English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning applications are drastically progressing in seismic processingand interpretation tasks. However, the majority of approaches subsample datavolumes and restrict model sizes to minimise computational requirements.Subsampling the data risks losing vital spatio-temporal information which couldaid training whilst restricting model sizes can impact model performance, or insome extreme cases, renders more complicated tasks such as segmentationimpossible. This paper illustrates how to tackle the two main issues oftraining of large neural networks: memory limitations and impracticably largetraining times. Typically, training data is preloaded into memory prior totraining, a particular challenge for seismic applications where data istypically four times larger than that used for standard image processing tasks(float32 vs. uint8). Using a microseismic use case, we illustrate how over750GB of data can be used to train a model by using a data generator approachwhich only stores in memory the data required for that training batch.Furthermore, efficient training over large models is illustrated through thetraining of a 7-layer UNet with input data dimensions of 4096X4096. Through abatch-splitting distributed training approach, training times are reduced by afactor of four. The combination of data generators and distributed trainingremoves any necessity of data 1 subsampling or restriction of neural networksizes, offering the opportunity of utilisation of larger networks,higher-resolution input data or moving from 2D to 3D problem spaces.

An introduction to distributed training of deep neural networks for segmentation tasks with large seismic datasets