[0])
#reshape test data as well
test_data_reshaped = X_test.reshape(X_test.shape[0],X_test.shape[1],1)
import keras
from keras.models import Sequential
from keras.layers import Dense, Conv1D
#create model as explained in the paper
model = Sequential()
model.add(Conv1D(filters=25, kernel_size=3, activation='relu', input_shape=(12,1)))
model.add(Conv1D(50, 3))
model.add(Conv1D(100, 3))
model.add(Dense(2200, activation='relu'))
model.add(Dense(2, activation='relu'))
model.add(Dense(2, activation='softmax'))
model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])
model.summary()
#output of model.summary()
Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv1d (Conv1D) (None, 10, 25) 100
conv1d_1 (Conv1D) (None, 8, 50) 3800
conv1d_2 (Conv1D) (None, 6, 100) 15100
dense (Dense) (None, 6, 2200) 222200
dense_1 (Dense) (None, 6, 2) 4402
dense_2 (Dense) (None, 6, 2) 6
=================================================================
Total params: 245,608
Trainable params: 245,608
Non-trainable params: 0
_________________________________________________________________
Are there any major flaws in my current approach that I am not aware of?
You should add a Flatten() layer before the output layer as the output shape is (6,2) and not (2,).
#Add Flatten layer
model.add(Flatten())
model.add(Dense(2, activation='softmax'))
Also kindly change the loss to sparse_categorical_crossentopy because you are using integers(0,1) as the labels.
#change loss to sparse_categorical_crossentropy
model.compile(optimizer = 'adam', loss = 'sparse_categorical_crossentropy', metrics = ['accuracy'])
Please refer to the gist for working code. Thank you!
Since tflearn is outdated and I am watching a chatbot tutorial that uses tflearn, I want to write the neural network model in keras. However, I got this error right here:
WARNING:tensorflow:Model was constructed with shape (None, 58) for input KerasTensor(type_spec=TensorSpec(shape=(None, 58), dtype=tf.float32, name='input_22'), name='input_22', description="created by layer 'input_22'"), but it was called on an input with incompatible shape (None,).
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-164-1a494613d0d2> in <module>()
1 convert_input("Hello")
----> 2 chat()
2 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in autograph_handler(*args, **kwargs)
1145 except Exception as e: # pylint:disable=broad-except
1146 if hasattr(e, "ag_error_metadata"):
-> 1147 raise e.ag_error_metadata.to_exception(e)
1148 else:
1149 raise
ValueError: in user code:
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1801, in predict_function *
return step_function(self, iterator)
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1790, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1783, in run_step **
outputs = model.predict_step(data)
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1751, in predict_step
return self(x, training=False)
File "/usr/local/lib/python3.7/dist-packages/keras/utils/traceback_utils.py", line 67, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/usr/local/lib/python3.7/dist-packages/keras/engine/input_spec.py", line 228, in assert_input_compatibility
raise ValueError(f'Input {input_index} of layer "{layer_name}" '
ValueError: Exception encountered when calling layer "sequential_24" (type Sequential).
Input 0 of layer "dense_59" is incompatible with the layer: expected min_ndim=2, found ndim=1. Full shape received: (None,)
Call arguments received:
• inputs=('tf.Tensor(shape=(None,), dtype=int64)',)
• training=False
• mask=None
I have tried to set up the keras model myself
model = keras.Sequential()
# model.add(keras.layers.InputLayer(input_shape = len(words)))
model.add(keras.Input(shape=len(training[0])))
model.add(keras.layers.Dense(8, activation = "relu"))
model.add(keras.layers.Dense(8, activation = "relu"))
model.add(keras.layers.Dense(len(output[0]), activation= "softmax"))
model.compile(optimizer = "adam", loss = "categorical_crossentropy", metrics=['accuracy'])
# convert_input(inp) -> return a 1D numpy array filled with 1 and 0
prediction = model.predict([convert_input(inp)])
versus the tflearn model
network = tflearn.input_data(shape=[None, len(training[0])])
network = tflearn.fully_connected(network, 8)
network = tflearn.fully_connected(network, 8)
network = tflearn.fully_connected(network, len(output[0]), activation = "softmax")
network = tflearn.regression(network)
model = tflearn.DNN(network)
model.fit(training, output, n_epoch = 1000, batch_size=8, show_metric=True)
# convert_input(inp) -> return a 1D numpy array filled with 1 and 0
prediction = model.predict([convert_input(inp)])
However, when I call model.predict only the tflearn model works and not the keras. Please help!
