In TensorFlow, suppose I have an image, and I would like to apply the same convolutional kernel to it multiple times. For example, I'd like to do something like the following:
for i in range(5):
output = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(output, weights, [1,1,1,1]), bias))
But this runs out of memory very quickly if I have many loops. I think this is because the system is allocating enough memory to hold the entire sequence at once.
What is an alternative way of doing this?
Thanks!
The way tensorflow works is as follows:
When you create any operation, even if this operation is not used, it will not be destroyed. Therefore, through each iteration, you are creating 3 operations. So after 5 iterations, you end up with 15 operations. So in reality, this is not how to apply the same convolutional kernel to the input. So you should rather create the whole graph first. That is:
x = tf.placeholder(tf.float32, shape=[...], name='input_frames')
weights: tf.Variable(tf.truncated_normal([5, 5, 3, 128], stddev=1e-1, dtype=tf.float32, name='weights1'))
output = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(output, weights, [1,1,1,1]), bias))
for i in range(5):
sess.run(output, feed_dict={x: "your_image"})
Hope this helps.
Related
I'd like to know the possible ways to implement batch normalization layers with synchronizing batch statistics when training with multi-GPU.
Caffe Maybe there are some variants of caffe that could do, like link. But for BN layer, my understanding is that it still synchronizes only the outputs of layers, not the means and vars. Maybe MPI can synchronizes means and vars but I think MPI is a little difficult to implemnt.
Torch I've seen some comments here and here, which show the running_mean and running_var can be synchronized but I think batch mean and batch var can not or are difficult to synchronize.
Tensorflow Normally, it is the same as caffe and torch. The implementation of BN refers this. I know tensorflow can distribute an operation to any device specified by tf.device(). But the computation of means and vars is in the middle of BN layer, so if I gather the means and vars in cpu, my code will be like this:
cpu_gather = []
label_batches = []
for i in range(num_gpu):
with tf.device('/gpu:%d' % i):
with tf.variable_scope('block1', reuse=i > 0):
image_batch, label_batch = cifar_input.build_input('cifar10', train_data_path, batch_size, 'train')
label_batches.append(label_batch)
x = _conv('weights', image_batch, 3, 3, 16, _stride_arr(1))
block1_gather.append(x)
with tf.device('/cpu:0'):
print block1_gather[0].get_shape()
x1 = tf.concat(block1_gather, 0)
# print x1.get_shape()
mean, variance = tf.nn.moments(x1, [0, 1, 2], name='moments')
for i in range(num_gpu):
with tf.device('/gpu:%d' % i):
with tf.variable_scope('block2', reuse=i > 0):
shape = cpu_gather[i].get_shape().as_list()
assert len(shape) in [2, 4]
n_out = shape[-1]
beta, gamma, moving_mean, moving_var = get_bn_variables(n_out, True, True)
x = tf.nn.batch_normalization(
cpu_gather[i], mean, variance, beta, gamma, 0.00001)
x = _relu(x)
That is just for one BN layer. For gathering statistics in cpu, I have to break the code. If I have more than 100 BN layers, that will be cumbersome.
I am not expert in those libraries so maybe there are some misunderstanding, feel free to point out my errors.
I do not care much about training speed. I am doing image segmentation which consumes much GPU memory and BN needs a reasonable batch size (e.g. larger than 16) for stable statistics. So using multi-GPU is inevitable. In my opinion, tensorflow might be the best choice but I can't resolve the breaking code problem. Solution with other libraries will be welcome too.
I'm not sure if I fully understand your question, but provided you set up your variable scope properly, the tf.GraphKeys.UPDATE_OPS collection should automatically have the update ops for batch_norm for each of your towers. If all of the update_ops are applied synchronously, they will be implicitly averaged by the parameter server, all you have to do is make sure the updates are applied before you average and apply gradients. (If I understand your intentions correctly).
