Keras vs. TensorFlow code comparison sources - tensorflow

This isn't really a question that's code-specific, but I haven't been able to find any answers or resources.
I'm currently trying to teach myself some "pure" TensorFlow rather than just using Keras, and I felt that it would be very helpful if there were some sources where they have TensorFlow code and the equivalent Keras code side-by-side for comparison.
Unfortunately, most of the results I find on the Internet talk about performance-wise differences or have very simple comparison examples (e.g. "and so this is why Keras is much simpler to use"). I'm not so much interested in those details as much as I am in the code itself.
Does anybody know if there are any resources out there that could help with this?

Here you have two models, in Tensorflow and in Keras, that are correspondent:
import tensorflow as tf
import numpy as np
import pandas as pd
from keras.datasets import mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
Tensorflow
X = tf.placeholder(dtype=tf.float64)
Y = tf.placeholder(dtype=tf.float64)
num_hidden=128
# Build a hidden layer
W_hidden = tf.Variable(np.random.randn(784, num_hidden))
b_hidden = tf.Variable(np.random.randn(num_hidden))
p_hidden = tf.nn.sigmoid( tf.add(tf.matmul(X, W_hidden), b_hidden) )
# Build another hidden layer
W_hidden2 = tf.Variable(np.random.randn(num_hidden, num_hidden))
b_hidden2 = tf.Variable(np.random.randn(num_hidden))
p_hidden2 = tf.nn.sigmoid( tf.add(tf.matmul(p_hidden, W_hidden2), b_hidden2) )
# Build the output layer
W_output = tf.Variable(np.random.randn(num_hidden, 10))
b_output = tf.Variable(np.random.randn(10))
p_output = tf.nn.softmax( tf.add(tf.matmul(p_hidden2, W_output), b_output) )
loss = tf.reduce_mean(tf.losses.mean_squared_error(
labels=Y,predictions=p_output))
accuracy=1-tf.sqrt(loss)
minimization_op = tf.train.AdamOptimizer(learning_rate=0.01).minimize(loss)
feed_dict = {
X: x_train.reshape(-1,784),
Y: pd.get_dummies(y_train)
}
with tf.Session() as session:
session.run(tf.global_variables_initializer())
for step in range(10000):
J_value = session.run(loss, feed_dict)
acc = session.run(accuracy, feed_dict)
if step % 100 == 0:
print("Step:", step, " Loss:", J_value," Accuracy:", acc)
session.run(minimization_op, feed_dict)
pred00 = session.run([p_output], feed_dict={X: x_test.reshape(-1,784)})
Keras
import tensorflow as tf
from tensorflow.keras.layers import Input, Dense
from keras.models import Model
l = tf.keras.layers
model = tf.keras.Sequential([
l.Flatten(input_shape=(784,)),
l.Dense(128, activation='relu'),
l.Dense(128, activation='relu'),
l.Dense(10, activation='softmax')
])
model.compile(loss='categorical_crossentropy', optimizer='adam',metrics = ['accuracy'])
model.summary()
model.fit(x_train.reshape(-1,784),pd.get_dummies(y_train),nb_epoch=15,batch_size=128,verbose=1)

You can take a look to this toy example, but it may be too simple.

