Question

I have some images of cats dogs and brids data set that I want to use to train a code using machine learning to distenguish new images by the whether they pictures od dogs, birds or cats. can you show how to do it in an example. please write the code.

Answer 1

CODE:

import numpy as np
import pandas as pd
from keras.preprocessing.image import ImageDataGenerator, img_to_array, load_img
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
import matplotlib.pyplot as plt

datagen = ImageDataGenerator(rescale=1/255)#The ImageDataGenerator class has two methods flow() and flow_from_directory() to read the images from a big numpy array and folders containing images.
train_generator = datagen.flow_from_directory(
directory='Path of your training data set',
target_size=(30, 30),
color_mode="grayscale",
batch_size=5,
class_mode="categorical",
shuffle=True,
seed=42)

valid_generator = datagen.flow_from_directory(
directory='Pat of your validation data set',
target_size=(30, 30),
color_mode="grayscale",
batch_size=5,
class_mode="categorical",
shuffle=True,
seed=42
)

test_generator = datagen.flow_from_directory(
directory='Path of your training data set',
target_size=(30, 30),
color_mode="grayscale",
batch_size=5,
class_mode=None,
shuffle=False,
seed=42
)

# 7. Define model architecture
from keras import backend as K
K.set_image_dim_ordering('th')
model = Sequential()
model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(1,30,30)))
model.add(Conv2D(32, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(79, activation='softmax'))

# 8. Compile model
x=model.compile(loss='mse', optimizer='adam', metrics=['accuracy'])
#..................................................................................................
#Fitting/Training the model
STEP_SIZE_TRAIN=train_generator.n//train_generator.batch_size
STEP_SIZE_VALID=valid_generator.n//valid_generator.batch_size
from keras.callbacks import History
history = History()
hist=model.fit_generator(generator=train_generator,
steps_per_epoch=STEP_SIZE_TRAIN,
validation_data=valid_generator,
validation_steps=STEP_SIZE_VALID,
epochs=10
)

train_loss=hist.history['loss']
val_loss=hist.history['val_loss']
train_acc=hist.history['acc']
val_acc=hist.history['val_acc']
xc=range(10)

plt.figure(1,figsize=(7,5))
plt.plot(xc,train_loss)
plt.plot(xc,val_loss)
plt.plot(xc,train_acc)
plt.plot(xc,val_acc)
plt.grid(True)
plt.xlabel('num of Epochs')
plt.ylabel('loss/accuracy')
plt.legend(['train_loss','val_loss','train_accu','val_accu'],loc=3)
plt.style.use(['classic'])

#Evaluate the model
model.evaluate_generator(generator=valid_generator)

#creating dataframe of train_loss,val_loss,train_acc,val_acc and saving it
dataframe1=pd.DataFrame({'train_loss':train_loss,
'val_loss':val_loss,
'train_acc':train_acc,
'val_acc':val_acc})
dataframe1.to_csv("values.csv",index=True)

#Predict the output
test_generator.reset()
pred=model.predict_generator(test_generator,verbose=1)

predicted_class_indices=np.argmax(pred,axis=1)
labels = (train_generator.class_indices)
labels = dict((v,k) for k,v in labels.items())
predictions = [labels[k] for k in predicted_class_indices]

#Finally save the result to CSV file
filenames=test_generator.filenames
results=pd.DataFrame({"Filename":filenames,
"Predictions":predictions})
results.to_csv("results.csv",index=True)