This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA2: can not parse the inference result file(stats_tool_out.bin)

Part Number: TDA2

Hi everybody 

I suffered from a problem when we try to deploy tensorflow pre-trained model on td2x.

Every step is fine(as described below). But, we can not parse the inference result file(stats_tool_out.bin)

What is the format of 1-D vector as described in http://software-dl.ti.com/processor-sdk-linux/esd/docs/latest/linux/Foundational_Components_TIDL.html

How can we parse it?

Does stats_tool_out.bin depend on training framework(e.g. caffe or tensorflow)?

Can anyone help us?

Steps are as follows:

We have transformed the pre-trained model(MobileNet_v1_1.0_224) in https://github.com/tensorflow/models/tree/master/research/slim through the command below

python "tensorflow\python\tools\optimize_for_inference.py" -- input=mobilenet_v1_1.0_224.pb --output=mobilenet_v1_1.0_224_final.pb -- input_names=input --output_names="softmax/Softmax" 

Then, we use tidl_model_import.out.exe and  mobilenet_v1_1.0_224_final.pb to produce two bin files.

Finally, we fed a image and the two bin files to get prediction file (stats_tool_out.bin)