Related

NaN loss in fit parametrs

import pyreadr
from sklearn.model_selection import train_test_split
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import numpy as np
from sklearn import preprocessing
# сбор тестовых данных в массивы
result = pyreadr.read_r("/home/ignat/Downloads/Telegram Desktop/TMS_coefficients.RData")
dataset = []
values = []
for i in range(694):
dataset.append((result['tms.coef']['bs0'][i],
result['tms.coef']['bs'][i],
result['tms.coef']['bi0'][i],
result['tms.coef']['bi'][i],
result['tms.coef']['b0'][i],
result['tms.coef']['b1'][i],
result['tms.coef']['b2'][i],
result['tms.coef']['a0'][i],
result['tms.coef']['a1'][i]))
values.append([0.0 if result['tms.coef']['Y'][i] == "НС"
else 1.0 if result['tms.coef']['Y'][i] == "AD"
else 2.0 if result['tms.coef']['Y'][i] == "DLB"
else 3.0])
dataset = np.array(dataset, dtype="float")
values = np.array(values, dtype="float")
print(dataset[0])
print(values[0])
(trainX, testX, trainY, testY) = train_test_split(dataset,
values, test_size=0.25, random_state=42)
# модель нейронки
visible = layers.Input(shape=(9,))
drop1 = layers.Dropout(0.5, input_shape=(9,))(visible)
hidden1 = layers.Dense(32, activation="relu")(drop1)
drop2 = layers.Dropout(0.5, input_shape=(9,))(hidden1)
output = layers.Dense(1, activation="relu")(drop2)
model = tf.keras.Model(inputs=visible, outputs=output)
# компиляция
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
loss='sparse_categorical_crossentropy',
metrics=['accuracy'],)
# обучение
model.fit(trainX, trainY, validation_data=(testX, testY), epochs=100)
model.save('my_model')
for i in range(1, 10):
print(model.predict(dataset), values[-i])
While go fit I have troubles - Nan Lose
dataset[i] example:
1.35684728e-01 -4.03785135e-02 -8.27514734e-02 4.21657613e-03
1.40184876e-01 1.06088863e-02 -1.31599134e-03 -1.77011366e+00
-7.19767825e-02
values[i] example:
1.
I beg your help for me in debug only this fragment of code. Im google the problem very much, because i wont to see abstract instruction
I think the main problem here might be... the loss function that you have chosen. Normally categorical_crossentropy is used in multiclass problems. As you are having only 1 neuron in your output layer this might lead to some problems. So if you are trying to make a classification you might want to switch the loss to:
# компиляция
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
loss='binary_crossentropy',
metrics=['accuracy'],)

How to find class labels from a keras model

I am predicting classes, but there is something I don't get. In the simplified example below, I train a model to predict MNIST handwritten digits. My test set has an accuracy of 95%, when I use
model.evaluate(test_image, test_label)
However, when I use
model.predict(test_image)
and the extract the predicted labels using np.argmax(), this accuracy drops. When I run all the code again and again, this accuracy changes a lot.
I suspect now that the classes in the model are not ordered 0, 1 ... 9. Is there a way to see the class labels of a model? Or did I make another mistake?
import matplotlib.pyplot as plt
import tensorflow as tf
from tensorflow.keras.datasets.mnist import load_data
from tensorflow.keras.layers import Dense, Flatten
from tensorflow.keras.models import Sequential
import numpy as np
# Load data
(train_image, train_label), (test_image, test_label) = load_data()
# Train
model = Sequential([
Flatten(input_shape=(28,28)),
Dense(100, activation="relu"),
Dense(100, activation="relu"),
Dense(10, activation="sigmoid")
])
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics='accuracy')
history = model.fit(train_image, train_label,
batch_size=32, epochs=50,
validation_data=(test_image, test_label),
verbose = 0)
eval = model.evaluate(test_image, test_label)
print('Accuracy (auto):', eval[1]) # This is always high
# Predict and evaluate manually
predictions = model.predict(test_image)
pred = np.array([np.argmax(pred) for pred in predictions])
true = test_label
print('Accuracy (manually):', np.mean(pred == true)) # This varies a lot

Keras generator function in model.fit()

