takahish

from keras.wrappers.scikit_learn import KerasClassifier
from sklearn.metrics import f1_score, make_scorer
from sklearn.model_selection import GridSearchCV
from keras.models import Sequential
from keras.layers import GlobalAveragePooling2D
from keras.layers import Dense
from keras.optimizers import RMSprop

def build_model(learning_rate):
    Resnet50_model = Sequential()

takahish

from keras import backend as K

def recall(y_true, y_pred):
    """Recall for each labels.
        
    Args:
        y_ture: 1d array-like, or label indicator array.
        y_pred: 1d array-like, or label indicator array.

    Return:

takahish

import cv2
import numpy as np
from matplotlib import pyplot as plt

def intensity_hist(image_file):
    """show some descriptive stats of some of the most relevant aspects in the dataset.
    Such as the min, max, median, mean of the image sizes.
    
    Args:
        image_files: string. Path of image file.

takahish

import cv2
import pandas as pd

def stats(image_files):
    """show some descriptive stats of some of the most relevant aspects in the dataset.
    Such as the min, max, median, mean of the image sizes.
    
    Args:
        image_files: sequence of string. Paths of image files.

takahish

import matplotlib.pyplot as plt
import seaborn as sns

%matplotlib inline

plt.figure(figsize=(24, 4))
plt.subplots_adjust(hspace=0.4)

ax1 = plt.subplot(1, 1, 1)
sns.countplot(x=(np.where(train_targets == 1))[1], ax=ax1)

takahish

def Resnet50_predict_breed(img_path):
    """Return the dog breed that is predicted by the transer learing model.

    Args:
        img_path: string. a file path to a color images.

    Returns:
        dog breed: string, and probability: float.
    """
    bottleneck_feature = extract_Resnet50(path_to_tensor(img_path))

takahish

Resnet50_model = Sequential()
Resnet50_model.add(GlobalAveragePooling2D(input_shape=train_Resnet50.shape[1:]))
Resnet50_model.add(Dense(133, activation='softmax'))

takahish

train_Resnet50 = extract_Resnet50(paths_to_tensor(train_files).astype('float32'))
valid_Resnet50 = extract_Resnet50(paths_to_tensor(valid_files).astype('float32'))
test_Resnet50 = extract_Resnet50(paths_to_tensor(test_files).astype('float32'))

takahish

from keras.applications.resnet50 import ResNet50, preprocess_input

def extract_Resnet50(tensor):
    """Extracting bottleneck features of ResNet-50 to use as input our classifying model.
    
    Args:
        tensor: numpy.array. a 4D tensor suitable for supplying to a Keras CNN.
    
    Retruns:
        numpy.array, bottleneck features of ResNet-50

takahish

model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=[f1])