I was in a similar predicament. However I was finally able to resolve my problem with the following code:
import numpy as np
from tensorflow.keras import Sequential
from tensorflow.keras.layers import Dense
model=Sequential()
model.add(Dense(8,input_shape=(len(train_x[0]),),kernel_initializer='normal'))
model.add(Dense(8,kernel_initializer='normal'))
model.add(Dense(len(train_y[0]),activation='softmax',kernel_initializer='normal'))
model.compile(optimizer='adam', loss='mean_squared_error', metrics=['accuracy'])
model.fit(np.array(train_x),np.array(train_y), epochs=1000, batch_size=8, verbose=1)
During prediction,
model.predict(np.array([p]))
Basically converting the list into an array using np.array() helped me solve the problem.
I'm trying to use tf.data.Dataset.list_files to load .tiff images and infer their labels from their names.
I use the following code but stumbled upon a strange issue, as described bellow:
import os
import datetime as dt
import numpy as np
import pathlib
from pathlib import Path
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import cv2
DATA_PATH = Path('PATH TO DATA')
BATCH_SIZE = 32
INPUT_IMAGE_SHAPE = (128, 128, 1)
CROP_SHAPE = INPUT_IMAGE_SHAPE
CENTRAL_CROP_PROP = .7
BRIGHTNESS_DELTA = 0.1
CONTRAST = (0.4, 0.6)
class ConvModel(keras.Model):
def __init__(self, input_shape):
super().__init__()
self.input_image_shape = input_shape
self.model = keras.Sequential([
layers.Input(shape=input_shape),
layers.Conv2D(32, 3),
layers.BatchNormalization(),
layers.ReLU(),
layers.MaxPool2D(),
layers.Conv2D(64, 5),
layers.BatchNormalization(),
layers.ReLU(),
layers.MaxPool2D(),
layers.Conv2D(128, 3, kernel_regularizer=keras.regularizers.l2(0.01)),
layers.BatchNormalization(),
layers.ReLU(),
layers.Flatten(),
layers.Dense(64, activation='relu', kernel_regularizer=keras.regularizers.l2(0.01)),
layers.Dropout(0.5),
layers.Dense(10)
])
def call(self, inputs):
return self.model(inputs)
def preprocessing_func(image):
img = tf.image.central_crop(image, CENTRAL_CROP_PROP)
if img.shape[2] == 3:
img = tf.image.rgb_to_grayscale(img)
return img
def augment(image):
img = tf.image.random_crop(image, size=CROP_SHAPE) # Slices a shape size portion out of value at a uniformly chosen offset. Requires value.shape >= size.
img = tf.image.random_brightness(img, max_delta=BRIGHTNESS_DELTA) # Equivalent to adjust_brightness() using a delta randomly picked in the interval [-max_delta, max_delta)
img = tf.image.random_contrast(img, lower=CONTRAST[0], upper=CONTRAST[1]) # Equivalent to adjust_contrast() but uses a contrast_factor randomly picked in the interval [lower, upper).
