We Keep Coding

Can a trained ANN (tensorflow) model be made predictable?

I'm new to ANN, but I've managed to train a convolutional model successfully (using some legacy tensorflow v1 code) up to ~90% accuracy or so on my data. But when I evaluate (test) it on any given batch, the result is somewhat random, even though it's 90% correct. I've tried to re-evaluate the data N times and averaging (using N's between 1 and 25), but still each evaluation differs from the others between 3% to 10% of the data points.
Is there any way to make the evaluation predictable, so that the evaluation of an input batch X always yield the exact same result Y every time I run it (once training is done)?
I'm not sure if it's relevant, but my layers are batch normalized like so:
inp = tf.identity(inp)
channels = inp.get_shape()[-1]
offset = tf.compat.v1.get_variable(
'offset', [channels],
dtype=tf.float32,
initializer=tf.compat.v1.zeros_initializer())
scale = tf.compat.v1.get_variable(
'scale', [channels],
dtype=tf.float32,
initializer=tf.compat.v1.random_normal_initializer(1.0, 0.02))
mean, variance = tf.nn.moments(x=inp, axes=[0, 1], keepdims=False)
variance_epsilon = 1e-5
normalized = tf.nn.batch_normalization(
inp, mean, variance, offset, scale, variance_epsilon=variance_epsilon)
The scale part is initialized with random data, but I assume that gets loaded when I do tf.compat.v1.train.Saver().restore(session, checkpoint_fname)?

I am assuming you are testing the model on your training batches?
You can't equate the accuracy of a portion of your total training dataset to the accuracy of the whole.
Think of it like a regression problem. If you only take a part of the dataset, there is no guarantee that it would average out close to the full dataset.
If you want consistent accuracy, evaluate on the full dataset.

training by batches leads to more over-fitting

I'm training a sequence to sequence (seq2seq) model and I have different values to train on for the input_sequence_length.
For values 10 and 15, I get acceptable results but when I try to train with 20, I get memory errors so I switched the training to train by batches but the model over-fit and the validation loss explodes, and even with the accumulated gradient I get the same behavior, so I'm looking for hints and leads to more accurate ways to do the update.
Here is my training function (only with batch section) :
if batch_size is not None:
k=len(list(np.arange(0,(X_train_tensor_1.size()[0]//batch_size-1), batch_size )))
for epoch in range(num_epochs):
optimizer.zero_grad()
epoch_loss=0
for i in list(np.arange(0,(X_train_tensor_1.size()[0]//batch_size-1), batch_size )): # by using equidistant batch till the last one it becomes much faster than using the X.size()[0] directly
sequence = X_train_tensor[i:i+batch_size,:,:].reshape(-1, sequence_length, input_size).to(device)
labels = y_train_tensor[i:i+batch_size,:,:].reshape(-1, sequence_length, output_size).to(device)
# Forward pass
outputs = model(sequence)
loss = criterion(outputs, labels)
epoch_loss+=loss.item()
# Backward and optimize
loss.backward()
optimizer.step()
epoch_loss=epoch_loss/k
model.eval
validation_loss,_= evaluate(model,X_test_hard_tensor_1,y_test_hard_tensor_1)
model.train()
training_loss_log.append(epoch_loss)
print ('Epoch [{}/{}], Train MSELoss: {}, Validation : {} {}'.format(epoch+1, num_epochs,epoch_loss,validation_loss))
EDIT:
here are the parameters that I'm training with :
batch_size = 1024
num_epochs = 25000
learning_rate = 10e-04
optimizer=torch.optim.Adam(model.parameters(), lr=learning_rate)
criterion = nn.MSELoss(reduction='mean')

