I am building an optimizer (https://github.com/keras-team/keras/blob/master/keras/optimizers.py) which calculates a search direction and then tries a few different step lengths to find which gives the lowest loss. However, I am running into problems when trying to change the step length depending on the value of the loss itself. It appears that the loss (which is a tensor dependant upon the weight of the network and the data) cannot be updated/recalculated more than once during each training loop, which I find very odd.
This is the relevant code I have in get_updates(self, loss, params):
L1 = loss
for p, direction in zip(params, directions):
self.updates.append(K.update(p, p+length*direction))
L2 = loss
for p, direction in zip(params, directions):
self.updates.append(K.update(p, tf.cond( L2<L1, lambda: p+0.5*length*direction, lambda: p))
The problem is that L1 and L2 are the same and no matter what I try I can't get the loss to update after I've updated the weights. I've also tried just p = p+length*direction and p.assign() but the loss doesn't update. Does anyone know how I can get an updated value of the loss?
Note, I am able to get the loss from the previous batch/epoch if I save a loss value and update using self.updates.append(K.update(self.prev_loss,loss)), however since the data will change between batches I am no longer working on the same loss function and thus my comparison between the losses to determine if the step length should be lower is not valid.
Related
I am training an autoencoder whose input is a matrix P in [0,1] using the following loss function:
And here is my code:
# Define loss and optimizer, minimize the squared error
with tf.device("/device:GPU:0"):
L = -tf.reduce_sum(self.p*tf.math.log(self.p_pred+1e-10) + 0.55*(1 - self.p)*tf.math.log(1-self.p_pred+1e-10), axis = 1)
self.loss = tf.reduce_mean(L)
self.optimizer = tf.compat.v1.train.AdamOptimizer(self.learning_rate).minimize(self.loss)
In main.py, I run session like this:
_, l, y_pred, y = sess.run([model.optimizer, model.loss, model.y_pred, model.y], feed_dict=...)
But the loss returns nan at different epoch whenever I'm training. The activation function is sigmoid and learning_rate = 0.01. The number of epoch is 20. I'm trying to save p and p_pred when the loss is nan, then I run the same loss function in google colab, the result is not nan! I don't understand.
Any ideas for what I did wrong?
The loss function you're using is BCELoss which gives "NaN" when the value inside log is less than or equal to 0. Sometimes, your term inside the log might go <=0 and sometimes, it might not. That's why you might not be getting NaN in colab.
Try adding sigmoid in the loss function (like sigmoid(self.p_pred) instead of self.p_pred. This will ensure the term is in (0,1) range.
Else, try increasing the epsilon value (1e-10 in your code. Increase that value to 1e-5 may be)
I want to calculate L1 loss in a neural network, I came across this example at https://discuss.pytorch.org/t/simple-l2-regularization/139/2, but there are some errors in this code.
Is this really how to calculate L1 Loss in a NN or is there a simpler way?
l1_crit = nn.L1Loss()
reg_loss = 0
for param in model.parameters():
reg_loss += l1_crit(param)
factor = 0.0005
loss += factor * reg_loss
Is this equivalent in any way to simple doing:
loss = torch.nn.L1Loss()
I assume not, because I am not passing along any network parameters. Just checking if there isn existing function to do this.
If I am understanding well, you want to compute the L1 loss of your model (as you say in the begining). However I think you might got confused with the discussion in the pytorch forum.
From what I understand, in the Pytorch forums, and the code you posted, the author is trying to normalize the network weights with L1 regularization. So it is trying to enforce that weights values fall in a sensible range (not too big, not too small). That is weights normalization using L1 normalization (that is why it is using model.parameters()). Normalization takes a value as input and produces a normalized value as output.
Check this for weights normalization: https://pytorch.org/docs/master/generated/torch.nn.utils.weight_norm.html
On the other hand, L1 Loss it is just a way to determine how 2 values differ from each other, so the "loss" is just measure of this difference. In the case of L1 Loss this error is computed with the Mean Absolute Error loss = |x-y| where x and y are the values to compare. So error compute takes 2 values as input and produces a value as output.
