data-augmentation | data augmentation on python | Wrapper library

How should I implement an augmentation pipeline in which my dataset gets extended instead of replacing the images with the augmented ones, that means, how to use map calls to augment and preserve the original samples?

threads I've checked: 1, 2

I am trying to create a binary CNN classifier for an unbalanced dataset (class 0 = 4000 images, class 1 = around 250 images), which I want to perform 5-fold cross validation on. Currently I am loading my training set into an ImageLoader that applies my transformations/augmentations(?) and loads it into a DataLoader. However, this results in both my training splits and validation splits containing the augmented data.

I originally applied transformations offline (offline augmentation?) to balance my dataset, but from this thread (https://stats.stackexchange.com/questions/175504/how-to-do-data-augmentation-and-train-validate-split), it seems it would be ideal to only augment the training set. I would also prefer to train my model on solely augmented training data and then validate it on non-augmented data in a 5-fold cross validation

My data is organized as root/label/images, where there are 2 label folders (0 and 1) and images sorted into the respective labels.

I would like to use a kind of data-augmentation method for NLP consisting of back-translating dataset.

Basically, I have a large dataset (SNLI), consisting of 1 100 000 english sentences. What I need to do is : translate these sentences in a language, and translate it back to English.

I may have to do this for several language. So I have a lot of translations to do.

I need a free solution.

I tried several python module for translation, but due to recent changes in Google Translate API, most of them do not work. googletrans seems to work if we apply this solution.

However, it is not working for big dataset. There is a limit of 15K characters by Google (as pointed out by this, this and this). The first link show a supposed work-around.

Even if I apply the work-around (initializing the Translator every iteration), it is not working, and I got the following error :

I want to create a model using the Functional Keras API that will have two inputs and two outputs. The model will be using two instances of the ImageDataGenerator.flow_from_directory() method to get images from two different directories (inputs1 and inputs2 respectively).

The model also is using two lambda layers to append the images procurred by the generators to a list for further inspection.

My question is how to train such a model. Here is some toy code:

I have a model which takes in a dataframe which looks like this

I am doing segmentation and my dataset is kinda small (1840 images) so I would like to use data-augmentation. I am using the generator provided in the keras documentation which yield a tuple with a batch of images and corresponding masks that got augmented the same way.

I'm changing my TensorFlow code from the old queue interface to the new Dataset API. In my old code I kept track of the epoch count by incrementing a tf.Variable every time a new input tensor is accessed and processed in the queue. I'd like to have this epoch count with the new Dataset API, but I'm having some trouble making it work.

Since I'm producing a variable amount of data items in the pre-processing stage, it is not a simple matter of incrementing a (Python) counter in the training loop - I need to compute the epoch count with respect to the input of the queues or Dataset.

I mimicked what I had before with the old queue system, and here is what I ended up with for the Dataset API (simplified example):

I try to augment my image data using the Keras ImageDataGenerator. My task is a regression task, where an input image results in another, transformed image. So far so good, works quite well.

Here I wanted to apply data augmentation by using the ImageDataGenerator. In order to transform both images the same way, I used the approach described in the Keras docs, where a transformation of an image with a corresponding mask is described. My case is a little bit different, as my images are already loaded and don't need to be fetched from a directory. This procedure was already described in another StackOverlow post.

To verify my implementation, I first used it without augmentation and using the ImageDataGenerator without any parameter specified. According to the class reference in the Keras docs, this should not alter the images. See this snippet:

How can I perform data augmentation when I use ROI-Pooling in a CNN network which I developed using MXnet ?

For example suppose I have a resnet50 architecture which uses a roi-pooling layer and I want to use random-crops data augmentation in the ImageRecord Iterator.

Is there an automatic way that data coordinates in the rois passed to the roi pooling layer, transform so as to be applied in images generated by the data-augmentation process of the ImageRecord Iterator ?