Release 4.3.1

• Fix a installation problem with PROJ due to OTB 9 packaging
• More tests for OTB

Release 4.3.0

• Add compute capability 8.0 and 9.0
• Remove OTB GUI dependencies
• Move OTB to version 9.0.0
• Move TensorFlow to version 2.14

Release 4.2.2

• Add a test to ensure that planetary computer works in the OTBTF docker image
• Move to OTB version d74ab47d4308591db4ed5a5ea3b820cef73a39fe since 8.2.1 looks not up-to-date

Release 4.2.1

• Fix a bug occurring when input images origins were not sharing the same pixel grid
• CI to push "latest-gpu" tag on dockerhub
• Move to OTB 8.2.1

Release 4.2.0

• Add new python modules: otbtf.layers (with new classes DilatedMask, ApplyMask, ScalarsTile, ArgMax, Max) and otbtf.ops (one_hot())
• Fix an error in the documentation
• Update the otbtf-keras tutorial
• Add decloud testing in CI
• Fix protobuf version in dockerfile (see tensorflow/tensorflow#61551)

Release 4.1.0

• Add no-data values support for inference in TensorflowModelServe application
• Fix CuDNN version detection in build-env-tf.sh
• Dockerfile args to build otbtf with TensorRT (experimental)

Release 4.0.0

Release 4.0.0 (5 april 2023)

• Big improvement of the documentation:
  • Re-structure the entire doc, remove deprecated stuff, etc.
  • Add a nice mkdocs template
  • Docs are now hosted at otbtf.readthedocs.io
  • Add a new section on the python API (end-to-end tutorial)
  • A lot of new sections: distributed training, etc...
• Refactoring all the python classes to enforce pep8
• Base images: ubuntu:22.04 (cpu), nvidia/cuda:12.0.1-cudnn8-devel-ubuntu22.04 (gpu)
• Tensorflow version: 2.12.0
• Latest OTB version (pre-9.0.0)
• Fixed Tensorflow error "Cannot register 2 metrics with the same name" + new test
• Faster CI build thanks to bazel remote cache
• /home/otbuser/.local/bin added to user path
• New tests to check the compatibility of numpy, tensorflow, gdal, rasterio, otbApplication, and otbtf modules together in the python environment

Release 3.4.0

• Up-to date OTB 8.1.1 version (5086d7601d80f2427f4d4d7f2398ec46e7efa300) with a lot of good bug fixes
• Fix the CI for docker images builds (docker/compose -> docker)

Release 3.3.0

Version 3.3.0 (27 jul 2022)

• Improves the dataset classes (DatasetFromPatchesImages, TFRecords) to use them easily in keras
• Add the ModelBase class, which eases considerably the creation of deep nets for Keras/TF/TensorflowModelServe
• Add an example explaining how to use python classes to build and train models with Keras, and use models in OTB.
• Document the python API (otbtf.dataset, otbtf.tfrecords, otbtf.ModelBase)
• Test the python API in the CI, using the (XS, labels) patches of the Amsterdam dataset from CRC book
• Upgrade OTB to version 8.0.1
• Upgrade GDAL to version 3.4.2

Release 3.2.1

Version 3.2.1 (1 jun 2022)

• Changing docker images naming convention (cpu/gpu-basic* --> cpu/gpu*, cpu/gpu* --> cpu/gpu-opt*) + only images without optimizations are pushed on dockerhub, image with optimization are pushed on gitlab.irstea.fr registry
• Enhance CI with automatic builds of docker images
• Fix a bug in dataset.py (output type was always Float32)
• Fix a bug in otbTensorflowSampler.hxx (see #81)
• Change the behavior of _read_extract_as_np_arr method: now returns 3D arrays even for singleband