remodels siamese network with vgg16 and 100x100 input

- worse performance than with initial design
- vgg16 pretrained weights are used for the base
  network, which is then piped into a custom head
  model, which
    - flattens the layer (previously done in the base model)
    + Dense Layer
    + Normalization
    + Activation
- training split with 360 fruits used, same as previous mode
- maximum prediction level around 0.95 after ca 60 epochs
This commit is contained in:
Raphael Maenle 2021-07-28 19:02:48 +02:00
parent e5058cc8cc
commit fbc6ee8187
8 changed files with 46 additions and 56 deletions

13
evaluate.py Normal file
View File

@ -0,0 +1,13 @@
import tensorflow.keras as keras
from PIL import Image
import numpy as np
import pdb
model = keras.models.load_model('./siamese_checkpoint')
image1 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_254_100.jpg').convert('RGB').resize((100,
100))) / 255
image2 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_250_100.jpg').convert('RGB').resize((100,
100))) / 255
output = model.predict([np.array([image2]), np.array([image1])])
pdb.set_trace()

View File

@ -28,8 +28,8 @@ the steps taken so far, which lead to a successfull detection of an image
import tensorflow.keras as keras import tensorflow.keras as keras
from PIL import image from PIL import image
model = keras.models.load_model('./siamese_checkpoint') model = keras.models.load_model('./siamese_checkpoint')
image1 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_254_100.jpg').convert('RGB').resize((28, 28))) / 255 image1 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_254_100.jpg').convert('RGB').resize((28, 28)))
image2 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_250_100.jpg').convert('RGB').resize((28, 28))) / 255 image2 = np.asarray(Image.open('../towards/data/fruits-360/Training/Avocado/r_250_100.jpg').convert('RGB').resize((28, 28)))
# note that the double division through 255 is only because the model bas taught with this double division, depends on # note that the double division through 255 is only because the model bas taught with this double division, depends on
# the input numbers of course # the input numbers of course

File diff suppressed because one or more lines are too long

Binary file not shown.

View File

@ -26,12 +26,11 @@ from siamese import SiameseNetwork
import os, math, numpy as np import os, math, numpy as np
from PIL import Image from PIL import Image
batch_size = 128 batch_size = 128
num_classes = 131 num_classes = 131
# input image dimensions # input image dimensions
img_rows, img_cols = 28, 28 img_rows, img_cols = 100, 100
def createTrainingData(): def createTrainingData():
base_dir = 'data/fruits-360/Training/' base_dir = 'data/fruits-360/Training/'
@ -133,22 +132,8 @@ x_train = x_train.astype('float32')
x_test = x_test.astype('float32') x_test = x_test.astype('float32')
def create_own_base_model(input_shape): def create_own_base_model(input_shape):
model_input = Input(shape=input_shape) return keras.applications.vgg16.VGG16(include_top=False, input_tensor=Input(shape=input_shape), weights='imagenet',
classes=1)
embedding = Conv2D(32, kernel_size=(10, 10), input_shape=input_shape)(model_input)
embedding = MaxPooling2D(pool_size=(2, 2))(embedding)
embedding = Conv2D(64, kernel_size=(7, 7))(embedding)
embedding = MaxPooling2D(pool_size=(2, 2))(embedding)
embedding = Conv2D(128, kernel_size=(4, 4))(embedding)
embedding = MaxPooling2D(pool_size=(2, 2))(embedding)
embedding = Conv2D(256, kernel_size=(4, 4))(embedding)
embedding = MaxPooling2D(pool_size=(2, 2))(embedding)
embedding = Flatten()(embedding)
embedding = Dense(4096, activation='sigmoid')(embedding)
embedding = BatchNormalization()(embedding)
embedding = Activation(activation='relu')(embedding)
return Model(model_input, embedding)
def create_base_model(input_shape): def create_base_model(input_shape):
model_input = Input(shape=input_shape) model_input = Input(shape=input_shape)
@ -168,6 +153,30 @@ def create_base_model(input_shape):
return Model(model_input, embedding) return Model(model_input, embedding)
def create_own_head_model(embedding_shape):
embedding_a = Input(shape=embedding_shape[1:])
embedding_b = Input(shape=embedding_shape[1:])
embedding_a_mod = Flatten()(embedding_a)
embedding_a_mod = Dense(128)(embedding_a_mod)
embedding_a_mod = BatchNormalization()(embedding_a_mod)
embedding_a_mod = Activation(activation='relu')(embedding_a_mod)
embedding_b_mod = Flatten()(embedding_b)
embedding_b_mod = Dense(128)(embedding_b_mod)
embedding_b_mod = BatchNormalization()(embedding_b_mod)
embedding_b_mod = Activation(activation='relu')(embedding_b_mod)
head = Concatenate()([embedding_a_mod, embedding_b_mod])
head = Dense(8)(head)
head = BatchNormalization()(head)
head = Activation(activation='sigmoid')(head)
head = Dense(1)(head)
head = BatchNormalization()(head)
head = Activation(activation='sigmoid')(head)
return Model([embedding_a, embedding_b], head)
def create_head_model(embedding_shape): def create_head_model(embedding_shape):
embedding_a = Input(shape=embedding_shape[1:]) embedding_a = Input(shape=embedding_shape[1:])
@ -214,10 +223,10 @@ def get_batch(x_train, y_train, x_test, y_test, cat_train, batch_size=64):
num_classes = 131 num_classes = 131
epochs = 20 epochs = 500
base_model = create_base_model(input_shape) base_model = create_own_base_model(input_shape)
head_model = create_head_model(base_model.output_shape) head_model = create_own_head_model(base_model.output_shape)
siamese_network = SiameseNetwork(base_model, head_model) siamese_network = SiameseNetwork(base_model, head_model)
siamese_network.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) siamese_network.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
@ -225,7 +234,7 @@ siamese_network.compile(loss='binary_crossentropy', optimizer='adam', metrics=['
siamese_checkpoint_path = "./siamese_checkpoint" siamese_checkpoint_path = "./siamese_checkpoint"
siamese_callbacks = [ siamese_callbacks = [
# EarlyStopping(monitor='val_accuracy', patience=10, verbose=0), EarlyStopping(monitor='val_accuracy', patience=10, verbose=0),
ModelCheckpoint(siamese_checkpoint_path, monitor='val_accuracy', save_best_only=True, verbose=0) ModelCheckpoint(siamese_checkpoint_path, monitor='val_accuracy', save_best_only=True, verbose=0)
] ]