I have a script that performs a Gatys-like neural style transfer. It uses style loss, and a total variation loss. I'm using the GradientTape() to compute my gradients. The losses that I have implemented seem to work fine, but a new loss that I added isn't being properly accounted for by the GradientTape(). I'm using TensorFlow with eager execution enabled.
I suspect it has something to do with how I compute the loss based on the input variable. The input is a 4D tensor (batch, h, w, channels). At the most basic level, the input is a floating point image, and in order to compute this new loss I need to convert it to a binary image to compute the ratio of one pixel color to another. I don't want to actually go and change the image like that during every iteration, so I just make a copy of the tensor(in numpy form) and operate on that to compute the loss. I do not understand the limitations of the GradientTape, but I believe it is "losing the thread" of how the input variable is used to get to the loss when it's converted to a numpy array.
Could I make a copy of the image tensor and perform binarizing operations & loss computation using that? Or am I asking tensorflow to do something that it just can not do?
My new loss function:
def compute_loss(self, **kwargs):
loss = 0
image = self.model.deprocess_image(kwargs['image'].numpy())
binarized_image = self.image_decoder.binarize_image(image)
volume_fraction = self.compute_volume_fraction(binarized_image)
loss = np.abs(self.volume_fraction_target - volume_fraction)
return loss
My implementation using the GradientTape:
def compute_grads_and_losses(self, style_transfer_state):
"""
Computes gradients with respect to input image
"""
with tf.GradientTape() as tape:
loss = self.loss_evaluator.compute_total_loss(style_transfer_state)
total_loss = loss['total_loss']
return tape.gradient(total_loss, style_transfer_state['image']), loss
An example that I believe might illustrate my confusion. The strangest thing is that my code doesn't have any problem running; it just doesn't seem to minimize the new loss term whatsoever. But this example won't even run due to an attribute error: AttributeError: 'numpy.float64' object has no attribute '_id'.
Example:
import tensorflow.contrib.eager as tfe
import tensorflow as tf
def compute_square_of_value(x):
a = turn_to_numpy(x['x'])
return a**2
def turn_to_numpy(arg):
return arg.numpy() #just return arg to eliminate the error
tf.enable_eager_execution()
x = tfe.Variable(3.0, dtype=tf.float32)
data_dict = {'x': x}
with tf.GradientTape() as tape:
tape.watch(x)
y = compute_square_of_value(data_dict)
dy_dx = tape.gradient(y, x) # Will compute to 6.0
print(dy_dx)
Edit:
From my current understanding the issue arises that my use of the .numpy() operation is what makes the Gradient Tape lose track of the variable to compute the gradient from. My original reason for doing this is because my loss operation requires me to physically change values of the tensor, and I don't want to actually change the values used for the tensor that is being optimized. Hence the use of the numpy() copy to work on in order to compute the loss properly. Is there any way around this? Or is shall I consider my loss calculation to be impossible to implement because of this constraint of having to perform essentially non-reversible operations on the input tensor?
The first issue here is that GradientTape only traces operations on tf.Tensor objects. When you call tensor.numpy() the operations executed there fall outside the tape.
The second issue is that your first example never calls tape.watche on the image you want to differentiate with respect to.
Related
I'm trying to implement a loss function that depends on the gradient of the network with respect to its inputs. That is, the loss function has a term like
sum(u - grad_x(network(x)))
where u is computed by forward propagating x through the network.
I'm able to compute the gradient by calling
funcApprox = funcNetwork.forward(X)
funcGrad = grad(funcApprox, X, grad_outputs=torch.ones_like(funcApprox))
Here, funcNetwork is my NN and X is the input. These computations are done in the loss function.
However, now if I attempt to do the following
opt.zero_grad()
loss = self.loss(X) # My custom loss function that calculates funcGrad, etc., from above
opt.zero_grad()
loss.backward()
opt.step()
I see the following error:
RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward.
on the loss.backward() line from above.
I've tried playing around with create_graph, retain_graph, etc. but to no avail.
Any help is appreciated!
As per comment by #aretor, setting retain_graph=True, create_graph=False in the grad call in the loss function, and retain_graph=True in backward solves the issue.
Let's take the following code:
'''
LSTM class
'''
import torch
import pandas as pd
import numpy as np
import torch.nn.CrossEntropyLoss
class LSTM(nn.Module):
def __init__(self, input_size, hidden_size, num_layers):
super (LSTM, self).__init__()
self.hidden_size = hidden_size
self.lstm = nn.LSTM(input_size, hidden_size, num_layers)
def forward(self, x):
# receive an input, create a new hidden state, return output?
# reset the hidden state?
hidden = (torch.zeros(num_layers, hidden_size), torch.zeros(num_layers, hidden_size))
x, hidden = self.lstm(x, hidden)
#since our observation has several sequences, we only want the output after the last sequence of the observation'''
x = x[:, -1]
return x
I have several questions here, and if permitted would rather ask them all at once rather than waiting 90 minutes between singular posts.