Because of variable scope each set of update ops will update the same variables, so to synchronize the update ops all you need to do is gate your gradient calculation on the complete set of update ops. You should also encapsulate all of your batch norm layers in a single name_scope to avoid grabbing any extraneous ops in UPDATE_OPS. Code skeleton below:
update_ops = []
for i, device in enumerate(devices):
with tf.variable_scope('foo', reuse=bool(i > 0)):
with tf.name_scope('tower_%d' % i) as name_scope:
with tf.device(device):
# Put as many batch_norm layers as you want here
update_ops.extend(tf.get_collection(tf.GraphKeys.UPDATE_OPS,
name_scope))
# make gradient calculation ops here
with tf.device(averaging_device):
with tf.control_dependencies(update_ops):
# average and apply gradients.
If you wanna try this on some existing code, try just deleting the if i == 0 line here: https://github.com/tensorflow/models/blob/master/tutorials/image/cifar10_estimator/cifar10_main.py#L115
You're going to see some slow down (we usually only use one tower to compute batch norm statistics for this reason), but it should do what you want.
A specialized keras layer SyncBatchNormalization is available Since TF2.2
https://www.tensorflow.org/api_docs/python/tf/keras/layers/experimental/SyncBatchNormalization
I've figured out a way to implement sync batch norm in pure tensorflow and pure python.
The code makes it possible to train PSPNet on Cityscapes and get comparable performance.
I have a TensorFlow CNN model that is performing well and we would like to implement this model in hardware; i.e., an FPGA. It's a relatively small network but it would be ideal if it were smaller. With that goal, I've examined the kernels and find that there are some where the weights are quite strong and there are others that aren't doing much at all (the kernel values are all close to zero). This occurs specifically in layer 2, corresponding to the tf.Variable() named, "W_conv2". W_conv2 has shape [3, 3, 32, 32]. I would like to freeze/lock the values of W_conv2[:, :, 29, 13] and set them to zero so that the rest of the network can be trained to compensate. Setting the values of this kernel to zero effectively removes/prunes the kernel from the hardware implementation thus achieving the goal stated above.
I have found similar questions with suggestions that generally revolve around one of two approaches;
Suggestion #1:
tf.Variable(some_initial_value, trainable = False)
Implementing this suggestion freezes the entire variable. I want to freeze just a slice, specifically W_conv2[:, :, 29, 13].
Suggestion #2:
Optimizer = tf.train.RMSPropOptimizer(0.001).minimize(loss, var_list)
Again, implementing this suggestion does not allow the use of slices. For instance, if I try the inverse of my stated goal (optimize only a single kernel of a single variable) as follows:
Optimizer = tf.train.RMSPropOptimizer(0.001).minimize(loss, var_list = W_conv2[:,:,0,0]))
I get the following error:
NotImplementedError: ('Trying to optimize unsupported type ', <tf.Tensor 'strided_slice_2228:0' shape=(3, 3) dtype=float32>)
Slicing tf.Variables() isn't possible in the way that I've tried it here. The only thing that I've tried which comes close to doing what I want is using .assign() but this is extremely inefficient, cumbersome, and caveman-like as I've implemented it as follows (after the model is trained):
for _ in range(10000):
# get a new batch of data
# reset the values of W_conv2[:,:,29,13]=0 each time through
for m in range(3):
for n in range(3):
assign_op = W_conv2[m,n,29,13].assign(0)
sess.run(assign_op)
# re-train the rest of the network
_, loss_val = sess.run([optimizer, loss], feed_dict = {
dict_stuff_here
})
print(loss_val)
The model was started in Keras then moved to TensorFlow since Keras didn't seem to have a mechanism to achieve the desired results. I'm starting to think that TensorFlow doesn't allow for pruning but find this hard to believe; it just needs the correct implementation.
A possible approach is to initialize these specific weights with zeros, and modify the minimization process such that gradients won't be applied to them. It can be done by replacing the call to minimize() with something like:
W_conv2_weights = np.ones((3, 3, 32, 32))
W_conv2_weights[:, :, 29, 13] = 0
W_conv2_weights_const = tf.constant(W_conv2_weights)
optimizer = tf.train.RMSPropOptimizer(0.001)
W_conv2_orig_grads = tf.gradients(loss, W_conv2)
W_conv2_grads = tf.multiply(W_conv2_weights_const, W_conv2_orig_grads)
W_conv2_train_op = optimizer.apply_gradients(zip(W_conv2_grads, W_conv2))
rest_grads = tf.gradients(loss, rest_of_vars)
rest_train_op = optimizer.apply_gradients(zip(rest_grads, rest_of_vars))
tf.group([rest_train_op, W_conv2_train_op])
I.e,
Preparing a constant Tensor for canceling the appropriate gradients
Compute gradients only for W_conv2, then multiply element-wise with the constant W_conv2_weights to zero the appropriate gradients and only then apply gradients.
