[D] State-of-the-Art for Self-Supervised (Pre-)Training of CNN architectures (e.g. ResNet)?

• solo-learn

  3 1,355 5.9 Python

  solo-learn: a library of self-supervised methods for visual representation learning powered by Pytorch Lightning

Hello, I lost a bit of touch to the current SOTA of self-supervised pretraining of CNNs, in particular ResNet. I found this repository https://github.com/vturrisi/solo-learn that has many methods implemented but I'm not really sure where to start. My goal is to pretrain a ResNet backbone on a decently large amount of image data that comes from a certain domain and after that fine-tune it for different downstream tasks (classification, segmentation, object detection) on a subset of the data I have labels for.

• mmselfsup

  5 3,084 5.3 Python

  OpenMMLab Self-Supervised Learning Toolbox and Benchmark

  

• InfluxDB

  www.influxdata.com featured

  Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.

  

Suggest a related project