Questions tagged [data-augmentation]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
0 votes
0 answers
13 views

NLP - F1 score for positive class drops to 0 after data augmentation

I'm working on a 3-class text classification problem where my initial class distribution looked like this: positive: 50% negative: 25% and neutral: 25% And training on a model on this slightly ...
0 votes
0 answers
8 views

What is the best way to acess the data in a pandas Window object?

I have a pandas DataFrame object, and I call its rolling method to get a Window object. Now I want to modify the data in each of the windows to get a new set of windows. Reading the source code, I ...
0 votes
0 answers
14 views

Is it correct to generate similar rows by reducing the time-frame of an instance?

I'm participating in a People Analytics project with a small historic dataset that includes event variables. The aim is to predict employee's attrition. I have variables like area, dept., company, etc....
  • 46
0 votes
1 answer
17 views

Model does not learn when using Keras 'flow_from_directory', but learns fine with 'image_dataset_from_directory'?

When classifying images with Keras, I am able to achieve a validation accuracy around 90-95%, however, I am trying to improve with the use of augmentation so have switched from ...
  • 1
0 votes
0 answers
7 views

Data augmentation useful for this or not?

I have a training set of medical breast x-ray images. Approximately half of them are flipped along the horizontal axis since some are from the right side and some are from the left side. See the ...
  • 1
0 votes
0 answers
19 views

Time Series data Augmentation in pytorch forecasting

I have count time series of demand data and some covariates like weather information every hour. I have used 168 hours (7 days) for encoder and 24 hours (next day) for decoder in DeepAR pytorch ...
1 vote
1 answer
98 views

CNN good results on train and test, bad results on real world data

I'm trying to build a neural network for an age detection task. Here some details : Dataset: I am using the "facial age" Kaggle dataset and the "UTKFace" dataset for a total of ...
0 votes
0 answers
11 views

Is possible to use GAN to generate images and masks for the semantic segmentation task?

I would like to know whether is possible to use Generative Adversarial Networks GAN to generate both images and masks for the semantic segmentation task.
  • 683
0 votes
0 answers
24 views

Data Augmentation avoid RAM problem

I have a dataset (imgs) which is a list of numpy.ndarray: images of dimension (200, 200, 3). When I try to augment it with imgaug library I always occur in RAM ...
0 votes
0 answers
73 views

Replace a lookup table with machine learning

I have a lookup table with 2 input columns and 2 output columns. I want to replace it with a value function such that with a given input pair, the function can give the output pair with minimal error. ...
  • 1
0 votes
0 answers
8 views

Data augmentation for region based time-series binary classification with contained feature values

I am working on time-series problem where I have feature values for various timesteps that are fed to a LSTM deep learning model. My features are all values within the range of [0, 1]. It is a binary ...
  • 73
1 vote
0 answers
17 views

What are some common data augmentation techniques used for code?

I understand that there are data augmentation techniques for natural language such as word/sentence shuffling, word replacement with synonyms and syntax-tree manipulation. However, I have hard time ...
  • 11
1 vote
1 answer
39 views

Certain Image Augmentation Prevent Unet Model from Learning

I am training a Unet model for cell image segmentation from microscopy images. In order to help the model generalize better to different microscopes, I attempted to apply brightness augmentation to ...
1 vote
1 answer
53 views

How to do Data Augmentation efficiently in Tensorflow 2?

First of all I'm asking that because of this tutorial. When I heard about Data Augmentation the definition I learned was something like: "It's a technique where we create more data to our dataset ...
0 votes
0 answers
18 views

Do we know why GAN-based data augmentation works?

Although I've seen many examples of GAN-generated synthetic data greatly improving the performance of models, I struggle to understand how this is possible. Say we are training a classifier $h$ to ...
0 votes
0 answers
19 views

Most efficient use of rare real input images - in training or validation set?

I want to do image segmentation with only very few realistic example images. Do I train on artificial data only and use the few real images as validation, thereby never directly learning from them at ...
  • 1
0 votes
0 answers
8 views

what libraries are available to generate augmented (synthetic) data at a vector level?

