I have about a 30% and 70% for class 0 (minority class) and class 1 (majority class). Since I do not have a lot of data, I am planning to oversample the minority class to balance out the classes to become a 50-50 split. I was wondering if oversampling should be done before or after splitting my data into train and test sets. I have generally seen it done before splitting in online examples, like this:
df_class0 = train[train.predict_var == 0]
df_class1 = train[train.predict_var == 1]
df_class1_over = df_class1.sample(len(df_class0), replace=True)
df_over = pd.concat([df_class0, df_class1_over], axis=0)
However, wouldn't that mean that the test data will likely have duplicated samples from the training set (because we have oversampled the training set)? This means that testing performance wouldn't necessarily be on new, unseen data. I am fine doing this, but I would like to know what is considered good practice. Thank you!
I was wondering if oversampling should be done before or after splitting my data into train and test sets.
It should certainly be done after splitting, i.e. it should be applied only to your training set, and not to your validation and test ones; see also my related answer here.
I have generally seen it done before splitting in online examples, like this
From the code snippet you show, it is not at all obvious that it is done before splitting, as you claim. It depends on what exactly the train variable is here: if it is the product of a train-test split, then the oversampling takes place after splitting indeed, as it should be.
However, wouldn't that mean that the test data will likely have duplicated samples from the training set (because we have oversampled the training set)? This means that testing performance wouldn't necessarily be on new, unseen data.
Exactly, this is the reason why the oversampling should be done after splitting to train-test, and not before.
(I once witnessed a case where the modeller was struggling to understand why he was getting a ~ 100% test accuracy, much higher than his training one; turned out his initial dataset was full of duplicates -no class imbalance here, but the idea is similar- and several of these duplicates naturally ended up in his test set after the split, without of course being new or unseen data...).
I am fine doing this
You shouldn't :)
From my experience this is a bad practice. As you mentioned test data should contain unseen samples so it would not overfit and give you better evaluation of training process. If you need to increase sample sizes - think about data transformation possibilities.
E.g. human/cat image classification, as they are symmetric you can double sample size by mirroring images.
Related
I'm working with an extremelly unbalanced and heterogeneous multiclass {K = 16} database for research, with a small N ~= 250. For some labels the database has a sufficient amount of examples for supervised machine learning, but for others I have almost none. I'm also not in a position to expand my database for a number of reasons.
As a first approach I divided my database into training (80%) and test (20%) sets in a stratified way. On top of that, I applied several classification algorithms that provide some results. I applied this procedure over 500 stratified train/test sets (as each stratified sampling takes individuals randomly within each stratum), hoping to select an algorithm (model) that performed acceptably.
Because of my database, depending on the specific examples that are part of the train set, the performance on the test set varies greatly. I'm dealing with runs that have as high (for my application) as 82% accuracy and runs that have as low as 40%. The median over all runs is around 67% accuracy.
When facing this situation, I'm unsure on what is the standard procedure (if there is any) when selecting the best performing model. My rationale is that the 90% model may generalize better because the specific examples selected in the training set are be richer so that the test set is better classified. However, I'm fully aware of the possibility of the test set being composed of "simpler" cases that are easier to classify or the train set comprising all hard-to-classify cases.
Is there any standard procedure to select the best performing model considering that the distribution of examples in my train/test sets cause the results to vary greatly? Am I making a conceptual mistake somewhere? Do practitioners usually select the best performing model without any further exploration?
I don't like the idea of using the mean/median accuracy, as obviously some models generalize better than others, but I'm by no means an expert in the field.
Confusion matrix of the predicted label on the test set of one of the best cases:
Confusion matrix of the predicted label on the test set of one of the worst cases:
They both use the same algorithm and parameters.
Good Accuracy =/= Good Model
I want to firstly point out that a good accuracy on your test set need not equal a good model in general! This has (in your case) mainly to do with your extremely skewed distribution of samples.
Especially when doing a stratified split, and having one class dominatingly represented, you will likely get good results by simply predicting this one class over and over again.
A good way to see if this is happening is to look at a confusion matrix (better picture here) of your predictions.
If there is one class that seems to confuse other classes as well, that is an indicator for a bad model. I would argue that in your case it would be generally very hard to find a good model unless you do actively try to balance your classes more during training.
Use the power of Ensembles
Another idea is indeed to use ensembling over multiple models (in your case resulting from different splits), since it is assumed to generalize better.