Compute and apply gradients "normally" to the rest of the variables.
Group the 2 train ops to a single training op.
It is is a common practice in convolutional neural networks to oversample a given image during inference,
I.e to create a batch from different transformation of the same image (most common - different crops and mirroring), transfer the entire batch through the network and average (or another kind of reducing function) over the results to get a single prediction (caffe example),
How can this approach be implemented in tensorflow?
You can take a look at the TF cnn tutorial. In particular, the function distorted_inputs does the image preprocessing step.
In short, there are a couple of TF functions in the tf.image package that help with distorting the images. You can use either them or regular numpy functions to create an extra dimension for the output, for which you can average the results:
Before:
input_place = tf.placeholder(tf.float32, [None, 256, 256, 3])
prediction = some_model(input_place) # size: [None]
sess.run(prediction, feed_dict={input_place: batch_of_images})
After:
input_place = tf.placeholder(tf.float32, [None, NUM_OF_DISTORTIONS, 256, 256, 3])
prediction = some_model(input_place) # make sure it is of size [None, NUM_DISTORTIONS]
new_prediction = tf.reduce_mean(prediction, axis=1)
new_batch = np.zeros(batch_size, NUM_OF_DISTORTIONS, 256, 256, 3)
for i in xrange(len(batch_of_images)):
for f in xrange(len(distortion_functions)):
new_batch[i, f, :, :, :] = distortion_functions[f](batch_of_images[i])
sess.run(new_prediction, feed_dict={input_place: new_batch})
Take a look at TF's image-related functions. You could apply those transformations at test time to some input image, and stack all of them together to make a batch.
I imagine you could also do this using OpenCV or some other image processing tool. I don't see a need to do it in the computation graph. You could create the batches beforehand, and pass it through in feed_dict.
I'm making my first steps learning TF and have some trouble training RNNs.
My toy problem goes like this: a two layers LSTM + dense layer network is fed with raw audio data and should test whether a certain frequency is present in the sound.
so the network should 1 to 1 map float(audio data sequence) to float(pre-chosen frequency volume)
I've got this to work on Keras and seen a similar TFLearn solution but would like to implement this on bare Tensorflow in a relatively efficient way.
what i've done:
lstm = rnn_cell.BasicLSTMCell(LSTM_SIZE,state_is_tuple=True,forget_bias=1.0)
lstm = rnn_cell.DropoutWrapper(lstm)
stacked_lstm = rnn_cell.MultiRNNCell([lstm] * 2,state_is_tuple=True)
outputs, states = rnn.dynamic_rnn(stacked_lstm, in, dtype=tf.float32)
outputs = tf.transpose(outputs, [1, 0, 2])
last = tf.gather(outputs, int(outputs.get_shape()[0]) - 1)
network= tf.matmul(last, W) + b
# cost function, optimizer etc...
during training I fed this with (BATCH_SIZE, SEQUENCE_LEN,1) batches and it seems like the loss converged correctly but I can't figure out how to predict with the trained network.
My (awful lot of) questions:
how do i make this network return a sequence right from Tensorflow without going back to python for each sample(feed a sequence and predict a sequence of the same size)?
If I do want to predict one sample at a time and iterate in python what is the correct way to do it?