I wrote a simple code to train a neural network with Keras model.fit() with generator. It is important for me to know
how many times the model.fit() calls the generator??
It seems that the generator is called one time before the first epoch and I don't know why?? This one extra call is visible in the attached figure --> 1
Here is my simple code.
import numpy
from keras.models import Model
from keras import layers, losses
from tensorflow.keras import optimizers
input_layer = layers.Input(shape=(256,))
x_layer = layers.BatchNormalization()(input_layer)
x_layer = layers.Dense(500, activation='relu')(input_layer)
x_layer = layers.BatchNormalization()(x_layer)
x_layer = layers.Dense(250, activation='relu')(x_layer)
x_layer = layers.BatchNormalization()(x_layer)
x_layer = layers.Dense(120, activation='relu')(x_layer)
x_layer = layers.BatchNormalization()(x_layer)
output_layer = layers.Dense(16, activation='sigmoid')(x_layer)
NN_model = Model(inputs=input_layer, outputs=output_layer)
NN_model.compile(optimizer=optimizers.RMSprop(), loss=losses.mean_squared_error)
NN_model.summary()
def training_generator(number_of_samples):
while True:
print('\n ---------- One Refer to This File')
NN_inputs = []
NN_labels = []
for _ in range(number_of_samples):
NN_inputs.append(numpy.arange(256))
NN_labels.append(numpy.arange(16))
yield numpy.asarray(NN_inputs), numpy.asarray(NN_labels)
NN_history = NN_model.fit(training_generator(1000),
steps_per_epoch=4,
epochs=10,
verbose=2)

degraded accuracy performance with overfitting when downgrading from tensorflow 2.3.1 to tensorflow 1.14 or 1.15 on multiclass categorization