Even if you might sacrifice a lot of accuracy on paper, I would bet that a confusion matrix of an ensemble is likely to look much better than the one of a single "high accuracy" model. Especially if you disregard the models that perform extremely poor (make sure that, again, the "poor" performance comes from an actual bad performance, and not just an unlucky split), I can see a very good generalization.
Try k-fold Cross-Validation
Another common technique is k-fold cross-validation. Instead of performing your evaluation on a single 80/20 split, you essentially divide your data in k equally large sets, and then always train on k-1 sets, while evaluating on the other set. You then not only get a feeling whether your split was reasonable (you usually get all the results for different splits in k-fold CV implementations, like the one from sklearn), but you also get an overall score that tells you the average of all folds.
Note that 5-fold CV would equal a split into 5 20% sets, so essentially what you are doing now, plus the "shuffling part".
CV is also a good way to deal with little training data, in settings where you have imbalanced classes, or where you generally want to make sure your model actually performs well.
I am experimenting with classification using neural networks (I am using tensorflow).
And unfortunately the training of my neural network gets stuck at 42% accuracy.
I have 4 classes, into which I try to classify the data.
And unfortunately, my data set is not well balanced, meaning that:
43% of the data belongs to class 1 (and yes, my network gets stuck predicting only this)
37% to class 2
13% to class 3
7% to class 4
The optimizer I am using is AdamOptimizer and the cost function is tf.nn.softmax_cross_entropy_with_logits.
I was wondering if the reason for my training getting stuck at 42% is really the fact that my data set is not well balanced, or because the nature of the data is really random, and there are really no patterns to be found.
Currently my NN consists of:
input layer
2 convolution layers
7 fully connected layers
output layer
I tried changing this structure of the network, but the result is always the same.
I also tried Support Vector Classification, and the result is pretty much the same, with small variations.
Did somebody else encounter similar problems?
Could anybody please provide me some hints how to get out of this issue?
Thanks,
Gerald
I will assume that you have already double, triple and quadruple checked that the data going in is matching what you expect.
The question is quite open-ended, and even a topic for research. But there are some things that can help.
In terms of better training, there's two normal ways in which people train neural networks with an unbalanced dataset.
Oversample the examples with lower frequency, such that the proportion of examples for each class that the network sees is equal. e.g. in every batch, enforce that 1/4 of the examples are from class 1, 1/4 from class 2, etc.
Weight the error for misclassifying each class by it's proportion. e.g. incorrectly classifying an example of class 1 is worth 100/43, while incorrectly classifying an example of class 4 is worth 100/7
That being said, if your learning rate is good, neural networks will often eventually (after many hours of just sitting there) jump out of only predicting for one class, but they still rarely end well with a badly skewed dataset.
If you want to know whether or not there are patterns in your data which can be determined, there is a simple way to do that.
Create a new dataset by randomly select elements from all of your classes such that you have an even number of all of them (i.e. if there's 700 examples of class 4, then construct a dataset by randomly selecting 700 examples from every class)
Then you can use all of your techniques on this new dataset.
Although, this paper suggests that even with random labels, it should be able to find some pattern that it understands.
Firstly you should check if your model is overfitting or underfitting, both of which could cause low accuracy. Check the accuracy of both training set and dev set, if accuracy on training set is much higher than dev/test set, the model may be overfiiting, and if accuracy on training set is as low as it on dev/test set, then it could be underfitting.
As for overfiiting, more data or simpler learning structures may work while make your structure more complex and longer training time may solve underfitting problem
I found that while training some CNNs and RNNs on imbalanced training data, that my training converges relatively quickly, with the accuracy being around the percentage of the bigger class (so e.g. if there are 80% yes examples it will probably always output yes). I find that explainable .. that this solution is a local optimum and the network cannot escape it while training. Is this explantion correct and this behaviour thus mostly found in these cases?
What can I do against it? Synthesize more training data to make the set more even? What else?
Thanks a lot!
Yes, you are right. Imbalanced training data do impacts the accuracy. Some of the solutions to come over imbalanced class problem are following:
1) More data collection: This is not easy in some cases. For example, there are a very small number of cases of frauds compared to non fraud cases.
2) Undersampling: Removing the data from the majority class. You can remove it randomly or informative (taking help from the distribution to decide what parts/patches to be removed)
3) Oversampling: Replicating observations belonging to minority class.
Your question has nothing to do with TF, this is a standard problem in machine learning. Just type "dealing with imbalanced data in machine learning" in google and read a few pages.