Check this for loss computing: https://pytorch.org/docs/master/generated/torch.nn.L1Loss.html
To answer your question: no, the above snippets are not equivalent, since the first is trying to do weights normalization and the second one, you are trying to compute a loss. This would be the loss computing with some context:
sample, target = dataset[i]
target_predicted = model(sample)
loss = torch.nn.L1Loss()
loss_value = loss(target, target_predicted)
Is there a difference between these two codes?
1
Loss.backward(retain_graph=True)
Loss.backward(retain_graph=True)
Loss.backward()
optimizer.step
2
Loss = 3 * Loss
Loss.backward()
optimizer.step
When I checked the gradient of the parameter after the last backward(), there was no difference between the two codes. However, there is a little difference in test accuracy after training.
I know this is not a common case, but it is related to the research I'm doing.
To me, it looks very different.
Computing the loss three time won't do anything (first code snippet). You just hold on to the gradient you have previously calculated. (Check on your leaf tensors the value of the .grad() attribute).
However, the second code snippet with just multiply the gradients by three, thus speeding up Gradient Descent. For a standard Gradient descent optimizer, it would be like mutliplying the learning rate by 3.
Hope this helps.
In option 1, every time you call .backward(), gradients are computed. After 3 calls, when you perform optimizer.step, the gradients are added and then weights are updated accordingly.
In option 2, you multiply the loss with a constant, so the gradients will be multiplied with that constant too.
So, adding a gradient value 3 times and multiplying the gradient value by 3 would result in the same parameter update.
Please note, I assume there is no loss due to floating point precision (as noted in the comments).
I'm following Tensorflow's Neural Machine Translation with Attention tutorial (link) but am unclear about some implementation details. It'd be great if someone could help clarify or refer me to a source/better place to ask:
1) def loss_function(real, pred): This function computes loss at a specific time step (say t), averaged over the entire the batch. Examples whose labels at t is <pad> (i.e. no real data, only padded so that all example sequences are of same length) are masked so as not to count towards loss.
My question: It seems loss should get smaller the bigger t is (since more examples are <pad> the further we get to maximum length). So why is loss averaged over the entire batch, and not just over the number of valid (non-<pad>) examples? (This is analogous to using tf.losses.Reduction.SUM_BY_NONZERO_WEIGHTS instead of tf.losses.Reduction.SUM_OVER_BATCH_SIZE)
2) for epoch in range(EPOCHS) ——> Two loss variables are defined in the training loop:
loss = sum of loss_function() outputs over all time steps
batch_loss = loss divided by number of time steps
My question: Why are gradients computed w.r.t. loss and not batch_loss? Shouldn't batch_loss be the average loss over all time steps and the entire batch?
Many thanks!
It seems loss should get smaller the bigger t
The loss does get smaller since the pad token is getting masked while calculating the loss.
Batch_loss is used only to print the loss calculated of each batch. Batch loss is calculated for every batch and across all the time steps.
for t in range(1, targ.shape[1])
This loop runs over the batch for all timesteps and calculates the loss by masking the padded values.
I hope this clears it up :)
Why does zero_grad() need to be called during training?
| zero_grad(self)
| Sets gradients of all model parameters to zero.
In PyTorch, for every mini-batch during the training phase, we typically want to explicitly set the gradients to zero before starting to do backpropragation (i.e., updating the Weights and biases) because PyTorch accumulates the gradients on subsequent backward passes. This accumulating behaviour is convenient while training RNNs or when we want to compute the gradient of the loss summed over multiple mini-batches. So, the default action has been set to accumulate (i.e. sum) the gradients on every loss.backward() call.
Because of this, when you start your training loop, ideally you should zero out the gradients so that you do the parameter update correctly. Otherwise, the gradient would be a combination of the old gradient, which you have already used to update your model parameters, and the newly-computed gradient. It would therefore point in some other direction than the intended direction towards the minimum (or maximum, in case of maximization objectives).
Here is a simple example:
import torch
from torch.autograd import Variable
import torch.optim as optim
def linear_model(x, W, b):
return torch.matmul(x, W) + b
data, targets = ...
