If Y_pred is very far off from Y, the Loss value will be very high. However, if both values are almost similar, the Loss value will be very low. Hence we need to keep a loss function which can penalize a model effectively while it is training on a dataset.
When a neural network is trying to predict a discrete value, we can consider it to be a classification model. This could be a network trying to predict what kind of animal is present in an image, or whether an email is a spam or not.
Questions tagged [loss-function]
1727 questions
0
votes
0 answers
Doing regression with Caffe using EuclideanLoss layer
I'm using the Caffenet model. I did a classification task with 9 classes successfully. Then I tried to change it to a regression network by preparing another LMDB file with labels ranging from 700 to 1400. I changed the original training code and…
behnam
- 11
- 2
0
votes
1 answer
How does xgboost split root node and question for Taylor expansion
I know xgboost use Gain = Score(L)+Score(R)-Score(L+R) to split node, but how does xgboost split root node? Also, why not use the fourth or fifth derivative in Taylor expansion for loss function?
tiezhuetc
- 41
- 1
- 7
0
votes
0 answers
Keras in_top_k loss
I'm using Keras (Tensorflow backend) to build a model that given an input predicts a single class (out of 64 classes), a multiclass model. Given the pretty large number of classes, I do not want to use the categorical_crossentropy or the…
Daniel Juravski
- 181
- 1
- 2
- 12
0
votes
0 answers
How to output the loss gradient backpropagation path through a PyTorch computational graph
I have implemented a new loss function in PyTorch.
#model_1 needs to be trained
outputs = model_1(input)
loss = myloss(outputs,labels)
#output is how much to resize an image
#label give the image file index
#Below I am explaining myloss()…
MSD Paul
- 1,648
- 3
- 13
- 31
0
votes
0 answers
Pytorch Autoencoder - How to improve loss?
I've a UNET style autoencoder below, with a filter I wrote in Pytorch at the end. The network seems to be converging faster than it should and I don't know why. I have a dataset of 4000 images and I'm taking a 128x128 crop every time. I'm employing…
Bled Clement
- 170
- 2
- 9
0
votes
0 answers
builtin_function_or_method' object has no attribute 'size'
optimizer = optim.SGD(model.parameters(), lr = lr)
criterion = nn.MSELoss()
valid_loss_min = np.Inf
def train(model, device, train_loader, optimizer, epoch):
model.train()
for batch_idx, (img_data32,img_data64, target) in…
Vahid Zarghami
- 1
- 3
0
votes
1 answer
Understanding CNN training results
I would appreciate your explanation on following :)
I trained CNN network that classifies TWO image classes.
I used 'SGD' optimizer and 'categorical_crossentropy' loss function.
My results are as follows:
- training loss = 0.28
- training accuracy =…
Ivan Geek
- 1
- 1
- 1
0
votes
1 answer
validation loss decreases dramatically in CNN, model fit or overfit?
I have problem with my CNN model.
I have 89 original fundus image with 5 images of a normal class, and 84 images un-normal class. Then, I augmented the normal class with OpenCV, so I have 85 images of normal class and 84 un-normal class.
I trained…
hilyap
- 1
- 2
0
votes
0 answers
Python Keras - Accuracy drops to zero
I have a problem, when training a U-Net, which has many similarities with a CNN, in Keras with Tensorflow. When starting the Training, the Accuracy increases and the loss steadily goes down. At around epoch 40, in my example, the validation loss…
Jacob
- 1
- 1
0
votes
1 answer
How to compile a keras model that has 2 outputs with a custom loss that takes 3 parameters?
I am trying to compile a model with 2 outputs using a custom loss function but I am failing at doing so.
Any ideas? Let me show you what I have done,
Here is the loss function:
def contrastive_loss(y_true, y_pred1, y_pred2):
'''Contrastive loss…
mj1261829
- 1,200
- 3
- 26
- 53
0
votes
1 answer
Difference between Logistic Loss and Cross Entropy Loss
I'm confused about logistic loss and cross entropy loss in binary classification scenario.
According to Wikipedia (https://en.wikipedia.org/wiki/Loss_functions_for_classification), the logistic loss is defined as:
where v=y*y_hat
The cross entropy…
YQ.Wang
- 1,090
- 1
- 17
- 43
0
votes
1 answer
PyTorch: Simple feedforward neural network not running without retain_graph=True
Below is my code for training a Feedforward neural network (FFNN).
The labels are numbers between 0 and 50. The FFNN comprises of a single hidden layer with 50 neurons and an output layer with 51 neurons. Furthermore, I have used negative log…
CS1999
- 23
- 5
0
votes
0 answers
Cross entropy loss returns infinity in neural network
I am implementing a neural network and using a cross-entropy loss function.
Cross entropy error is given by:
error = - np.sum((actual_Y * np.log2 (Y_pred)) + ((1-actual_Y)* np.log2 (1 - Y_pred)))
after few iterations (1- Y_pred) inside…
Irfan Umar
- 196
- 3
- 18
0
votes
1 answer
how important is the loss difference between training and validation data at the beginning when training a neuronal network?
Short question:
Is the difference between validation and training loss at the beginning of the training (first epochs) a good indicator for the amount of data that should be used?
E.g would it be a good method to increase the amount of data until…
Khan
- 1,418
- 1
- 25
- 49
0
votes
1 answer
SSD Resnet 50 FPN Loss function clarification
I am using tensorflow object detection api on my dataset. I am using ssd-resnet50-fpn model. While training, I see that classification loss and localization loss has converged but the total loss is still decreasing. Also total loss is not coming out…
Aashish Kumar
- 21
- 2