1 # TensorFlow Lite operators that the Arm NN SDK supports
3 This reference guide provides a list of TensorFlow Lite operators the Arm NN SDK currently supports.
7 The Arm NN SDK TensorFlow Lite parser currently supports the following operators:
11 * AVERAGE_POOL_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE
15 * CONCATENATION, Supported Fused Activation: RELU , RELU6 , TANH, NONE
17 * CONV_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE
19 * DEPTHWISE_CONV_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE
25 * FULLY_CONNECTED, Supported Fused Activation: RELU , RELU6 , TANH, NONE
31 * MAX_POOL_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE
55 * RESIZE_NEAREST_NEIGHBOR
83 * TFLite_Detection_PostProcess
87 Arm tested these operators with the following TensorFlow Lite neural network:
89 * [Quantized MobileNet](http://download.tensorflow.org/models/mobilenet_v1_2018_02_22/mobilenet_v1_1.0_224_quant.tgz)
91 * [Quantized SSD MobileNet](http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_quantized_300x300_coco14_sync_2018_07_18.tar.gz)
93 * DeepSpeech v1 converted from [TensorFlow model](https://github.com/mozilla/DeepSpeech/releases/tag/v0.4.1)
97 * [DeepLab v3+](https://www.tensorflow.org/lite/models/segmentation/overview)
101 * RDN converted from [TensorFlow model](https://github.com/hengchuan/RDN-TensorFlow)
103 * Quantized RDN (CpuRef)
105 * [Quantized Inception v3](http://download.tensorflow.org/models/tflite_11_05_08/inception_v3_quant.tgz)
107 * [Quantized Inception v4](http://download.tensorflow.org/models/inception_v4_299_quant_20181026.tgz) (CpuRef)
109 * Quantized ResNet v2 50 (CpuRef)
111 * Quantized Yolo v3 (CpuRef)
113 More machine learning operators will be supported in future releases.