I made a script in tensorflow 2.x but I had to downconvert it to tensorflow 1.x (tested in 1.14 and 1.15). However, the tf1 version performs very differently (10% accuracy lower on the test set). See also the plot for train and validation performance (diagram is attached below).
Looking at the operations needed for the migration from tf1 to tf2 it seems that only the Adam learning rate may be a problem but I'm defining it explicitly tensorflow migration
I've reproduced the same behavior both locally on GPU and CPU and on colab. The keras used was the one built-in in tensorflow (tf.keras). I've used the following functions (both for train,validation and test), using a sparse categorization (integers):
train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
horizontal_flip=horizontal_flip,
#rescale=None, #not needed for resnet50
preprocessing_function=None,
validation_split=None)
train_dataset = train_datagen.flow_from_directory(
directory=train_dir,
target_size=image_size,
class_mode='sparse',
batch_size=batch_size,
shuffle=True)
And the model is a simple resnet50 with a new layer on top:
IMG_SHAPE = img_size+(3,)
inputs = Input(shape=IMG_SHAPE, name='image_input',dtype = tf.uint8)
x = tf.cast(inputs, tf.float32)
# not working in this version of keras. inserted in imageGenerator
x = preprocess_input_resnet50(x)
base_model = tf.keras.applications.ResNet50(
include_top=False,
input_shape = IMG_SHAPE,
pooling=None,
weights='imagenet')
# Freeze the pretrained weights
base_model.trainable = False
x=base_model(x)
# Rebuild top
x = GlobalAveragePooling2D(data_format='channels_last',name="avg_pool")(x)
top_dropout_rate = 0.2
x = Dropout(top_dropout_rate, name="top_dropout")(x)
outputs = Dense(num_classes,activation="softmax", name="pred_out")(x)
model = Model(inputs=inputs, outputs=outputs,name="ResNet50_comp")
optimizer = tf.keras.optimizers.Adam(lr=learning_rate)
model.compile(optimizer=optimizer,
loss="sparse_categorical_crossentropy",
metrics=['accuracy'])
And then I'm calling the fit function:
history = model.fit_generator(train_dataset,
steps_per_epoch=n_train_batches,
validation_data=validation_dataset,
validation_steps=n_val_batches,
epochs=initial_epochs,
verbose=1,
callbacks=[stopping])
I've reproduced the same behavior for example with the following full script (applied to my dataset and changed to adam and removed intermediate final dense layer):
deep learning sandbox
The easiest way to replicate this behavior was to enable or disable the following line on a tf2 environment with the same script and add the following line to it. However, I've tested also on tf1 environments (1.14 and 1.15):
tf.compat.v1.disable_v2_behavior()
Sadly I cannot provide the dataset.
Update 26/11/2020
For full reproducibility I've obtained a similar behaviour by means of the food101 (101 categories) dataset enabling tf1 behaviour with 'tf.compat.v1.disable_v2_behavior()'. The following is the script executed with tensorflow-gpu 2.2.0:
#%% ref https://medium.com/deeplearningsandbox/how-to-use-transfer-learning-and-fine-tuning-in-keras-and-tensorflow-to-build-an-image-recognition-94b0b02444f2
import os
import sys
import glob
import argparse
import matplotlib.pyplot as plt
import tensorflow as tf
# enable and disable this to obtain tf1 behaviour
tf.compat.v1.disable_v2_behavior()
from tensorflow.keras import __version__
from tensorflow.keras.applications.resnet50 import ResNet50, preprocess_input
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
from tensorflow.keras.optimizers import Adam
# since i'm using resnet50 weights from imagenet, i'm using food101 for
# similar but different categorization tasks
# pip install tensorflow-datasets if tensorflow_dataset not found
import tensorflow_datasets as tfds
(train_ds,validation_ds),info= tfds.load('food101', split=['train','validation'], shuffle_files=True, with_info=True)
assert isinstance(train_ds, tf.data.Dataset)
print(train_ds)
#%%
IM_WIDTH, IM_HEIGHT = 224, 224
NB_EPOCHS = 10
BAT_SIZE = 32
def get_nb_files(directory):
"""Get number of files by searching directory recursively"""
if not os.path.exists(directory):
return 0
cnt = 0
for r, dirs, files in os.walk(directory):
for dr in dirs:
cnt += len(glob.glob(os.path.join(r, dr + "/*")))
return cnt
def setup_to_transfer_learn(model, base_model):
"""Freeze all layers and compile the model"""
for layer in base_model.layers:
layer.trainable = False
model.compile(optimizer='rmsprop', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
def add_new_last_layer(base_model, nb_classes):
"""Add last layer to the convnet
Args:
base_model: keras model excluding top
nb_classes: # of classes
Returns:
new keras model with last layer
"""
x = base_model.output
x = GlobalAveragePooling2D()(x)
#x = Dense(FC_SIZE, activation='relu')(x) #new FC layer, random init
predictions = Dense(nb_classes, activation='softmax')(x) #new softmax layer
model = Model(inputs=base_model.input, outputs=predictions)
return model
def train(nb_epoch, batch_size):
"""Use transfer learning and fine-tuning to train a network on a new dataset"""
#nb_train_samples = train_ds.cardinality().numpy()
nb_train_samples=info.splits['train'].num_examples
nb_classes = info.features['label'].num_classes
classes_names = info.features['label'].names
#nb_val_samples = validation_ds.cardinality().numpy()
nb_val_samples = info.splits['validation'].num_examples
#nb_epoch = int(args.nb_epoch)
#batch_size = int(args.batch_size)
def preprocess(features):
#print(features['image'], features['label'])
image = tf.image.resize(features['image'], [224,224])
#image = tf.divide(image, 255)
#print(image)
# data augmentation
image=tf.image.random_flip_left_right(image)
image = preprocess_input(image)
label = features['label']
# for categorical crossentropy
#label = tf.one_hot(label,101,axis=-1)
#return image, tf.cast(label, tf.float32)
return image, label
#pre-processing the dataset to fit a specific image size and 2D labelling
train_generator = train_ds.map(preprocess).batch(batch_size).repeat()
validation_generator = validation_ds.map(preprocess).batch(batch_size).repeat()
#train_generator=train_ds
#validation_generator=validation_ds
#fig = tfds.show_examples(validation_generator, info)
# setup model
base_model = ResNet50(weights='imagenet', include_top=False) #include_top=False excludes final FC layer
model = add_new_last_layer(base_model, nb_classes)
# transfer learning
setup_to_transfer_learn(model, base_model)
history = model.fit(
train_generator,
epochs=nb_epoch,
steps_per_epoch=nb_train_samples//BAT_SIZE,
validation_data=validation_generator,
validation_steps=nb_val_samples//BAT_SIZE)
#class_weight='auto')
#execute
history = train(nb_epoch=NB_EPOCHS, batch_size=BAT_SIZE)
And the performance on food101 dataset:
update 27/11/2020
It's possible to see the discrepancy also in the way smaller oxford_flowers102 dataset:
(train_ds,validation_ds,test_ds),info= tfds.load('oxford_flowers102', split=['train','validation','test'], shuffle_files=True, with_info=True)
Nb: the above plot shows confidences given by running the same training multiple times and evaluatind mean and std to check for the effects on random weights initialization and data augmentation.
Moreover I've tried some hyperparameter tuning on tf2 resulting in the following picture:
changing optimizer (adam and rmsprop)
not applying horizontal flipping aumgentation
deactivating keras resnet50 preprocess_input
Thanks in advance for every suggestion. Here are the accuracy and validation performance on tf1 and tf2 on my dataset:
Update 14/12/2020
I'm sharing the colab for reproducibility on oxford_flowers at the clic of a button:
colab script
I came across something similar, when doing the opposite migration (from TF1+Keras to TF2).
Running this code below:
# using TF2
import numpy as np
from tensorflow.keras.applications.resnet50 import ResNet50
fe = ResNet50(include_top=False, pooling="avg")
out = fe.predict(np.ones((1,224,224,3))).flatten()
sum(out)
>>> 212.3205274187726
# using TF1+Keras
import numpy as np
from keras.applications.resnet50 import ResNet50
fe = ResNet50(include_top=False, pooling="avg")
out = fe.predict(np.ones((1,224,224,3))).flatten()
sum(out)
>>> 187.23898954353717
you can see the same model from the same library on different versions does not return the same value (using sum as a quick check-up). I found the answer to this mysterious behavior in this other SO answer: ResNet model in keras and tf.keras give different output for the same image
Another recommendation I'd give you is, try using pooling from inside applications.resnet50.ResNet50 class, instead of the additional layer in your function, for simplicity, and to remove possible problem-generators :)