For instance, I have an embedding and I wish to generate a sampling of vectors similar to a sample embedding that can then be used for training a model.
  • 143
0 votes
0 answers
41 views

Data Augmentation Keras length of data

I'm confused when I add data augmentation should I get more data or the same data I tested my x_train length to confirm but I got the same length before augmentation and after augmentation is that ...
  • 1
1 vote
0 answers
74 views

Can data augmentation techniques be misleading?

In an attempt to handle imbalance in data, especially in the case of extremely imbalanced data, can the various data augmentation techniques create some bias?
  • 23
0 votes
0 answers
24 views

Is There Techniques for creating synthetic Data for Regression Problem i tried SMOTE and its variant but these are for classification problem

This is my data "Volume" is my Target variable and all other are Independent variables i just applied labelencoder on Area_categ , wind_direction_labelencod and on current _label_encode and ...
2 votes
1 answer
64 views

Data augmentation in images

Suppose there is a ML network that takes grayscale images as the input. The images that I have are RGB images. So, instead of converting these RGB images to grayscale, I treat each individual colour ...
1 vote
1 answer
25 views

Does synthetic data be over sampled as well?

I'm building a binary text classifier, the ratio between the positives and negatives is 1:100 (100 / 10000). By using back translation as an augmentation, I was able to get 400 more positives. Then I ...
0 votes
0 answers
23 views

Is there any papers on adaptive discriminator augmentation in 3D?

I am really impressed with the results of ADA in action. Currently I work with 2D data (normal png images) but I would like to train StyleGAN2 + ADA in 3D space. Is there any papers/implementations ...
0 votes
0 answers
37 views

How should I improve my CNN binary classification model from overfitting and underfitting [duplicate]

I am trying to do the cats & dogs classification problem, the problem is that my model is overfitting and I have tried all the techniques I know in order to solve but nothing is working such as ...
0 votes
0 answers
23 views

Baseline model and transfer learning

I've tried to find any guidance on using transfer learning when building baseline models for ML projects (CNN in my case) but found no clues on good practices in the matter. My logic says that no ...
1 vote
1 answer
173 views

Why we call Mix-up method is a data augmentation technique?

I am bit confused in the Mixup data augmentation technique, let me explain the problem briefly: What is Mixup For further detail you may refer to original paper . We double or quadruple the data ...
  • 138
0 votes
0 answers
27 views

Augmentation for sound recognition of dog barks for CNNs

I am training CNNs to recognize dog barking, and for this I would like to augment the data sets I have (~30'000 10s clips with either barks, or no-barks in them). The straight forward idea was to mix ...
1 vote
1 answer
113 views

ValueError: Error when checking input: expected time_distributed_6_input to have 5 dimensions, but got array with shape (32, 224, 224, 3) [closed]

I am trying to apply data augmentation to avoid overffiting in my CNN-LSTM image classification model. My training data has the shape: (1882, 1, 224, 224, 3) My ...
  • 111
1 vote
1 answer
174 views

how does zoom out works in data augmentation?

how does zoom out works in data augmentation? I'm reading a doc on data augmentation in Keras ,and it says that randomZoom(0.2) zooms in and out by a factor in rage ...
0 votes
2 answers
219 views

Removing outliers from a multi-dimensional dataset & Data augmentation

Removing the outliers of a single-dimensional data can be easily done by removing the points that are outside of the IQR range. But how should the process of detecting and removing outliers be done if ...
1 vote
0 answers
35 views

Methods to combine datasets from different time periods

Consider a multivariate time series forecasting task where I have two datasets A and B. A goes from 1960 to 2020 and B goes from 2010 to 2020. There is a feature f ...
  • 155
0 votes
2 answers
174 views

How much data augmentation is required on an imbalanced dataset?

Imagine I have a dataset with positive and negative sentences, and I need to train a transformer (Like BERT) to do the binary classification. The problem is that there are 100 negative sentences and ...
0 votes
1 answer
39 views

Batch normalization for multiple datasets?

I am working on a task of generating synthetic data to help the training of my model. This means that the training is performed on synthetic + real data, and tested on real data. I was told that batch ...
  • 111
1 vote
0 answers
41 views

Specify torchvision transforms depending on the properties of an image and a mask

I have a dataset 1000 of images and corresponding segmentation masks from dermatologists. The images come in different sizes (as low as 400x600 and as large as 4Kx4K). 95% of image pixels are not ...
0 votes
0 answers
25 views

How can I get dark, median-brightness and bright distribution histograms from an image?

