NVIDIA Optimized Frameworks
NVIDIA Optimized Frameworks (Latest Release) Download PDF

TensorFlow Release 19.11

The NVIDIA container image of TensorFlow, release 19.11, is available on NGC.

Contents of the TensorFlow container

This container image contains the complete source of the version of NVIDIA 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 19.11 is based on NVIDIA CUDA 10.2.89, which requires NVIDIA Driver release 440.30. However, if you are running on Tesla (for example, T4 or any other Tesla board), you may use NVIDIA driver release 396, 384.111+, 410 or 418.xx. 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.

GPU Requirements

Release 19.11 supports CUDA compute capability 6.0 and higher. This corresponds to GPUs in the Pascal, Volta, and Turing 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.

Announcements

  • We will stop support for Python 2.7 in a future TensorFlow container release.
  • Deep learning framework containers 19.11 and later include experimental support for Singularity v3.0.

Accelerating Inference In TensorFlow With TensorRT (TF-TRT)

For step-by-step instructions on how to use TF-TRT, see Accelerating Inference In TensorFlow With TensorRT User Guide.

Key Features And Enhancements
  • New converters were added. Refer to the Supported Operators section in the Accelerating Inference In TensorFlow With TensorRT User Guide for the list of new converters.

  • TensorFlow 2.0:
    • A new API is introduced for TF-TRT in TensorFlow 2.0. This new API can only be used in TensorFlow 2.0. Refer to the User Guide for more information about the new API.

    • Introduced a new API method (converter.build()) for optimizing TensorRT engines during graph optimization. Previously, the optimization during preprocessing (before deployment) was possible by using is_dynamic_op=False.

    • converter.convert() no longer returns a tf.function. Now, the function must be accessed from the saved model. This encapsulates the function in the converter for better safety.

    • The converter.calibrate() method has been removed. To trigger calibration, a calibration_input_fn should be provided to converter.convert().

Deprecated Features
  • The old API of TF-TRT is deprecated. It still works in TensorFlow 1.14 and 1.15, however, it is removed in TensorFlow 2.0. The old API is a Python function named create_inference_graph which is now replaced by the Python class TrtGraphConverterin TensorFlow 1.x and TrtGraphConverterV2 in TensorFlow 2.0 with a number of methods. Refer to TF-TRT User Guide for more information about the API and how to use it.

Known Issues
  • We have observed a regression in the performance of certain TF-TRT benchmarks in TensorFlow 1.15 including image classification models with precision INT8. We are still investigating this. Since 19.11 comes with a new version of TensorFlow (1.15), which includes a lot of changes in the TensorFlow backend, it’s very possible that the regression is caused by a change in the TensorFlow backend.

  • CUDA 10.2 and NCCL 2.5.x libraries require slightly more device memory than previous releases. As a result, some models that ran previously may exhaust device memory.

  • TensorRT INT8 calibration algorithm (see the TF-TRT User Guide for more information about how to use INT8) is very slow for certain models such as NASNet and Inception. We are working on optimizing the calibration algorithm in TensorRT.

  • The accuracy of Faster RCNN with the backbone ResNet-50 using TensorRT6.0 INT8 calibration is lower than expected. We are investigating the issue.

  • The following sentence that appears in the log of TensorRT 6.0 can be safely ignored. This will be removed in the future releases of TensorRT.
    Copy
    Copied!
                

    Calling isShapeTensor before the entire network is constructed may result in an inaccurate result.

  • The following warning is issued when the method build() from the API is not called. This warning can be ignored.
    Copy
    Copied!
                

    OP_REQUIRES failed at trt_engine_resource_ops.cc:183 : Not found: Container TF-TRT does not exist. (Could not find resource: TF-TRT/TRTEngineOp_...

  • The following warning is issued because internally TensorFlow calls the TensorRT optimizer for certain objects unnecessarily. This warning can be ignored.
    Copy
    Copied!
                

    OP_REQUIRES failed at trt_engine_resource_ops.cc:183 : Not found: Container TF-TRT does not exist. (Could not find resource: TF-TRT/TRTEngineOp_...

  • We have seen failures when using INT8 calibration (post-training) within the same process that does FP32/FP16 conversion. We recommend to use separate processes for different precisions until this issue gets resolved.

  • We have seen failures when calling the TensorRT optimizer on models that are already optimized by TensorRT. This issue will be fixed in a future release.

  • In case you import nets from models/slim, you might see the following error:
    Copy
    Copied!
                

    AttributeError: module 'tensorflow_core.contrib' has no attribute 'tensorrt'

    Changing the order of imports can fix the issue. Therefore, import TensorRT before importing nets as follows:

    Copy
    Copied!
                

    import tensorflow.contrib.tensorrt as trt import nets.nets_factory

Automatic Mixed Precision (AMP)

Automatic mixed precision converts certain float32 operations to operate in float16 which can run much faster on Tensor Cores. Automatic mixed precision is built on two components:

  • a loss scaling optimizer
  • graph rewriter

For models already using an optimizer from tf.train or tf.keras.optimizers for both compute_gradients() and apply_gradients() operations (for example, by calling optimizer.minimize() or model.fit(), automatic mixed precision can be enabled by wrapping the optimizer with tf.train.experimental.enable_mixed_precision_graph_rewrite().

For more information on this function, see the TensorFlow documentation here. For backward compatibility with previous container releases, AMP can also be enabled for tf.train optimizers by defining the following environment variable:

Copy
Copied!
            

export TF_ENABLE_AUTO_MIXED_PRECISION=1


For more information about how to access and enable Automatic mixed precision for TensorFlow, see Automatic Mixed Precision Training In TensorFlow from the TensorFlow User Guide, along with Training With Mixed Precision.

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

  • There are known issues in the 19.11 release for NCF inference with XLA and VGG16 training without XLA; these benchmarks have performance that is lower than expected.

  • There are known issues regarding TF-TRT INT8 accuracy issues. See the Accelerating Inference In TensorFlow With TensorRT (TF-TRT) section above for more information.

  • For BERT Large training with the 19.08 release on Tesla V100 boards with 16 GB memory, performance with batch size 3 per GPU is lower than expected; batch size 2 per GPU may be a better choice for this model on these GPUs with the 19.08 release. 32 GB GPUs are not affected.

  • TensorBoard has a bug in its IPv6 support which can result in the following error: Tensorboard could not bind to unsupported address family ::. To workaround this error, pass the --host <IP>flag when starting TensorBoard.

  • Automatic Mixed Precision (AMP) does not support the Keras LearningRateSchedulerin the 19.08 release. A fix will be included in a future release.

  • A known issue in TensorFlow results in the error Cannot take the length of Shape with unknown rank when training variable sized images with the Keras model.fit API. Details are provided here and a fix will be available in a future release.

  • Support for CUDNN float32 Tensor Op Math mode first introduced in the 18.09 release is now deprecated in favor of Automatic Mixed Precision. It is scheduled to be removed after the 19.11 release.

  • There is a known issue when your NVIDIA driver release is older than 418.xx in the 19.10 release, the Nsight Systems profiling tool (for example, the nsys) might cause CUDA runtime API error. A fix will be included in a future release.

© Copyright 2024, NVIDIA. Last updated on Dec 2, 2024.