W = Variable(torch.randn(4, 3), requires_grad=True)
b = Variable(torch.randn(3), requires_grad=True)
optimizer = optim.Adam([W, b])
for sample, target in zip(data, targets):
# clear out the gradients of all Variables
# in this optimizer (i.e. W, b)
optimizer.zero_grad()
output = linear_model(sample, W, b)
loss = (output - target) ** 2
loss.backward()
optimizer.step()
Alternatively, if you're doing a vanilla gradient descent, then:
W = Variable(torch.randn(4, 3), requires_grad=True)
b = Variable(torch.randn(3), requires_grad=True)
for sample, target in zip(data, targets):
# clear out the gradients of Variables
# (i.e. W, b)
W.grad.data.zero_()
b.grad.data.zero_()
output = linear_model(sample, W, b)
loss = (output - target) ** 2
loss.backward()
W -= learning_rate * W.grad.data
b -= learning_rate * b.grad.data
Note:
The accumulation (i.e., sum) of gradients happens when .backward() is called on the loss tensor.
As of v1.7.0, Pytorch offers the option to reset the gradients to None optimizer.zero_grad(set_to_none=True) instead of filling them with a tensor of zeroes. The docs claim that this setting reduces memory requirements and slightly improves performance, but might be error-prone if not handled carefully.
Although the idea can be derived from the chosen answer, but I feel like I want to write that explicitly.
Being able to decide when to call optimizer.zero_grad() and optimizer.step() provides more freedom on how gradient is accumulated and applied by the optimizer in the training loop. This is crucial when the model or input data is big and one actual training batch do not fit in to the gpu card.
Here in this example from google-research, there are two arguments, named train_batch_size and gradient_accumulation_steps.
train_batch_size is the batch size for the forward pass, following the loss.backward(). This is limited by the gpu memory.
gradient_accumulation_steps is the actual training batch size, where loss from multiple forward pass is accumulated. This is NOT limited by the gpu memory.
From this example, you can see how optimizer.zero_grad() may followed by optimizer.step() but NOT loss.backward(). loss.backward() is invoked in every single iteration (line 216) but optimizer.zero_grad() and optimizer.step() is only invoked when the number of accumulated train batch equals the gradient_accumulation_steps (line 227 inside the if block in line 219)
https://github.com/google-research/xtreme/blob/master/third_party/run_classify.py
Also someone is asking about equivalent method in TensorFlow. I guess tf.GradientTape serve the same purpose.
(I am still new to AI library, please correct me if anything I said is wrong)
zero_grad() restarts looping without losses from the last step if you use the gradient method for decreasing the error (or losses).
If you do not use zero_grad() the loss will increase not decrease as required.
For example:
If you use zero_grad() you will get the following output:
model training loss is 1.5
model training loss is 1.4
model training loss is 1.3
model training loss is 1.2
If you do not use zero_grad() you will get the following output:
model training loss is 1.4
model training loss is 1.9
model training loss is 2
model training loss is 2.8
model training loss is 3.5
You don't have to call grad_zero() alternatively one can decay the gradients for example:
optimizer = some_pytorch_optimizer
# decay the grads :
for group in optimizer.param_groups:
for p in group['params']:
if p.grad is not None:
''' original code from git:
if set_to_none:
p.grad = None
else:
if p.grad.grad_fn is not None:
p.grad.detach_()
else:
p.grad.requires_grad_(False)
p.grad.zero_()
'''
p.grad = p.grad / 2
this way the learning is much more continues
During the feed forward propagation the weights are assigned to inputs and after the 1st iteration the weights are initialized what the model has learnt seeing the samples(inputs). And when we start back propagation we want to update weights in order to get minimum loss of our cost function. So we clear off our previous weights in order to obtained more better weights. This we keep doing in training and we do not perform this in testing because we have got the weights in training time which is best fitted in our data. Hope this would clear more!
In simple terms We need ZERO_GRAD
because when we start a training loop we do not want past gardients or past results to interfere with our current results beacuse how PyTorch works as it collects/accumulates the gradients on backpropagation and if the past results may mixup and give us the wrong results so we set the gradient to zero every time we go through the loop.
Here is a example:
`
# let us write a training loop
torch.manual_seed(42)
epochs = 200
for epoch in range(epochs):
model_1.train()
y_pred = model_1(X_train)
loss = loss_fn(y_pred,y_train)
optimizer.zero_grad()
loss.backward()
optimizer.step()
`
In this for loop if we do not set the optimizer to zero every time the past value it may get add up and changes the result.
So we use zero_grad to not face the wrong accumulated results.