I'm trying to replicate an augmentation technique used in this paper. This is how they explain the procedure: [...] the augmentation technique was used to adjust the histogram because the intensity ...
0 votes
1 answer
185 views

What is the scope of Keras' ImageDataGenerator.flow_from_dataframe seed parameter?

I've been working on a U-Net model using training images stored on my local drive. To load these I have been using Keras' ImageDataGenerator.flow_from_dataframe ...
  • 111
1 vote
0 answers
16 views

Working on an image classification project (microscopic images) , have some doubts [closed]

Currently, I am working on an image classification project. The data set contains very high resolution images taken via an electron microscope. Hence, I have few and limited instances. I have done EDA ...
  • 11
4 votes
1 answer
2k views

Why does adding data augmentation decrease training accuracy a tiny bit?

Before data augmentation, my model clearly overfits and hits a 100% training accuracy and a 52% validation accuracy. When only adding data augmentation with Keras, as a regularization technique, it ...
  • 43
0 votes
1 answer
227 views

Does Mixup requires two loss functions?

I created a neural network with multi-label classification using MSE. Now, I would like to use Mixup. Do I need two loss functions (for each target one) or is the result the same if I just combine the ...
1 vote
1 answer
329 views

How to mantain the nested structure of a tf.dataset after applying map?

I'm creating a tf.dataset object containing 2 images as inputs and a mask as target. All of them are 3D in grayscale. After applying a custom map, the shape of the object changes from ...
0 votes
1 answer
274 views

How to implement random cropping during training?

I'm developing a U-net like model which segments the damaged tissue of the brain between two time-points in Multiple Sclerosis patients. The model is given the baseline and follow-up images as x and ...
1 vote
1 answer
1k views

Rescale parameter in data augmentation

I'm a little bit lost about the rescale parameter in the ImageDatagenerator function. I know that the rescale argument by itself does not augment my data and that ...
  • 113
0 votes
1 answer
1k views

How many images are generated when ImageDataGenerator is used, and when data augmentation is included as a part of the model?

Is there any way to know the number of images generated by the ImageDataGenerator class and loading data using flow_from_directory method? I searched everywhere for the same but couldn't find anything ...
3 votes
0 answers
231 views

Is it possible to increase the number of images of one class using data augmentation, which is not applied on the other class, in the same dataset?

I have 2 classes for my image classification problem, say class A and class B, and I am using tensorflow and keras for the same. One of them have around 5K images while the other have just around 2K ...
2 votes
2 answers
512 views

What is the use of using width/height shift in data augmentation?

I'm not sure to understand the use of augmentation data using width shift and height shift. Say I have limited image data, and I want to create new data using Keras' ImageDataGenerator. To classify ...
0 votes
0 answers
24 views

Should I resize my object localization dataset in order to train a fully convolutional network?

I want to create a fully-convolutional neural net that trains on wider face datasets in order to draw bounding box around faces. The dataset is highly diverse in the image sizes. So my question is, ...
1 vote
1 answer
61 views

Is it better to augment both training and validation sets or just training set?

Is it better to augment data both training and validation sets or just the training set in order to achieve the best accuracy possible on a convolutional neural network? why?
  • 461
0 votes
1 answer
25 views

Textbook definition of Data augmentation

I'm trying to find a textbook definition of Data augmentation. We all know what it is, but I'm having a hard time finding a reliable source which I can cite for my paper: The Wikipedia article on ...
  • 123
2 votes
1 answer
1k views

First perform data augmentation or normalization?

Should I first perform data augmentation or normalization in deep learning? I am mainly interested in 2D and 3D input data. In tutorials that I have seen so far the data augmentation always comes ...
  • 141
1 vote
2 answers
354 views

Why should I use data augmentation as Keras layer

I've seen in several Tensorflow/Keras tutorials that data augmentation functions are added as keras layers. When I converted my Keras Python model (for production purpose) to TensorflowJS I faced the ...
  • 123