During testing is dynamic_rnn needed or it's just used for unrolling for BPTT during training? why is dynamic_rnn returning all the back propagation steps Tensors? these are the outputs of each layer of the unrolled network right?
after some research:
how do i make this network return a sequence right from Tensorflow
without going back to python for each sample(feed a sequence and
predict a sequence of the same size)?
you can use state_saving_rnn
class Saver():
def __init__(self):
self.d = {}
def state(self, name):
if not name in self.d:
return tf.zeros([1,LSTM_SIZE],tf.float32)
return self.d[name]
def save_state(self, name, val):
self.d[name] = val
return tf.identity('save_state_name') #<-important for control_dependencies
outputs, states = rnn.state_saving_rnn(stacked_lstm, inx, Saver(),
('lstmstate', 'lstmstate2', 'lstmstate3', 'lstmstate4'),sequence_length=[EVAL_SEQ_LEN])
#4 states are for two layers of lstm each has hidden and CEC variables to restore
network = [tf.matmul(outputs[-1], W) for i in xrange(EVAL_SEQ_LEN)]
one problem is that state_saving_rnn is using rnn() and not dynamic_rnn() therefore unroll at compile time EVAL_SEQ_LEN steps you might want to re-implement state_saving_rnn with dynamic_rnn if you want to input long sequences
If I do want to predict one sample at a time and iterate in python what is the correct way to do it?
you can use dynamic_rnn and supply initial_state. this is probably just as efficient as state_saving_rnn. look at state_saving_rnn implementations for reference
During testing is dynamic_rnn needed or it's just used for unrolling for BPTT during training? why is dynamic_rnn returning all the back propagation steps Tensors? these are the outputs of each layer of the unrolled network right?
dynamic_rnn does do unrolling at runtime similarly to compile time rnn(). I guess it returns all the steps for you to branch the graph in some other places - after less time steps. in a network that use [one time step input * current state -> one output, new state] like the one described above it's not needed in testing but could be used for training truncated time back propagation
I have been using Tensorflow with the l-bfgs optimizer from openopt. It was pretty easy to setup callbacks to allow Tensorflow to compute gradients and loss evaluations for the l-bfgs, however, I am having some trouble figuring out how to introduce stochastic elements like dropout into the training procedure.
During the line search, l-bfgs performs multiple evaluations of the loss function, which need to operate on the same network as the prior gradient evaluation. However, it seems that for each evaluation of the tf.nn.dropout function, a new set of dropouts is created. I am looking for a way to fix the dropout over multiple evaluations of the loss function, and then allow it to change between the gradient steps of the l-bfgs. I'm assuming this has something to do with the control flow ops in tensorflow, but there isn't really a good tutorial on how to use these and they are a little enigmatic to me.
Thanks for your help!
Drop-out relies on uses random_uniform which is a stateful op, and I don't see a way to reset it. However, you can hack around it by substituting your own random numbers and feeding them to the same input point as random_uniform, replacing the generated values
Taking the following code:
tf.reset_default_graph()
a = tf.constant([1, 1, 1, 1, 1], dtype=tf.float32)
graph_level_seed = 1
operation_level_seed = 1
tf.set_random_seed(graph_level_seed)
b = tf.nn.dropout(a, 0.5, seed=operation_level_seed)
Visualize the graph to see where random_uniform is connected
You can see dropout takes input of random_uniform through the Add op which has a name mydropout/random_uniform/(random_uniform). Actually the /(random_uniform) suffix is there for UI reasons, and the true name is mydropout/random_uniform as you can see by printing tf.get_default_graph().as_graph_def(). That gives you shortened tensor name. Now you append :0 to get actual tensor name. (side-note: operation could produce multiple tensors which correspond to suffixes :0, :1 etc. Since having one output is the most common case, :0 is implicit in GraphDef and node input is equivalent to node:0. However :0 is not implicit when using feed_dict so you have to explicitly write node:0)
So now you can fix the seed by generating your own random numbers (of the same shape as incoming tensor), and reusing them between invocations.
tf.reset_default_graph()
a = tf.constant([1, 1, 1, 1, 1], dtype=tf.float32)
graph_level_seed = 1
operation_level_seed = 1
tf.set_random_seed(graph_level_seed)
b = tf.nn.dropout(a, 0.5, seed=operation_level_seed, name="mydropout")
random_numbers = np.random.random(a.get_shape()).astype(dtype=np.float32)
sess = tf.Session()
print sess.run(b, feed_dict={"mydropout/random_uniform:0":random_numbers})
print sess.run(b, feed_dict={"mydropout/random_uniform:0":random_numbers})
You should see the same set of numbers with 2 run calls.