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 the difference at the beginning is as small as possible? It would save me time and computation.
backround:
I am working on a neuronal network that overfits very fast. The best result after applying many different techniques like dropouts, batch normalization, reducing learning rate, reducing batch size, increasing variety of data, reducing layers, increasing filter sizes ..... is still very bad.
While the training loss decreases very well, validation loss overfits too early(with too early I mean, the desired loss is not reached, it should be many times less)
Since the training with my dataset ~200 samples took 24 hours for 50 epochs, I was hoping to find a way to fight against overfitting with all the methods I described above, before increasing the amount of data. Because nothing helped I am at the point of increasing the amount of data.
I am thinking about how much data could be enough for my network to eliminate overfitting. I know that this it is not easy to answer because it depends on the complexity of the data and the task I am trying to solve.. therefore I try to generalize my question to:
Short answer to the short question: No
explanation: There's a correlation between (train_loss - val_loss) and the amount of data you need to train your model, but there's a bunch of other factors that could be the source of the big (train_loss - val_loss). For example, your network architecture is too small, and therefor your model quickly overfits. Or, your validation set doesn't reflect the training data. Or your learning rate is too big. Or...
So my recommendation: formulate your problem in another SO question, and ask "what might I be doing wrong?"
Related
I am working on deep learning model to detect regions of timesteps with anomalies. This model should classify each timestep as possessing the anomaly or not.
My labels are something like this:
labels = [0 0 0 1 0 0 0 0 1 0 0 0 ...]
The 0s represent 'normal' timesteps and the 1s represent the existence of an anomaly. In reality, my dataset is very very imbalanced:
My training set consists of over 7000 samples, where only 1400 samples = 20% of those contain at least 1 anomaly (timestep = 1)
I am feeding samples with 4096 timesteps each. The average number of anomalies, in the samples that contain them, is around 2. So, assuming there is an anomaly, the % of anomalous timesteps ranges from 0.02% to 0.04% for each sample.
With that said, I do need to shift from the standard binary cross entropy to something that highlights the anomalous timesteps from the anomaly free timesteps.
So, I experimented adding weights to the anomalous class in such a way that the model is forced to learn from the anomalies and not just reduce its loss from the anomaly-free timesteps. It actually worked well and the model seems to learn to detect anomalous timesteps. One problem however is that training can become quite unstable (and unpredictable), with sudden loss spikes appearing and affecting the learning process. Below, you can see the effects on the loss and metrics charts for two of my trainings:
After going through a debugging process for the trainings, I am confident that the problem comes from ocasional predictions given for the anomalous timesteps. That is, in some samples of a certain epoch, and in some anomalous timesteps, the model is giving a very low prediction, e.g. 0.01, for the 1s label (should be close to 1 ofc). Considering the very high (but supposedly necessary) weights given to the anomalous timesteps, the penaly is really extreme and the loss just skyrockets.
Going deeper, if I inspect the losses of the sample where the jump happened and look for the batch right before the loss jumped, I see that the losses are all around 10^-2 - 0.0053, 0.004, 0.0041... - not a single sample with a loss over those values. Overall, an average loss of 0.005. However, if I inspect the loss of the following batch, in that same sample, the avg. loss of the batch is already 3.6, with a part of the samples with a low loss but another part with a very high loss - e.g. 9.2, 7.7, 8.9... I can confirm that all the high losses come from the penalties given at predicting the 1s timesteps. The following batches of the same sample and some of the batches of the next epoch get affected and take some time to start decreasing again and going back to a stable learning process.
With this said, I am having this problem for some weeks already and really need some guidance in what I could try to deal with the spikes, which I assume that arise on the gradient updates associated with anomalous timesteps that are harder to learn.
I am currently using a simple 2-layer keras LSTM model with 64 units each and a dense as the last layer with a 1 unit dense layer with sigmoid activation. As for the optimizer I am using Adam. I am training with batch size 128. Some things to consider also:
I have tried changes in weights and other loss functions. Ultimately, if I reduce the weights given to the anomalous timesteps the model doesn't give so much importance to them and the loss reduces by considering only the anomalous free timesteps. I have also considered focal binary cross entropy loss but it doesn't seem to do anything that could avoid those jumps as, in the end, it is all about adding or reducing weights for certain timesteps.
My current learning rate is the Adam's default, 10⁻3. I have tried reducing the learning rate which leads to less impactful spikes (they're still there though) but the model also takes much more time or gets stuck. Not sure if it would be the way to go in this case, as the training seems to go well except for these cases. Decaying learning rate might also not make too much sense as the spikes can happen earlier in the training and not only on later epochs. Not sure if this is the way to go.
