load_model() doesn't load a model - function

Question

I have Python version 3.6 and tensorflow version 2.0.1 installed. I created a model and tried to load it, but ran into a problem:
Traceback (most recent call last):
File "C:/Users/Irina/Documents/PYTHON4/main.py", line 86, in <module>
main()
File "C:/Users/Irina/Documents/PYTHON4/main.py", line 73, in main
predictions = get_predictions(data, model)
File "C:/Users/Irina/Documents/PYTHON4/main.py", line 59, in get_predictions
predictions = [predict(np.array([image]), model) for image in data]
File "C:/Users/Irina/Documents/PYTHON4/main.py", line 59, in <listcomp>
predictions = [predict(np.array([image]), model) for image in data]
File "C:\Users\Irina\Documents\PYTHON4\eval.py", line 53, in predict
pred = model.predict(data)
AttributeError: 'list' object has no attribute 'predict'
Process finished with exit code 1
I suspect that the error is in the non-working load_model () function, although it may be due to the incompatibility of Python and tensorflow.
It's main.py:
# -*- coding: utf-8 -*-
import cv2 # computer vision library
import os
from sklearn.metrics import f1_score
import numpy as np
from imutils import paths
from eval import standardize_input, predict, load_final_model
def unison_shuffled_copies(a, b):
assert len(a) == len(b)
p = np.random.permutation(len(a))
return a[p], b[p]
# Image data directories
def one_hot_encode(label):
dictAnimal = {'cats': 0, 'dogs': 1}
return dictAnimal[label]
def load_data():
IMAGE_DIR_VALIDATION = "animalsval"
imagePaths = sorted(list(paths.list_images(IMAGE_DIR_VALIDATION)))
data = []
labels = []
for imagePath in imagePaths:
data.append(standardize_input(imagePath))
label = imagePath.split(os.path.sep)[-2]
labels.append(one_hot_encode(label))
data = np.array(data, dtype="float")
labels = np.array(labels)
data, labels = unison_shuffled_copies(data, labels)
return data, labels
def get_predictions(data, model):
predictions = [predict(np.array([image]), model) for image in data]
return predictions
def main():
data, labels = load_data()
try:
model = load_final_model()
except:
print('The model is not loaded, we use a constant classifier')
model = None
predictions = get_predictions(data, model)
try:
f1 = f1_score(labels, predictions)
print('F1-Classifier measure:', f1)
except Exception as e:
print('Error: ', e)
file = open("score.txt", "w")
file.write(str(f1))
file.close()
if __name__ == '__main__':
main()
It's eval.py:
# -*- coding: utf-8 -*-
import numpy as np
import cv2
from tensorflow.keras.models import load_model
def standardize_input(image):
standard_im = cv2.imread(image)
standard_im = cv2.resize(standard_im, (32, 32))
standard_im = np.reshape(standard_im, (1, 32, 32, 3))
return standard_im
MODEL_FILE_NAME = 'EasyNet.h5'
def load_final_model():
try:
model = load_model(MODEL_FILE_NAME)
print(type(model))
except Exception:
print("Error")
model = []
return model
def predict(image, model):
data = image
label = 0
pred = model.predict(data)
if pred[0][0] >= pred[0][1]:
label = 0
else:
label = 1
return label
It's model:
import matplotlib
matplotlib.use("Agg")
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report
from tensorflow.keras.models import Sequential
from tensorflow.keras.layer import Dense
from tensorflow.keras.optimizers import SGD
from imutils import paths
import matplotlib.pyplot as plt
import numpy as np
import random
import pickle
import cv2
import os
from tensorflow.keras.callbacks import EarlyStopping
from tensorflow.keras.callbacks import ModelCheckpoint
ImagePaths = list(paths.list_images("/content/drive/My Drive/GoogleColab/dogs/"))
random.seed(42)
ImagePaths = ImagePaths[:1800] + list(paths.list_images("/content/drive/My Drive/GoogleColab/cats/"))
print(len(ImagePaths))
random.shuffle(ImagePaths)
data = []
labels = []
i=0
for imagepath in ImagePaths:
if i % 100 == 0:
print(str(i)+"/3555")
i+=1
image = cv2.imread(imagepath)
image = cv2.resize(image, (32, 32))
data.append(image)
label = imagepath.split(os.path.sep)[-2]
if label == "cats":
label = [1,0]
else:
label = [0,1]
labels.append(label)
print(labels)
data = np.array(data, dtype="float") / 255.0
labels = np.array(labels)
with open("/content/drive/My Drive/GoogleColab/Lesson_4/data.pickle", 'wb') as f:
pickle.dump(data, f)
print("Data saved")
with open("/content/drive/My Drive/GoogleColab/Lesson_4/labels.pickle", 'wb') as f:
pickle.dump(labels, f)
print("Labels saved")
with open("/content/drive/My Drive/GoogleColab/Lesson_4/data.pickle", 'rb') as f:
data = pickle.load(f)
print("Data loaded")
with open("/content/drive/My Drive/GoogleColab/Lesson_4/labels.pickle", 'rb') as f:
labels = pickle.load(f)
print("Labels loaded")
(trainX, testX, trainY, testY) = train_test_split(data, labels,
test_size=0.15,
random_state=42)
print("Dataset prepared")
from tensorflow.keras.layers import Conv2D, Flatten, Dropout, Activation, MaxPooling2D
from tensorflow.keras.optimizers import Adam
model = Sequential()
model.add(Conv2D(32, (3, 3), padding="same", input_shape=(32, 32, 3)))
model.add(Activation("relu"))
model.add(Conv2D(32, (3, 3), padding="same", activation="relu"))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(64, (3, 3), padding="same", activation="relu"))
model.add(Conv2D(64, (3, 3), padding="same", activation="relu"))
model.add (MaxPooling2D( pool_size = ( 2 , 2 )))
model.add(Dropout(0.25))
model.add(Flatten())
#model.add(Dense(1024,input_shape=(3072,), activation='sigmoid'))
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(2, activation='softmax'))
INIT_LR = 0.01
#opt = SGD(lr=INIT_LR)
opt = Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999, amsgrad=False)
model.compile(loss="binary_crossentropy",optimizer=opt,
metrics=["accuracy"]) #categorial_crosentropy
print ("Model compiled")
model.summary()
EPOCHS = 30
checkpointer = ModelCheckpoint(filepath='/content/drive/My Drive/GoogleColab/Lesson_4/ConvNN.h5', verbose=1, save_best_only=True)
H = model.fit(trainX, trainY, validation_data=(testX,testY),
epochs=EPOCHS, batch_size=32,
shuffle=True,
callbacks=[checkpointer])
print("Model trained")
predictions = model.predict (testX, batch_size = 32)
print(predictions)
print(classification_report(testY.argmax(axis=1),
predictions.argmax(axis=1), target_names=("cats","dogs")))
N = np.arange(0, EPOCHS)
plt.style.use("ggplot")
plt.figure()
plt.plot(N, H.history["loss"], label="train_loss")
plt.plot(N, H.history["val_loss"], label="vall_loss")
plt.plot(N, H.history["accuracy"], label="train_acc")
plt.plot(N, H.history["val_accuracy"], label="val_acc")
plt.title("Results")
plt.xlabel("Epoch #")
plt.ylabel("Loss/Accuracy/")
plt.legend()
plt.savefig("/content/drive/My Drive/GoogleColab/Lesson_4/Loss.png")
model.save("/content/drive/My Drive/GoogleColab/Lesson_4/EasyNet.h5")
print("End")