I've seen and followed quite a few examples of LSTMs in pytorch and each example seems to treat different pieces a bit differently. Since i'm not in expert in either python, or neural networks, this has lead me to a lot of confusion. I'll ask my question sequentially in order of how they appear in the code above.
I've seen the hidden layer both defined, zeroed, left out and ignored entirely in a few different implementations. I know what its for, but in the implementation i've produced (which is itself an amalgamation of several tutorials) the hidden layer doesn't appear to be connected to anything. in the forward function we take a single input pass it to the hidden layer (which is first zero'd) then call self.lstm on it. Is this the equivalent of letting lstm "handle" the hidden layer itself?
Will this properly produce a hidden state?
Am I correct that the optimization only occurs during the training loop? I was using this particular tutorial as an example:
https://pytorch.org/tutorials/beginner/basics/optimization_tutorial.html
def train_loop(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset)
for batch, (X, y) in enumerate(dataloader):
# Compute prediction and loss
pred = model(X)
loss = loss_fn(pred, y)
# Backpropagation
optimizer.zero_grad()
loss.backward()
optimizer.step()
in the optimizer tutorial, i assume y is the true label for the observation, is that correct?
My intention is to use cross entropy loss, as that seems like the right one to define what i'm doing with my data (the labels are not discrete, and are real positive floats with a range, and there are 3 of them), so the output size should be 3. Given the optimizer tutorial, all i need to do is hand the loss function the output from the training step, and the correct label, and then backpropagate. Is that correct as well?
I know there's a lot here, so i'd appreciate answers to any of the questions from anyone inclined to help, even if you cannot answer all of them. Thank you for your time.
I've had a talk with a colleague and I've been able to answer some of my own questions, so I'll post those, since it may help others.
The hidden state zero-out is NOT modifying the hidden_layer, it is zero-ing out the hidden state at the start because the cells start empty, as you'd expect in any language object-oriented. It turns out this is unnecessary, since for quite some time the pytorch default is to zero out these values if they're not initialized manually.
This simple implementation will produce a hidden state as written.
The answer to this is yes. We don't "grade" the result of network until we get to the optimization section, or, more specifically, the loss function.
y is the true label, it was not identified in the tutorial. Also, important to note, pred is not a "prediction" but a pytorch object that points to the result of the network acting upon the observation that was fed in. In other words, printing out "pred" would not show you a vector of values that represents a prediction
This is also correct. Pytorch handles the "distance measure" between the true label and the predicted label, on its own.
Suppose I have my custom loss function and I want to fit the solution of some differential equation with help of my neural network. So in each forward pass, I am calculating the output of my neural net and then calculating the loss by taking the MSE with the expected equation to which I want to fit my perceptron.
Now my doubt is: should I use grad(loss) or should I do loss.backward() for backpropagation to calculate and update my gradients?
I understand that while using loss.backward() I have to wrap my tensors with Variable and have to set the requires_grad = True for the variables w.r.t which I want to take the gradient of my loss.
So my questions are :
Does grad(loss) also requires any such explicit parameter to identify the variables for gradient computation?
How does it actually compute the gradients?
Which approach is better?
what is the main difference between the two in a practical scenario.
It would be better if you could explain the practical implications of both approaches because whenever I try to find it online I am just bombarded with a lot of stuff that isn't much relevant to my project.
TLDR; Both are two different interfaces to perform gradient computation: torch.autograd.grad is non-mutable while torch.autograd.backward is.
Descriptions
The torch.autograd module is the automatic differentiation package for PyTorch. As described in the documentation it only requires minimal change to code base in order to be used:
you only need to declare Tensors for which gradients should be computed with the requires_grad=True keyword.
The two main functions torch.autograd provides for gradient computation are torch.autograd.backward and torch.autograd.grad:
torch.autograd.backward (source)
torch.autograd.grad (source)
Description
Computes the sum of gradients of given tensors with respect to graph leaves.
Computes and returns the sum of gradients of outputs with respect to the inputs.
Header
torch.autograd.backward( tensors, grad_tensors=None, retain_graph=None, create_graph=False, grad_variables=None, inputs=None)
torch.autograd.grad( outputs, inputs, grad_outputs=None, retain_graph=None, create_graph=False, only_inputs=True, allow_unused=False)
Parameters
- tensors – Tensors of which the derivative will be computed.- grad_tensors – The "vector" in the Jacobian-vector product, usually gradients w.r.t. each element of corresponding tensors.- retain_graph – If False, the graph used to compute the grad will be freed. [...] - inputs – Inputs w.r.t. which the gradient be will be accumulated into .grad. All other Tensors will be ignored. If not provided, the gradient is accumulated into all the leaf Tensors that were used [...].
