使用 TensorFlow 的二进制图像分类

Question

I am writing a code to classify between dogs and cats in python(Tensorflow) but the code is displaying this error:

IndexError: index 0 is out of bounds for axis 0 with size 0

I am stuck here. Any help is appreciated.

Also can you please help me I cant figure out here how OneHotEncoder is working .I didn't understand the logic here. I spent a lot of time on the same.

def reset_graph(seed=42):
    tf.reset_default_graph()
    tf.set_random_seed(seed)
    np.random.seed(seed)

reset_graph()
img_size = 64
num_channels = 3
img_size_flat = img_size * img_size * num_channels
img_shape = (img_size, img_size)
trainpath='C:\ProgramData\Anaconda3\train'
testpath='C:\ProgramData\Anaconda3\test'
labels = {'cats': 0, 'dogs': 1}
fc_size=32 #size of the output of final FC layer
num_steps=300
tf.logging.set_verbosity(tf.logging.INFO)

def read_images_classes(basepath,imgSize=img_size):
    image_stack = []
    label_stack = []
    for counter, l in enumerate(labels):
        path = os.path.join(basepath, l,'*g')
        for img in glob.glob(path):
            one_hot_vector = np.zeros(len(labels),dtype=np.int16)
            one_hot_vector[counter]=1
            image = cv2.imread(img)
            image_stack.append(im_resize)
            label_stack.append(labels[l])            
    return np.array(image_stack), np.array(label_stack)

X_train, y_train = read_images_classes(trainpath)
X_test, y_test = read_images_classes(testpath)b 
print('length of train image set',len(X_train))
print('X_data shape:', X_train.shape)
print('y_data shape:', y_train.shape)

fig1 = plt.figure() 
ax1 = fig1.add_subplot(2,2,1) 
img = cv2.resize(X_train[0],(64,64), interpolation=cv2.INTER_CUBIC)
ax1.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
plt.title(y_train[0])
plt.show()

Answer 1

IndexError: index 0 is out of bounds for axis 0 with size 0

It means you don't have the index you are trying to reference.

Since this is a binary classification problem, you don't required one_hot encoding for pre-processing labels. if you have more than two labels then you can use one_hot encoding.

Please refer binary classification code using Tensorflow for Cats and Dogs Dataset

import os
import numpy as np
from keras import layers
import pandas as pd
from tensorflow.keras.layers import Input, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D
from tensorflow.keras.layers import AveragePooling2D, MaxPooling2D, Dropout, GlobalMaxPooling2D, GlobalAveragePooling2D
from tensorflow.keras.models import Sequential
from tensorflow.keras import regularizers, optimizers
from tensorflow.keras.preprocessing import image
from tensorflow.keras.preprocessing.image import ImageDataGenerator

import keras.backend as K

import keras.backend as K
K.set_image_data_format('channels_last')

from google.colab import drive
drive.mount('/content/drive')

train_dir  = '/content/drive/My Drive/Dogs_Vs_Cats/train'
test_dir = '/content/drive/My Drive/Dogs_Vs_Cats/test'

img_width, img_height = 300, 281
input_shape = img_width, img_height, 3

train_samples = 2000
test_samples = 1000
epochs = 30
batch_size = 32

train_datagen = ImageDataGenerator(
    rescale = 1. /255,
    shear_range = 0.2,
    zoom_range = 0.2,
    horizontal_flip = True)

test_datagen = ImageDataGenerator(
    rescale = 1. /255)

train_data = train_datagen.flow_from_directory(
    train_dir,
    target_size = (img_width, img_height),
    batch_size = batch_size,
    class_mode = 'binary')

test_data = test_datagen.flow_from_directory(
    test_dir,
    target_size = (img_width, img_height),
    batch_size = batch_size,
    class_mode = 'binary')


model = Sequential()

model.add(Conv2D(32, (7, 7), strides = (1, 1), input_shape = input_shape))
model.add(BatchNormalization(axis = 3))
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2)))

model.add(Conv2D(64, (7, 7), strides = (1, 1)))
model.add(BatchNormalization(axis = 3))
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2)))

model.add(Flatten())
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))

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

model.build(input_shape)
model.summary() 


model.fit_generator(
        train_data,
        steps_per_epoch = train_samples//batch_size,
        epochs = epochs,
        validation_data = test_data,
        verbose = 1,
        validation_steps = test_samples//batch_size)