Batch size affects regularization. Training on a single example at a time is quite noisy, which makes it harder to overfit. Training on batches smoothes everything out, which makes it easier to overfit. Translating back to regularization:
Smaller batches add regularization.
Larger batches reduce regularization.
I am also curious about your learning rate. Every call to loss.backward() will accumulate the gradient. If you have set your learning rate to expect a single example at a time, and not reduced it to account for batch accumulation, then one of two things will happen.
The learning rate will be too high for the now-accumulated gradient, training will diverge, and both training and validation errors will explode.
The learning rate won't be too high, and nothing will diverge. The model will just train more quickly and effectively. If the model is too large for the data being fit, then training error will go to 0 but validation error will explode due to overfitting.
Update
Here is a bit more detail regarding the gradient accumulation.
Every call to loss.backward() will accumulate gradient, until you reset it with optimizer.zero_grad(). It will be acted on when you call optimizer.step(), based on whatever it has accumulated.
The way your code is written, you call loss.backward() for every pass through the inner loop, then you call optimizer.step() in the outer loop before resetting. So the gradient has been accumulated, that is summed, over all examples in the batch and not just one example at a time.
Under most assumptions, that will make the batch-accumulated gradient larger than the gradient for a single example. If the gradients are all aligned, for B batches, it will be larger by B times. If the gradients are i.i.d. then it will be more like sqrt(B) times larger.
If you do not account for this, then you have effectively increased your learning rate by that factor. Some of that will be mitigated by the smoothing effect of larger batches, which can then tolerate a higher learning rate. Larger batches reduce regularization, larger learning rates add it back. But that will not be a perfect match to compensate, so you will still want to adjust accordingly.
In general, whenever you change your batch size you will also want to re-tune your learning rate to compensate.
Leslie N. Smith has written some excellent papers on a methodical approach to hyperparameter tuning. A great place to start is A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay. He recommends you start by reading the diagrams, which are very well done.

What Does Accuracy Metrics Mean in Keras' Sample Denoising Autoencoder?

I am working with Keras' sample denoising autoencoder;
https://keras.io/examples/mnist_denoising_autoencoder/
As I compile it, I use the following options:
autoencoder.compile(loss='mse', optimizer= Adadelta, metrics=['accuracy'])
Followed by training. I did training deliberately WITHOUT using noisy training data(x_train_noisy), but merely tried to recover x_train.
autoencoder.fit(x_train, x_train, epochs=30, batch_size=128)
After training 60,000 inputs of MNIST digits, it gives me an accuracy of 81.25%. Does it mean there are 60000*81.25% images are PERFECTLY recovered (equaling to the original input pixel by pixel), that is, 81.25% output images from the autoencoder are IDENTICAL to their input counterparts, or something else?
Furthermore, I also conducted a manual check by comparing output and the original data (60000 28X28 matrices) pixel by pixel--counting non-zeros elements from their differences:
x_decoded = autoencoder.predict(x_train)
temp = x_train*255
x_train_uint8 = temp.astype('uint8')
temp = x_decoded*255
x_decoded_uint8 = temp.astype('uint8')
c = np.count_nonzero(x_train_uint8 - x_decoded_uint8)
cp = 1-c /60000/28/28
Yet cp is only about 71%. Could any tell me why there is a difference?

Accuracy doesn't make sense for a regression problem, hence the keras sample doesn't use that metric during autoencoder.compile.
In this case, keras calculates the accuracy as per this metric.
binary_accuracy
def binary_accuracy(y_true, y_pred):
return K.mean(K.equal(y_true, K.round(y_pred)), axis=-1)
Using this numpy implementation, you should get the same value as output by Keras for validation accuracy at the end of training.
x_decoded = autoencoder.predict(x_test_noisy)
acc = np.mean(np.equal(x_test, np.round(x_decoded)))
print(acc)
Refer this answer for more details:
What function defines accuracy in Keras when the loss is mean squared error (MSE)?