- outputs – outputs of the differentiated function.- inputs – Inputs w.r.t. which the gradient will be returned (and not accumulated into .grad).- grad_tensors – The "vector" in the Jacobian-vector product, usually gradients w.r.t. each element of corresponding tensors.- retain_graph – If False, the graph used to compute the grad will be freed. [...].
Usage examples
In terms of high-level usage, you can look at torch.autograd.grad as a non-mutable function. As mentioned in the documentation table above, it will not accumulate the gradients on the grad attribute but instead return the computed partial derivatives. In contrast torch.autograd.backward will be able to mutate the tensors by updating the grad attribute of leaf nodes, the function won't return any value. In other words, the latter is more suitable when computing gradients for a large number of parameters.
In the following, we will take two inputs (x1 and, x2), calculate a tensor y with them, and then compute the partial derivatives of the result w.r.t both inputs, i.e. dL/dx1 and dL/dx2:
>>> x1 = torch.rand(1, requires_grad=True)
>>> x2 = torch.rand(1, requires_grad=True)
>>> x1, x2
(tensor(0.3939, grad_fn=<UnbindBackward>),
tensor(0.7965, grad_fn=<UnbindBackward>))
Inference:
>>> y = x1**2 + 5*x2
>>> y
tensor(4.1377, grad_fn=<AddBackward0>)
Since y was computed using tensor(s) requiring gradients (i.e. with requires_grad=True) - *outside of a torch.no_grad context. It will have a grad_fn function attached. This callback is used to backpropagate onto the computation graph to compute the gradients of preceding tensor nodes.
torch.autograd.grad:
Here we provide torch.ones_like(y) as the grad_outputs.
>>> torch.autograd.grad(y, (x1, x2), torch.ones_like(y))
(tensor(0.7879), tensor(5.))
The above output is a tuple containing the two partial derivatives w.r.t. to the provided inputs respectively in order of appearance, i.e. dL/dx1 and dL/dx2.
This corresponds to the following computation:
# dL/dx1 = dL/dy * dy/dx1 = grad_outputs # 2*x1
# dL/dx2 = dL/dy * dy/dx2 = grad_outputs # 5
torch.autograd.backward: in contrast it will mutate the provided tensors by updating the grad of the tensors which have been used to compute the output tensor and that require gradients. It is equivalent to the torch.Tensor.backward API. Here, we go through the same example by defining x1, x2, and y again. We call backward:
>>> # y.backward(torch.ones_like(y))
>>> torch.autograd.backward(y, torch.ones_like(y))
None
Then you can retrieve the gradients on x1.grad and x2.grad:
>>> x1.grad, x2.grad
(tensor(0.7879), tensor(5.))
In conclusion: both perform the same operation. They are two different interfaces to interact with the autograd library and perform gradient computations. The latter, torch.autograd.backward (equivalent to torch.Tensor.backward), is generally used in neural networks training loops to compute the partial derivative of the loss w.r.t each one of the model's parameters.
You can read more about how torch.autograd.grad works by reading through this other answer I made on: Meaning of grad_outputs in PyTorch's torch.autograd.grad.
In addition to Ivan's answer, having torch.autograd.grad not accumulating gradients into .grad can avoid racing conditions in multi-thread scenarios.
Quoting PyTorch doc https://pytorch.org/docs/stable/notes/autograd.html#non-determinism
If you are calling backward() on multiple thread concurrently but with shared inputs (i.e. Hogwild CPU training). Since parameters are automatically shared across threads, gradient accumulation might become non-deterministic on backward calls across threads, because two backward calls might access and try to accumulate the same .grad attribute. This is technically not safe, and it might result in racing condition and the result might be invalid to use.
But this is expected pattern if you are using the multithreading approach to drive the whole training process but using shared parameters, user who use multithreading should have the threading model in mind and should expect this to happen. User could use the functional API torch.autograd.grad() to calculate the gradients instead of backward() to avoid non-determinism.
implementation details https://github.com/pytorch/pytorch/blob/7e3a694b23b383e38f5e39ef960ba8f374d22404/torch/csrc/autograd/functions/accumulate_grad.h
I'm developing a machine learning model using keras and I notice that the available losses functions are not giving the best results on my test set.
I am using an Unet architecture, where I input a (16,16,3) image and the net also outputs a (16,16,3) picture (auto-encoder). I notice that maybe one way to improve the model would be if I used a loss function that compares pixel to pixel on the gradients (laplacian) between the net output and the ground truth. However, I did not found any tutorial that would handle this kind of application, because it would need to use opencv laplacian function on each output image from the net.