I am still investigating gradient clipping as a solution. I am still not sure on what values to use and if it is actually an effective solution for my case, but from what I understood of it, it should allow to counter those jumps resulting from those 'almost' exploding gradients.
The spikes could originate from sample noise / bad samples. However, since I am already using batch size 128 and I have already tested training with simple synthetic samples I have created and the spikes were still there, I guess it is not a problem with specific samples.
The imbalance obviously plays the bigger role here. Not sure if undersampling the majority class of samples of 4096 timesteps (like increasing from 20% to 50% the amount of samples with at least an anomalous timestep) would make a big difference here since each sample of timesteps is by itself very imbalanced as it contains around 2 timesteps with anomalies. It is a problem with the imbalance within each sample.
I know it might be quite some context but honestly I am already into my limit of trying stuff for weeks.
The solutions I am inclined to go for next are either gradient clipping or just changing my samples to be more centered around the anomalous timesteps, in such a way that it contains less anomaly free timesteps and hopefully allows for convergence without having to apply such drastic weights to anomalous timesteps. This last option is more difficult for me to opt for due to some restrictions, but I might look at it if I have nothing else available.
What do you think? I am able to provide more information if needed.
I have some data (reuters).
I combine the training and test data into one large data set.
I shuffle the entire data set.
I then split the data set into 50%. 50% for training and 50% for testing.
I then setup a sequential model (Embedding, GRU, BatchNormalization, Dense).
I compile with adam optimizer, and loss = sparse_categorical_crossentropy
I run fit on it, for 5 epochs, with validation_split = 0.2
I get a val_acc of 95%. I am very happy with this.
I then call evaluate with the test data, and get acc = 71%
When I repeat evaluate with the train data (instead of test data), just to see what happens, I get 95% acc (as I should).
I am trying to understand what is wrong with my acc score with the test data.
The data is shuffled between train and test each time, so it can not be related to the data.
I tried the checkpoint save/restore trick, that did not seem to help.
I have reduced epochs in case of overfitting, but this has not helped.
I am curious what is going on. The only thing I can think of is there is some sort of overfitting going on that is carrying over to test, but is NOT impacting val_acc.
(Side note: the 20% data saved for validation during fit should not cause an overfitting problem, correct?)
Any ideas what could be wrong with my approach?
Thank you.
Edit 1:
Okey, I might be onto something. I noticed something odd about the test data that is included with the reuters dataset, vs the training data.
In previous experiments, results were always poor evaluating against test set, vs an unused portion of the training data.
This time, I combined the training and test data sets into one set, and shuffled.
Then shuffled some more, and then generated pseudo data around it, and shuffled some more.
Then I peeled off 20% to use as test data.
This time it worked. I got 95% for training, validation and evaluation accuracy.
I tried searching for information on the test set to see if anyone came up with anything but found no results. So I'm going to presently chalk it up to test data that is significantly different from the training set.
Edit 2:
Nevermind edit 1. I think I was corrupting my test data with pseudo-generated version of training data, that was close enough to work.
The only conclusion I can draw is that there is a lot of overfitting going on during my training AND using validation data during training is misleading, as it is also being overfit.
(Why validation data is being used to help overfit, I do not know).
Overfitting occurs due to many reasons, to overcome this problem you can try different methods like:
L1/L2 regularization.
Dropout layers: By using dropout layers in the network, we ignore a subset of units of our network with a set probability. Using dropout, we can reduce interdependent learning among units, which may have led to overfitting.
Early stopping: Monitors the performance of the model for every epoch on a held-out validation set during the training, and terminates the training conditional on the validation performance.
Feature selection: Improves the machine learning process and increases the predictive power of machine learning algorithms by selecting the most important variables and eliminating unnecessary and irrelevant features.
For more details refer to this link.
I am working with a large dataset (e.g. large for a single machine) - with 1,000,000 examples.
I split my dataset to as follows: (80% Training Data, 10% Validation Data, 10% Testing Data). Every time I retrain the model, I shuffle the data first - such that some of the data from the validation / testing set ends up into the training set and vice versa.)
My thinking is this:
Ideally I would want all possible available data for the model to learn. The more the better - for improved accuracy.
Even though 20% of the data is dedicated to validation and testing, that is still 100,000 examples per piece - (i.e. I may potentially miss out on some crucial data that exists within the validation or testing set that the previous training set may not have accounted for.)
Shuffling prevents the training set from learning order where it is not important (at least in my particular dataset).