Add or remove loss function based on wether sample is labeled or not in tensorflow

I want to implement PI Model architecture proposed in this paper "TEMPORAL ENSEMBLING FOR SEMI-SUPERVISED
LEARNING".
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjKiqyMmOfZAhXGz4MKHcFJCxEQFgguMAA&url=https%3A%2F%2Farxiv.org%2Fabs%2F1610.02242&usg=AOvVaw3VuNNeChU0O7WYXZQsfxOh.
The authors provided the code in theano but theano is no longer supported so I am writing the algorithm in tensorflow.
The paper is about semi-supervised model where a batch of training data contains both labeled and unlabeled samples. The algorithm has two branches - one for labeled and the other for unlabeled samples. One sample serves as the input to the two branches, however in ech branch the input undergoes random augmentation. In the labeled branch the cross entropy loss is used between the label and the output of a CNN while for the unlabeled branch, the mean square loss between the outputs of a CNN for the two random augmented versions of the same input is used.
I am using cifar10 dataset and I have randomly changed the lables of some samples to 10 to represent the unlabled samples. For either labeled or unlabled sample, the mean square will always be part of the training loss, however if the sample is unlabeled(i.e, the label == 10) then cross entropy should be zero.
I have come up with the following snippet in my code
sentinel = tf.Variable(-1, name='sentinel')
if tf.argmax(y_, axis=1)[0] != tf.constant(10):
sentinel = sentinel
else:
sentinel = 1
cross_entropy = tf.cond(tf.greater(sentinel, tf.constant(0)), lambda: tf.constant(0.0), lambda: tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=train_prediction)), name ='cross_entropy')
train_loss = cross_entropy
mean_sq_loss = tf.losses.mean_squared_error(train_prediction, train_prediction_b)
train_loss += unsup_weight * mean_sq_loss
train_op = tf.train.AdamOptimizer(config.adam_epsilon).minimize(train_loss)
but the condition
if tf.argmax(y_, axis=1)[0] != tf.constant(10):
does not evaluate to true even when
tf.argmax(y_, axis=1)[0] equals 10.
I used sentinel to track if the sample is labeled or not but when I checked the output for a batch size of 1, sentinel does not change even when the sample is unlabeled. I will appreciate any contribution.
Thanks

Why does recognition rate drop after multiple online training epochs?

I am using tensorflow to do image recognition on the MNIST dataset. In each training epoch, I picked 10,000 random images and conducted online training with batch size of 1. The recognition rate increased for the first few epochs, however, after several epochs the recognition rate started to drop greatly. (In the first 20 epochs, the recognition rate goes up to ~94%. Afterwards, the recognition rate went from 90->50->40->30->20). What is the reason for this?
Also, with a batch size of 1, the performance is worse than when using a batch size of 100 (max recognition rate 94% vs. 96%). I looked through several references but there seems to be contradictory results on whether small or large batch sizes achieve better performance. What would be this case in this situation?
Edit: I also added a figure of the recognition rate of the training dataset and the test dataset.Recognition rate vs. epoch
I have attached a copy of the code below. Thanks for the help!
import tensorflow as tf
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/", one_hot = True)
#parameters
n_nodes_hl1 = 500
n_nodes_hl2 = 500
n_nodes_hl3 = 500
n_classes = 10
batch_size = 1
x = tf.placeholder('float', [None, 784])
y = tf.placeholder('float')
#model of neural network
def neural_network_model(data):
hidden_1_layer = {'weights':tf.Variable(tf.random_normal([784, n_nodes_hl1]) , name='l1_w'),
'biases': tf.Variable(tf.random_normal([n_nodes_hl1]) , name='l1_b')}
hidden_2_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl1, n_nodes_hl2]) , name='l2_w'),
'biases' :tf.Variable(tf.random_normal([n_nodes_hl2]) , name='l2_b')}
hidden_3_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl2, n_nodes_hl3]) , name='l3_w'),
'biases' :tf.Variable(tf.random_normal([n_nodes_hl3]) , name='l3_b')}
output_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl3, n_classes]) , name='lo_w'),
'biases' :tf.Variable(tf.random_normal([n_classes]) , name='lo_b')}
l1 = tf.add(tf.matmul(data,hidden_1_layer['weights']), hidden_1_layer['biases'])
l1 = tf.nn.relu(l1)
l2 = tf.add(tf.matmul(l1,hidden_2_layer['weights']), hidden_2_layer['biases'])
l2 = tf.nn.relu(l2)
l3 = tf.add(tf.matmul(l2,hidden_3_layer['weights']), hidden_3_layer['biases'])
l3 = tf.nn.relu(l3)
output = tf.matmul(l3,output_layer['weights']) + output_layer['biases']
return output
#train neural network
def train_neural_network(x):
prediction = neural_network_model(x)
cost = tf.reduce_mean( tf.nn.softmax_cross_entropy_with_logits(logits=prediction, labels=y))
optimizer = tf.train.AdamOptimizer().minimize(cost)
hm_epoches = 100
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
for epoch in range(hm_epoches):
epoch_loss=0
for batch in range (10000):
epoch_x, epoch_y=mnist.train.next_batch(batch_size)
_,c =sess.run([optimizer, cost], feed_dict = {x:epoch_x, y:epoch_y})
epoch_loss += c
correct = tf.equal(tf.argmax(prediction, 1), tf.argmax(y,1))
accuracy = tf.reduce_mean(tf.cast(correct, 'float'))
print(epoch_loss)
print('Accuracy_test:', accuracy.eval({x:mnist.test.images, y:mnist.test.labels}))
print('Accuracy_train:', accuracy.eval({x:mnist.train.images, y:mnist.train.labels}))
train_neural_network(x)