The loss function would be something like this:
def laplacian_loss(y_true, y_pred):
# y_true already is the calculated gradients, only needs to compute on the y_pred
# calculates the gradients for each predicted image
y_pred_lap = []
for img in y_pred:
laplacian = cv2.Laplacian( np.float64(img), cv2.CV_64F )
y_pred_lap.append( laplacian )
y_pred_lap = np.array(y_pred_lap)
# mean squared error, according to keras losses documentation
return K.mean(K.square(y_pred_lap - y_true), axis=-1)
Has anyone done something like that for loss calculation?
Given the code above, it seems that it would be equivalent to using a Lambda() layer as the output layer that applies that transformation in the image, before considering the mean square error.
Regardless as whether it is implemented as a Lambda() layer or in the loss function; the transformation needs to be such that Tensorflow understands how to calculate the gradients. The simplest was to do this would probably be to reimplement the cv2.Laplacian computation using Tensorflow math operations.
In order to use the cv2 library directly, you need to create a function that calculates the gradients for what happens inside the cv2 lib; that seems significantly more error prone.
Gradient descent optimisation relies on being able to compute gradients from the inputs to the loss; and back. Any operation in the middle must be differentiable; and Tensorflow must understand the math operations for auto differentiation to work; or you need to add them manually.
I managed to reach a easy solution. The main feature was that the gradient calculation is actually a 2D filter. For more information about it, please follow the link about the laplacian kernel. In that matter, is necessary that the output of my network be filtered by the laplacian kernel. For that, I created an extra convolutional layer with fixed weights, exactly as the laplacian kernel. After that, the network will have two outputs (one been the desired image, and the other been the gradient's image). So, is also necessary to define both losses.
To make it clearer, I'll exemplify. In the end of the network you'll have something like:
channels = 3 # number of channels of network output
lap = Conv2D(channels , (3,3), padding='same', name='laplacian') (net_output)
model = Model(inputs=[net_input], outputs=[net_out, lap])
Define how you want to calculate the losses for each output:
# losses for output, laplacian and gaussian
losses = {
"enhanced": "mse",
"laplacian": "mse"
}
lossWeights = {"enhanced": 1.0, "laplacian": 0.6}
Compile the model:
model.compile(optimizer=Adam(), loss=losses, loss_weights=lossWeights)
Define the laplacian kernel, apply its values in the weights of the above convolutional layer and set trainable equals False (so it won't be updated).
bias = np.asarray([0]*3)
# laplacian kernel
l = np.asarray([
[[[1,1,1],
[1,-8,1],
[1,1,1]
]]*channels
]*channels).astype(np.float32)
bias = np.asarray([0]*3).astype(np.float32)
wl = [l,bias]
model.get_layer('laplacian').set_weights(wl)
model.get_layer('laplacian').trainable = False
When training, remember that you need two values for the ground truth:
model.fit(x=X, y = {"out": y_out, "laplacian": y_lap})
Observation: Do not utilize the BatchNormalization layer! In case you use it, the weights in the laplacian layer will be updated!
I am attempting to implement a Lambda layer that will produce a custom loss function. In the layer, I need to be able to compare every element in a batch to every other element in the batch in order to calculate the cost. Ideally, I want code that looks something like this:
for el_1 in zip(y_pred, y_true):
for el_2 in zip(y_pred, y_true):
if el_1[1] == el_2[1]:
# Perform a calculation
else:
# Perform a different calculation
When I true this, I get:
TypeError: TensorType does not support iteration.
I am using Keras version 2.0.2 with a Theano version 0.9.0 backend. I understand that I need to use Keras tensor functions in order to do this, but I can't figure out any tensor functions that do what I want.
Also, I am having difficulty understanding precisely what my Lambda function should return. Is it a tensor of the total cost for each sample, or is it just a total cost for the batch?
I have been beating my head against this for days. Any help is deeply appreciated.
A tensor in Keras commonly has at least 2 dimensions, the batch and the neuron/unit/node/... dimension. A dense layer with 128 units trained with a batch size of 64 would therefore yields a tensor with shape (64,128).
Your LambdaLayer processes tensors as any other layer does, plugging it in after your dense layer from before will give you a tensor with shape (64,128) to process. Processing a tensor works similar to how calculations on numpy arrays works (or any other vector processing library really): you specify one operation to broadcast over all elements in the data structure.
For example, your custom cost is the difference for each value in the batch, you would implement it like so:
cost_layer = LambdaLayer(lambda a,b: a - b)
The - operation is broadcasted over a and b and will return a suitable result provided the dimensions match. The takeaway is that you really only can specify one operation for every value. If you want to do more complex tasks, for example computations based on the value you need single operations that take two operations and apply the correct one accordingly, i.e. the switch operation.
The syntax for K.switch is
K.switch(condition, then_expression, else_expression)
For example, if you want to subtract both values when a != b but add them when they are equal, you would write:
import keras.backend as K
cost_layer = LambdaLayer(lambda a,b: K.switch(a != b, a - b, a + b))