Here is my workflow process:
The Test Accuracy is more or less the equivalent to the Validation Accuracy (plus or minus 0.5%)
Per each retrain, the results usually ends up something like this: where the accuracy keeps improving (until it runs out of total epoch), but the validation accuracy ends up stuck at a particular percentage. I then save that model. Start the retraining process again. Shuffles data occurs. The training accuracy drops, but validation accuracy jumps up. The training accuracy improves until total epoch. The validation accuracy, converges downward (still greater than the previous run).
See Example:
I plan on doing this until the training accuracy data reaches 99%. (Note: I used Keras-Tuner to find the best architecture/model for my particular problem)
I can't help but think, that I am doing something wrong by doing this. From my perspective, this is just the model eventually learning all 1,000,000 examples. It feels like "mild overfitting" because of the shuffling per each retrain.
Is it a good idea to mix the validation / testing data with the training data?
Am I wrong by doing it this way? If so, why should I not do this method? Is there a better way to approach this?
If you mix your test/validation data with training data, you then can not evaluate your model on that data, since that data has been seen by your model. The model evaluation is done on the basis of how well it is able to make predictions/classification on data which your model has not seen (assuming that the data you are using to evaluate your model is coming from the same distribution as your training data). If you also mix your test set data with training set data, you will eventually end up with really good test set accuracy since that data has been seen by your model, but it might not perform well on new unseen data coming from the same distribution.
If you are worried size of test/validation data, I suggest you further reduce the size of your test/validation data. Use 99.9% instead of 99%. Also, the random shuffling will take care of learning almost every feature of your data.
After all, my point is, never ever evaluate your model on the data it has seen before. It will always give you better results (assuming you have trained your model well untill it memorizes the training data). The validation data is used when you have multiple algorithms/models and you need to select one algorithm/model from all those available models. Here, the validation data is used to select the model. The algo/model which gives good results on validation data is selected (again you do not evaluate your model based on validation set accuracy, it is just used for the selection of the model.) Once you have selected your model based on validation set accuracy, you then evaluate it on new unseen data (called test data) and report the prediction/classification accuracy on test data as your model accuracy.
I am not completely familiar with batch normalization layers,. As I understand it, it is going to compute normalization at training time using mini-batch statistics.
Do any of you have experience using thesen layers when the minibatch size is very small (for example using 2 or 4 images per iteration for the minibatch size) ? Is there any reason for it not to work efficiently ?
My feeling would be that the statistics is computed on a very small sample at training time, and could negativaly affect the training, what do you think ?
You are right in your intuition that the samples might be different from the population (mini-batch vs all samples), but this problem was addressed in the batch normalization paper. Specifically, during train time, you find the variance of your samples by dividing with the batch size (N), but during test time you account for this by using the unbiased variance estimate (multiplication by N/(N-1)):
Have a look here for a more detailed and easy to understand explanation:
Batch Normalization
I have a multilayer perceptron with 5 hidden layers and 256 neurons each. When I start training, I get different prediction probabilities for each train sample until epoch 50, but then the number of duplicate predictions increases, on epoch 300 I already have 30% of duplicate predictions which does not make sense since the input data is different for all training samples. Any idea what causes this behavior?
Clarifications:
with "duplicate predictions", I mean items with the exactly same predicted probability to belong to class A (it's a binary classification problem)
I have 4000 training samples with 200 features each and all samples are different, it does not make sense that the number of duplicate predictions increases to 30% while training. So I wonder what can cause this behavior.
One point, you say you are doing a binary prediction, and when you say "duplicate predictions", even with your clarification it's hard to understand your meaning. I am guessing that you have two outputs for your binary classifier, one for class A and one for class B and you are getting roughly the same value for a given sample. If that's the case, then the first thing to do is to use 1 output. A binary classification problem is better modeled with 1 output that ranges between 0 and 1 (sigmoid the output neuron). This way there will be no ambiguity, the network will have to choose one or the other, or when it's confused you'll get ~0.5 and it will be clear.
Second, it is very common for a network to start learning well and then to perform more poorly after overtraining. Especially with small datasets such as what you have. In fact, even with the little knowledge I have of your dataset I would put a small bet on you getting better performance out of an algorithm like XGA Boost than a neural network (I assume you're using a neural net and not literally a perceptron).
But regarding the performance degrading over time. When this happens you want to look into something called "early stopping". At some point the network will start memorizing the input, and may be part of what's happening. Essentially you train until the performance on your held out test data starts to worsen.
To address this you can apply various forms of regularization (L2 regularization, dropout, batch normalization all come to mind). You can also reduce the size of your network. 5 layers of 256 neurons sounds too big for the problem. Try trimming this down and I bet your results will improve. There is a sweet spot for architecture size in neural networks. When your network is too large it can, and often will, over fit. When it's too small it won't be expressive enough for the data. Angrew Ng's coursera class has some helpful practical advice on dealing with this.