ML Bench

Machine Learning Benchmark Tool (ML Bench) (AI Benchmark Tool)

Supported models :

- MobileNet v1

- MobileNet v2

- Inception v3

- Resnet v2 50

- SSD Mobilenet v1 (Object Detection)

Supported runtime :

- Tensorflow Lite

- Tensorflow Mobile

- Android NN

- SNPE (for Qualcomm)

- PyTorth

- TVM

How to side load your model :

1. Prepare your model file (supported types are tflite, pytorch, tvm, dlc)

2. On your local machine, create [Model Name] directory

3. Copy your model file to the [Model Name] directory

4. Create a file called meta-data.json in the [Model Name] directory

example of meta-data.json :

{

"xres" : 299,

"yres" : 299,

"depth" : 3,

"input_type" : "float",

"output_type" : "float",

"input_name" : "input:0",

"output_name" : "InceptionV3/Predictions/Reshape_1:0",

"image_mean" : 0,

"image_std" : 0,

"label_offset": 1

}

* you can also set your own label, ground truth and image files

{

...

"labels": "my_labels.txt",

"ground_truth": "my_ground_truth.txt",

"image_path": "my/path/to/images",

...

}

5. push [Model Name] directory to your device using below command

adb push ./[Model Name] /sdcard/Download/MlBenchModels/