TensorFlow 2.x [Deprecated]

This section outlines all computer-vision training and finetuning pipelines that are implemented with TensorFlow 2.x. As of TAO 6.25.09, support for these models have been deprecated and will be removed in a future release. FTMS API’s do not support these models as of TAO 6.25.09. For support with TAO APIs, please fall back to TAO 6.0.0 or by building the Docker container from source.

The source code for these networks are hosted on GitHub.

• Image Classification (TF2)
  • Preparing the Input Data Structure
  • Creating an Experiment Spec File - Specification File for Classification
  • Training the model
  • Evaluating the Model
  • Running Inference on a Model
  • Pruning the Model
  • Re-training the Pruned Model
  • Exporting the model
  • TensorRT Engine Generation, Validation, and int8 Calibration
  • Deploying to DeepStream
• EfficientDet (TF2)
  • Data Input for EfficientDet
  • Pre-processing the Dataset
  • Creating a Configuration File
  • Training the Model
  • Evaluating the Model
  • Running Inference with an EfficientDet Model
  • Pruning the Model
  • Re-training the Pruned Model
  • Exporting the Model
  • TensorRT Engine Generation, Validation, and int8 Calibration
  • Deploying to DeepStream