TensorFlow Release 21.07

The NVIDIA container image of TensorFlow, release 21.07, is available on NGC.

Contents of the TensorFlow container

This container image includes the complete source of the NVIDIA version of TensorFlow in /opt/tensorflow. It is pre-built and installed as a system Python module.

To achieve optimum TensorFlow performance, for image based training, the container includes a sample script that demonstrates efficient training of convolutional neural networks (CNNs). The sample script may need to be modified to fit your application.

The container also includes the following:

Driver Requirements

Release 21.07 is based on NVIDIA CUDA 11.4.0, which requires NVIDIA Driver release 470 or later. However, if you are running on Data Center GPUs (formerly Tesla), for example, T4, you may use NVIDIA driver release 418.40 (or later R418), 440.33 (or later R440), 450.51 (or later R450), or 460.27 (or later R460). The CUDA driver's compatibility package only supports particular drivers. For a complete list of supported drivers, see the CUDA Application Compatibility topic. For more information, see CUDA Compatibility and Upgrades and NVIDIA CUDA and Drivers Support.

GPU Requirements

Release 21.07 supports CUDA compute capability 6.0 and higher. This corresponds to GPUs in the NVIDIA Pascal, Volta, Turing, and Ampere Architecture GPU families. Specifically, for a list of GPUs that this compute capability corresponds to, see CUDA GPUs. For additional support details, see Deep Learning Frameworks Support Matrix.

Key Features and Enhancements

This TensorFlow release includes the following key features and enhancements.
  • Increased GPU memory reservation to avoid OOM errors in some cases.
  • Integrated TRT 8 Support.
  • Improved NVTX markers to include XLA cluster names.
  • Fixed a deadlock in XLA by backporting upstream PR 50280 to TF1 and TF2.
  • Fixed an issue so that CUDNN now respects the TF32 disable switch.
  • TF2 implements support for embedding ops on GPU:
    • SparseFillEmptyRows[Grad]
    • fp16 embedding_lookup_sparse
    • fp16 SparseSegmentSumGrad
    • SparseSegmentSum/Mean
    • SparseSegmentSum/MeanGrad
    • hash value to string
  • TF2 - Use CUDA occupancy calculator to improve the performance of BiasAdd.
  • TensorFlow container images version 21.07 are based on Tensorflow 1.15.5 and 2.5.0

Tensor Core Examples

The tensor core examples provided in GitHub focus on achieving the best performance and convergence by using the latest deep learning example networks and model scripts for training.

Each example model trains with mixed precision Tensor Cores on Volta, therefore you can get results much faster than training without tensor cores. This model is tested against each NGC monthly container release to ensure consistent accuracy and performance over time. This container includes the following tensor core examples.

Known Issues

Note: If you encounter functional or performance issues when XLA is enabled, please refer to the XLA Best Practices document. It offers pointers on how to diagnose symptoms and possibly address them.
  • There is a known issue in TensorRT 8.0 regarding accuracy for a certain case of int8 inferencing on A40 and similar GPUs. The version of TF-TRT in TF2 21.07 includes a feature that works around this issue, but TF1 21.07 does not include that feature and may experience the accuracy drop for a small subset of model/data type/batch size combinations on A40. This will be fixed in the next version of TensorRT.
  • The TF1 21.07 container includes Django 3.2.2, which has a known vulnerability that was discovered late in our QA process. See CVE-2021-35042 for details. This will be fixed in the next release. TF2 21.07 is not vulnerable to this issue.
  • The 21.07 release includes libsystemd and libudev versions that have a known vulnerability that was discovered late in our QA process. See CVE-2021-33910 for details. This will be fixed in the next release.
  • A known regression can reduce the training performance of VGG-16 by up to 12% at certain batch sizes.
  • There is a known performance regression of up to 30% when training SSD models with fp32 data type on T4 GPUs. This will be addressed in a future release.
  • There is a known issue where attempting to convert some models using TF-TRT produces an error "Failed to import metagraph". This issue is still under investigation and will be resolved in a future release.