.npy files under this folder are downloaded from model of SSD MobileNet V1 generated by following commands, which is licensed under the Apache 2.0.
How to Generate SSD MobileNet V1 TFLite Model
Download and extract the origin SSD MobileNet V1 model, use following converter command to remove the preprocess and postprocess graphs by using tensorflow's optimize_for_inference tool:
python3 -m tensorflow.python.tools.optimize_for_inference \
--input=./frozen_inference_graph.pb \
--output=./frozen_inference_graph_stripped.pb --frozen_graph=True \
--input_names=Preprocessor/sub \
--output_names='concat,concat_1' \
--alsologtostderr
Use 'TensorFlow Lite converter' tool to convert frozen graph to tflite model:
tflite_convert \
--graph_def_file=./frozen_inference_graph_stripped.pb \
--output_file=./ssd_mobilenet_v1_coco.tflite \
--input_format=TENSORFLOW_GRAPHDEF --output_format=TFLITE \
--input_shapes=1,300,300,3 --input_arrays=Preprocessor/sub \
--enable_v1_converter \
--output_arrays='concat,concat_1' \
--inference_type=FLOAT --logtostderr