img = tf.image.random_flip_left_right(img)
img = tf.image.random_flip_up_down(img)
return img
def load_image(image_file):
# 1) Decode the path
image_file = image_file.decode('utf-8')
# 2) Read the image
img = cv2.imread(image_file)
if len(img.shape) < 3:
img = np.expand_dims(img, axis=-1)
img = preprocessing_func(image=img)
img = augment(img)
img = tf.cast(img, tf.float32)
img.set_shape(INPUT_IMAGE_SHAPE)
# 3) Get the label
label = tf.strings.split(image_file, "\\")[-1]
label = tf.strings.substr(label, pos=0, len=1)
label = tf.strings.to_number(label, out_type=tf.float32)
label = tf.cast(label, tf.float32)
label.set_shape([])
return img, label
def _fixup_shape(images, labels):
images.set_shape(INPUT_IMAGE_SHAPE)
labels.set_shape([])
return images, labels
if __name__=='__main__':
train_ds = tf.data.Dataset.list_files(str(DATA_PATH / '*.tiff'))
train_ds = train_ds.map(lambda x: tf.numpy_function(load_image, [x], (tf.float32, tf.float32)))
# train_ds = train_ds.map(_fixup_shape)
train_ds = train_ds.batch(BATCH_SIZE)
train_ds = train_ds.shuffle(buffer_size=1000)
train_ds = train_ds.prefetch(tf.data.AUTOTUNE)
train_ds = train_ds.repeat()
model = ConvModel(input_shape=INPUT_IMAGE_SHAPE)
model.compile(
loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),
optimizer=keras.optimizers.Adam(learning_rate=3e-4),
metrics=['accuracy']
)
train_log_dir = f'./logs/{dt.datetime.now().strftime("%Y-%m-%d_%H-%M-%S")}/train_data'
callbacks = [
keras.callbacks.TensorBoard(
log_dir=train_log_dir,
write_images=True
)
]
model.fit(
train_ds,
batch_size=32,
steps_per_epoch=10,
epochs=10,
callbacks=callbacks
)
While I try to run it it throws up an error :
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-213-b1f3d317135b> in <module>
----> 1 model.fit(
2 train_ds,
3 batch_size=32,
4 steps_per_epoch=10,
5 epochs=10,
~\anaconda3\lib\site-packages\keras\utils\traceback_utils.py in error_handler(*args, **kwargs)
65 except Exception as e: # pylint: disable=broad-except
66 filtered_tb = _process_traceback_frames(e.__traceback__)
---> 67 raise e.with_traceback(filtered_tb) from None
68 finally:
69 del filtered_tb
~\anaconda3\lib\site-packages\tensorflow\python\framework\func_graph.py in autograph_handler(*args, **kwargs)
1127 except Exception as e: # pylint:disable=broad-except
1128 if hasattr(e, "ag_error_metadata"):
-> 1129 raise e.ag_error_metadata.to_exception(e)
1130 else:
1131 raise
ValueError: in user code:
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\training.py", line 878, in train_function *
return step_function(self, iterator)
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\training.py", line 867, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\training.py", line 860, in run_step **
outputs = model.train_step(data)
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\training.py", line 817, in train_step
self.compiled_metrics.update_state(y, y_pred, sample_weight)
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\compile_utils.py", line 439, in update_state
self.build(y_pred, y_true)
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\compile_utils.py", line 359, in build
self._metrics = tf.__internal__.nest.map_structure_up_to(y_pred, self._get_metric_objects,
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\compile_utils.py", line 485, in _get_metric_objects
return [self._get_metric_object(m, y_t, y_p) for m in metrics]
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\compile_utils.py", line 485, in <listcomp>
return [self._get_metric_object(m, y_t, y_p) for m in metrics]
File "C:\Users\mchls\anaconda3\lib\site-packages\keras\engine\compile_utils.py", line 506, in _get_metric_object
y_t_rank = len(y_t.shape.as_list())
ValueError: as_list() is not defined on an unknown TensorShape.
though manually running X.shape.as_list() and y.shape.as_list() works, as shown below:
X, y = next(iter(train_ds))
X.shape.as_list(), y.shape.as_list()
[OUT] ([16, 128, 128, 1], [16])
This issue is fixed, as described in this GitHub thread and in this answer, by manually mapping the following function on the dataset by train_ds = train_ds.map(_fixup_shape).batch(BATCH_SIZE):
def _fixup_shape(images, labels):
images.set_shape([128, 128, 1])
labels.set_shape([]) # I have 19 classes
# weights.set_shape([None])
return images, labels
if __name__=='__main__':
train_ds = tf.data.Dataset.list_files(str(DATA_PATH / '*.tiff'))
train_ds = train_ds.map(lambda x: tf.numpy_function(load_image, [x], (tf.float32, tf.float32)))
train_ds = train_ds.map(_fixup_shape)
train_ds = train_ds.batch(BATCH_SIZE)
train_ds = train_ds.shuffle(buffer_size=1000)
train_ds = train_ds.prefetch(tf.data.AUTOTUNE)
train_ds = train_ds.repeat()
but I can't figure out why is the _fixup_shape is needed in the first place, as I do state the shape inside the load_image function.