Here are a few approaches:
get more data
use other metric (f1)
undersampling/oversampling/weighting
I know you're supposed to separate your training data from your testing data, but when you make predictions with your model is it OK to use the entire data set?
I assume separating your training and testing data is valuable for assessing the accuracy and prediction strength of different models, but once you've chosen a model I can't think of any downsides to using the full data set for predictions.
You can use full data for prediction but better retain indexes of train and test data. Here are pros and cons of it:
Pro:
If you retain index of rows belonging to train and test data then you just need to predict once (and so time saving) to get all results. You can calculate performance indicators (R2/MAE/AUC/F1/precision/recall etc.) for train and test data separately after subsetting actual and predicted value using train and test set indexes.
Cons:
If you calculate performance indicator for entire data set (not clearly differentiating train and test using indexes) then you will have overly optimistic estimates. This happens because (having trained on train data) model gives good results of train data. Which depending of % split of train and test, will gives illusionary good performance indicator values.
Processing large test data at once may create memory bulge which is can result in crash in all-objects-in-memory languages like R.
In general, you're right - when you've finished selecting your model and tuning the parameters, you should use all of your data to actually build the model (exception below).
The reason for dividing data into train and test is that, without out-of-bag samples, high-variance algorithms will do better than low-variance ones, almost by definition. Consequently, it's necessary to split data into train and test parts for questions such as:
deciding whether kernel-SVR is better or worse than linear regression, for your data
tuning the parameters of kernel-SVR
However, once these questions are determined, then, in general, as long as your data is generated by the same process, the better predictions will be, and you should use all of it.
An exception is the case where the data is, say, non-stationary. Suppose you're training for the stock market, and you have data from 10 years ago. It is unclear that the process hasn't changed in the meantime. You might be harming your prediction, by including more data, in this case.
Yes, there are techniques for doing this, e.g. k-fold cross-validation:
One of the main reasons for using cross-validation instead of using the conventional validation (e.g. partitioning the data set into two sets of 70% for training and 30% for test) is that there is not enough data available to partition it into separate training and test sets without losing significant modelling or testing capability. In these cases, a fair way to properly estimate model prediction performance is to use cross-validation as a powerful general technique.
That said, there may not be a good reason for doing so if you have plenty of data, because it means that the model you're using hasn't actually been tested on real data. You're inferring that it probably will perform well, since models trained using the same methods on less data also performed well. That's not always a safe assumption. Machine learning algorithms can be sensitive in ways you wouldn't expect a priori. Unless you're very starved for data, there's really no reason for it.
I have a 6-dimensional training dataset where there is a perfect numeric attribute which separates all the training examples this way: if TIME<200 then the example belongs to class1, if TIME>=200 then example belongs to class2. J48 creates a tree with only 1 level and this attribute as the only node.
However, the test dataset does not follow this hypothesis and all the examples are missclassified. I'm having trouble figuring out whether this case is considered overfitting or not. I would say it is not as the dataset is that simple, but as far as I understood the definition of overfit, it implies a high fitting to the training data, and this I what I have. Any help?
However, the test dataset does not follow this hypothesis and all the examples are missclassified. I'm having trouble figuring out whether this case is considered overfitting or not. I would say it is not as the dataset is that simple, but as far as I understood the definition of overfit, it implies a high fitting to the training data, and this I what I have. Any help?
Usually great training score and bad testing means overfitting. But this assumes IID of the data, and you are clearly violating this assumption - your training data is completely different from the testing one (there is a clear rule for the training data which has no meaning for testing one). In other words - your train/test split is incorrect, or your whole problem does not follow basic assumptions of where to use statistical ml. Of course we often fit models without valid assumptions about the data, in your case - the most natural approach is to drop a feature which violates the assumption the most - the one used to construct the node. This kind of "expert decisions" should be done prior to building any classifier, you have to think about "what is different in test scenario as compared to training one" and remove things that show this difference - otherwise you have heavy skew in your data collection, thus statistical methods will fail.
Yes, it is an overfit. The first rule in creating a training set is to make it look as much like any other set as possible. Your training set is clearly different than any other. It has the answer embedded within it while your test set doesn't. Any learning algorithm will likely find the correlation to the answer and use it and, just like the J48 algorithm, will regard the other variables as noise. The software equivalent of Clever Hans.
You can overcome this by either removing the variable or by training on a set drawn randomly from the entire available set. However, since you know that there is a subset with an embedded major hint, you should remove the hint.
You're lucky. At times these hints can be quite subtle which you won't discover until you start applying the model to future data.