DROPPING ACCURACY
You're over-fitting. This is when the model learns false features that are specific to artifacts of the images in the training data, at the expense of important features. One of the main experimental results of any application is to determine the optimal number of training iterations.
For instance, perhaps 80% of the 7's in your training data happen to have a little extra slant to the right near the bottom of the stem, where 4's and 1's do not. After too much training, your model "decides" that the best way to tell a 7 from another digit is from that extra slant, despite any other features. As a result, some 1's and 4's now get classed as 7's.
BATCH SIZE
Again, the best batch size is one of the experimental results. Typically, a batch size of 1 is too small: this gives the first few input images too much influence on the early weights in kernel or perceptron training. This is a minor case of over-fitting: one item having undue influence on the model. However, it's significant enough to alter your best results by 2%.
You need to balance the batch size with the other hyper-parameters to find the model's "sweet spot", optimum performance followed by shortest training time. In my experience, it's been best to increase the batch size until my time per image degraded. The models I've used most (MNIST, CIFAR-10, AlexNet, GoogleNet, ResNet, VGG, etc.) had very little loss of accuracy once we reached a rather minimal batch size; from there, the training speed was usually a matter of choosing the batch size the best used available RAM.

There are a few possibilities, although you'll need to do some experimentation to find out which it is.
Overfitting
Prune did a good job of explaining this. I'll add that the simplest way to avoid overfitting is to just remove 10-15% of the training set and evaluate the recognition rate on this held out validation set after every few epochs. If you graph the change in recognition rate on both the training and validation sets, you'll eventually reach a point on the graph where the training error keeps going down but the validation error starts going up. Stop training at that point; that's where overfitting is starting in earnest. Note that it's important that there be no overlap between the training/validation/test sets.
This was more likely before you mentioned that the training error wasn't also decreasing, but it's possible that it's overfitting on a fairly homogeneous part of your training set at the expense of the outliers, or something like this. Try randomizing the order of your training set after each epoch; if it's fitting one section of the set at the expense of the others, this might help.
Addendum: The massive instantaneous drop in quality around epoch 20 makes this even less likely; that is not what overfitting looks like.
Numerical Instability
If you get a particularly incorrect input at a point on the activation function with a large gradient, it's possible to end up with a gigantic weight update that screws up everything it's learned thus far. It's common to put a hard limit on the gradient magnitude for this reason. But you're using AdamOptimizer, which has an epsilon parameter for avoiding instability. I haven't read the paper it references, so I don't know exactly how it works, but the fact that it's there makes instability less likely.
Saturated Neurons
Some activation functions have regions with very small gradients, so if you end up with weights such that the function is almost always in that region, you have a tiny gradient and thus can't learn effectively. Sigmoids and Tanh are particularly prone to this since they have flat regions on both sides of the function. ReLUs don't have a flat region on the high end, but do on the low end. Try replacing your activation functions with Softplus; those are similar to ReLU, but with a continuous nonzero gradient.