Is it a bug in TF 2.6.1 or is it an expected behavior?
Thanks
I am trying to do semantic segmentation with two classes.
I have 224x224x3 images and 224x224 binary segmentation masks. I am reshaping the masks to be 224x224x1 (I read somewhere that this is the format that I should pass to the model).
When I try to loop through the train data loader it either runs without errors or I get the following error (randomly):
Traceback (most recent call last):
File "/Users/nikolaykolibarov/Desktop/GATE/rooftop-recognition/rooftop-edge-segmentation-ml/main.py", line 43, in <module>
for i, sample in enumerate(train_loader):
File "/Users/nikolaykolibarov/opt/anaconda3/envs/dtcc/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 346, in __next__
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/Users/nikolaykolibarov/opt/anaconda3/envs/dtcc/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 47, in fetch
return self.collate_fn(data)
File "/Users/nikolaykolibarov/opt/anaconda3/envs/dtcc/lib/python3.7/site-packages/torch/utils/data/_utils/collate.py", line 79, in default_collate
return [default_collate(samples) for samples in transposed]
File "/Users/nikolaykolibarov/opt/anaconda3/envs/dtcc/lib/python3.7/site-packages/torch/utils/data/_utils/collate.py", line 79, in <listcomp>
return [default_collate(samples) for samples in transposed]
File "/Users/nikolaykolibarov/opt/anaconda3/envs/dtcc/lib/python3.7/site-packages/torch/utils/data/_utils/collate.py", line 55, in default_collate
return torch.stack(batch, 0, out=out)
RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 0. Got 224 and 244 in dimension 2 at /tmp/pip-req-build-9oilk29k/aten/src/TH/generic/THTensor.cpp:689
Most of the issues with the same exception I found online don’t have matching numbers like I do (Got 224 and 224). I don’t understand, this exception should occur when they aren’t equal (if I am not wrong).
When I run it, occasionally I get this error a few times, and then it doesn’t occur a few times.
I am not sure if I messed something up with the data types and shapes or it is something else.
Here is the code:
roof_edges_dataset.py
import os
import cv2 as cv
from torch.utils.data import Dataset
from torchvision.transforms import transforms
from utils import create_binary_mask, get_labelme_shapes, plot_segmentation_dataset
class RoofEdgesDataset(Dataset):
def __init__(self, im_path, ann_path, transform=None):
self.im_path = im_path
self.ann_path = ann_path
self.transform = transform
self.im_fn_list = sorted(os.listdir(im_path), key=lambda x: int(x.split('.')[0]))
self.ann_fn_list = sorted(os.listdir(ann_path), key=lambda x: int(x.split('.')[0]))
def __len__(self):
return len(self.im_fn_list)
def __getitem__(self, index):
im_path = os.path.join(self.im_path, self.im_fn_list[index])
im = cv.imread(im_path)
ann_path = os.path.join(self.ann_path, self.ann_fn_list[index])
ann = create_binary_mask(im, get_labelme_shapes(ann_path))
ann = ann.reshape(ann.shape[0], ann.shape[1], 1)
ann = transforms.ToTensor()(ann)
if self.transform:
im = self.transform(im)
return im, ann
main.py:
import torch
import torchvision.transforms as transforms
from torch.utils.data import DataLoader
from roof_edges_dataset import RoofEdgesDataset
# Device
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# Hyperparameters
in_im_shape = (3, 224, 224)
num_classes = 2 # Edge / Non-edge
learning_rate = 0.001
batch_size = 4
n_epochs = 10
# Data - 60% Train - 20% Val - 20% Test
transformations = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
dataset = RoofEdgesDataset(im_path='data/images', ann_path='data/annotations', transform=transformations)
train_size = int(0.8 * len(dataset))
test_size = len(dataset) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(dataset, [train_size, test_size])
train_size = int(0.8 * len(train_dataset))
val_size = len(train_dataset) - train_size
train_dataset, val_dataset = torch.utils.data.random_split(train_dataset, [train_size, val_size])
train_loader = DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True)
val_loader = DataLoader(dataset=val_dataset, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(dataset=test_dataset, batch_size=batch_size, shuffle=True)
# Model
# Loss and Optimizer
# Train
for epoch in range(n_epochs):
for batch_idx, (image, annotation) in enumerate(train_loader):
image = image.to(device=device)
annotation = annotation.to(device=device)
print(image.shape)
print(annotation.shape)
break
# Evaluate
I would be also very thankful if you let me know if something is done in a wrong or stupid way and how to do it better.