How do I call model.predict with data stored in GPU in Keras

I have built and trained a Sequential model.
Now before each model.predict call I want to upload the data into GPU, do some operations and then call model.predict using the output stored in GPU without downloading to memory and handover to keras model for it to upload to gpu again.
Edit:
I would like to use opencv operations on the input image in gpu and use the output directly to call model.predict if possible.
You can easily achieve this by adding the operations as Lambda layers on the top of the model.
Here is a very simple example.. you can extend from here:
import numpy as np
from keras import backend as K
from keras.models import Sequential, Model
from keras.layers import Dense, Lambda, Input, merge
X = np.random.random((1000,5))
Y = np.random.random((1000,1))
inp = Input(shape = (5,))
d1 = Dense(60, input_dim=5, init='normal', activation='relu')
d2 = Dense(1, init='normal', activation='sigmoid')
out = d2(d1(inp))
model = Model(input=[inp], output=[out])
model.compile(loss='binary_crossentropy', optimizer='adam')
model.summary()
model.fit(X, Y, nb_epoch=1)
X1 = np.random.random((10,3))
X2 = np.random.random((10,2))
inp1 = Input(shape = (3,))
inp2 = Input(shape = (2,))
p1 = Lambda(lambda x: K.sqrt(x))(inp1)
p2 = Lambda(lambda x: K.tf.exp(x))(inp2)
mer = merge([p1, p2], mode='concat')
out2 = d2(d1(mer))
model2 = Model(input=[inp1, inp2], output=[out2])
model2.summary()
ypred = model2.predict([X1, X2])
print ypred.shape
Here from model.summary() you can see both the models are sharing the upper layers so in essence is using the weights already learnt during the training of the first model