Thanks in advance!
It seems like your training images have different sizes, e.g. one is 224x224 while another is 244x244.
Add a crop/resize transformation to all images so they can be concatenated together to form a batch.
I am trying to follow the "Fine-tune InceptionV3 on a new set of classes" sample code to freeze the first 172 layers and re-train the last layers on cats/dogs dataset. I keep getting an error which I have noted at the bottom. Please help. I am using Ubuntu 16.04, keras 1.2.1, theano 0.9.0beta1.dev, numpy 1.12.0 and python 3.5.
from PIL import Image
import os
import matplotlib.pyplot as plt
import numpy as np
data_root_dir = "/home/ubuntu/ML/data/dogscats/"
train_dir = os.path.join(data_root_dir,"sample", "train")
valid_dir = os.path.join(data_root_dir, "valid")
from keras.applications.inception_v3 import InceptionV3
from keras.preprocessing import image
from keras.models import Model
from keras.layers import Dense, GlobalAveragePooling2D
from keras import backend as K
# create the base pre-trained model
base_model = InceptionV3(weights='imagenet', include_top=True)
# add a global spatial average pooling layer
x = base_model.output
#x = GlobalAveragePooling2D()(x)
# let's add a fully-connected layer
x = Dense(1024, activation='relu')(x)
# and a logistic layer -- let's say we have 200 classes
predictions = Dense(2, activation='softmax')(x)
# this is the model we will train
model = Model(input=base_model.input, output=predictions)
for layer in model.layers[:172]:
layer.trainable = False
for layer in model.layers[172:]:
layer.trainable = True
from keras.optimizers import SGD
model.compile(optimizer=SGD(lr=0.0001, momentum=0.9), loss='categorical_crossentropy')
from sklearn.preprocessing import OneHotEncoder
def get_data(path, target_size=(299,299)):
batches = get_batches(path, shuffle=False, batch_size=1, class_mode=None, target_size=target_size)
return np.concatenate([batches.next() for i in range(batches.nb_sample)])
def get_batches(dirname, gen=image.ImageDataGenerator(), shuffle=True, batch_size=2, class_mode='categorical',
target_size=(299,299)):
return gen.flow_from_directory(dirname, target_size=target_size,
class_mode=class_mode, shuffle=shuffle, batch_size=batch_size)
def onehot(x): return np.array(OneHotEncoder().fit_transform(x.reshape(-1,1)).todense())
# Use batch size of 1 since we're just doing preprocessing on the CPU
val_batches = get_batches(valid_dir, shuffle=False, batch_size=10)
train_batches = get_batches(train_dir, shuffle=False, batch_size=10)
val_classes = val_batches.classes
trn_classes = train_batches.classes
val_labels = onehot(val_classes)
trn_labels = onehot(trn_classes)
model.fit_generator(train_batches, samples_per_epoch=train_batches.n, nb_epoch=10,
validation_data=val_batches, nb_val_samples=val_batches.n)
The exception is: padding must be zero for average_exc_pad
Here is the full stack-trace:
ValueError Traceback (most recent call last)
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/theano/compile/function_module.py in __call__(self, *args, **kwargs)
883 outputs =\
--> 884 self.fn() if output_subset is None else\
885 self.fn(output_subset=output_subset)
ValueError: padding must be zero for average_exc_pad
During handling of the above exception, another exception occurred:
ValueError Traceback (most recent call last)
<ipython-input-4-369d7760ec6e> in <module>()
34
35 model.fit_generator(train_batches, samples_per_epoch=train_batches.n, nb_epoch=10,
---> 36 validation_data=val_batches, nb_val_samples=val_batches.n)
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/keras/engine/training.py in fit_generator(self, generator, samples_per_epoch, nb_epoch, verbose, callbacks, validation_data, nb_val_samples, class_weight, max_q_size, nb_worker, pickle_safe, initial_epoch)
1551 outs = self.train_on_batch(x, y,
1552 sample_weight=sample_weight,
-> 1553 class_weight=class_weight)
1554
1555 if not isinstance(outs, list):
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/keras/engine/training.py in train_on_batch(self, x, y, sample_weight, class_weight)
1314 ins = x + y + sample_weights
1315 self._make_train_function()
-> 1316 outputs = self.train_function(ins)
1317 if len(outputs) == 1:
1318 return outputs[0]
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/keras/backend/theano_backend.py in __call__(self, inputs)
957 def __call__(self, inputs):
958 assert isinstance(inputs, (list, tuple))
--> 959 return self.function(*inputs)
960
961
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/theano/compile/function_module.py in __call__(self, *args, **kwargs)
896 node=self.fn.nodes[self.fn.position_of_error],
897 thunk=thunk,
--> 898 storage_map=getattr(self.fn, 'storage_map', None))
899 else:
900 # old-style linkers raise their own exceptions
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/theano/gof/link.py in raise_with_op(node, thunk, exc_info, storage_map)
323 # extra long error message in that case.
324 pass
--> 325 reraise(exc_type, exc_value, exc_trace)
326
327
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/six.py in reraise(tp, value, tb)
683 value = tp()
684 if value.__traceback__ is not tb:
--> 685 raise value.with_traceback(tb)
686 raise value
687
/home/ubuntu/anaconda3/envs/tensorflow/lib/python3.5/site-packages/theano/compile/function_module.py in __call__(self, *args, **kwargs)
882 try:
883 outputs =\
--> 884 self.fn() if output_subset is None else\
885 self.fn(output_subset=output_subset)
886 except Exception:
ValueError: padding must be zero for average_exc_pad
Apply node that caused the error: AveragePoolGrad{ignore_border=True, mode='average_exc_pad', ndim=2}(Join.0, IncSubtensor{InplaceInc;::, ::, :int64:, :int64:}.0, TensorConstant{(2,) of 3}, TensorConstant{(2,) of 1}, TensorConstant{(2,) of 1})
Toposort index: 5270
Inputs types: [TensorType(float32, 4D), TensorType(float32, 4D), TensorType(int64, vector), TensorType(int64, vector), TensorType(int64, vector)]
Inputs shapes: [(10, 2048, 8, 8), (10, 2048, 8, 8), (2,), (2,), (2,)]
Inputs strides: [(524288, 256, 32, 4), (524288, 256, 32, 4), (8,), (8,), (8,)]
Inputs values: ['not shown', 'not shown', array([3, 3]), array([1, 1]), array([1, 1])]
Outputs clients: [[Elemwise{add,no_inplace}(CorrMM_gradInputs{half, (1, 1), (1, 1)}.0, CorrMM_gradInputs{half, (1, 1), (1, 1)}.0, CorrMM_gradInputs{half, (1, 1), (1, 1)}.0, AveragePoolGrad{ignore_border=True, mode='average_exc_pad', ndim=2}.0)]]
Fine-tuning in that situation possibly means using the convolutional layers as pre-trained feature extractors. So you don't really want the top layers (densely connected layers) of the Inception network.
Changing
base_model = InceptionV3(weights='imagenet', include_top=True)
to
base_model = InceptionV3(weights='imagenet', include_top=False)
should work.
Also, if you have 200 classes you should change
# and a logistic layer -- let's say we have 200 classes
predictions = Dense(2, activation='softmax')(x)
to
predictions = Dense(200, activation='softmax')(x)
So your last layer will have the desired 200 elements.