version: /ti-processor-sdk-rtos-j721e-evm-07_01_00_11/tidl_j7_01_03_00_11/
我原来训练好的模型输入是一个480行640列的张量，元素为取值0.0～1.0之间的float,
example:
shape(480.640)
[[0.98039216 0.98039216 0.98039216 ... 0.57254905 0.5803922 0.5882353 ]
[0.98039216 0.98039216 0.98039216 ... 0.5803922 0.5882353 0.59607846]
[0.98039216 0.98039216 0.98039216 ... 0.5882353 0.59607846 0.6117647 ]
...
[0.5529412 0.53333336 0.5254902 ... 0.25882354 0.25490198 0.2509804 ]
[0.5411765 0.5411765 0.53333336 ... 0.25490198 0.25882354 0.2627451 ]
[0.5529412 0.5686275 0.56078434 ... 0.2509804 0.2627451 0.27450982]]
请问我的indata文件.y应该是怎么样的编码格式？
config:
modelType = 2
numParamBits = 8
numFeatureBits = 8
inputNetFile = "/home/swl/ws/tidl/superpoint_v1.onnx"
outputNetFile = "/home/swl/ws/tidl/int8_model/tidl_net_superpoint_net1.1.bin"
outputParamsFile = "/home/swl/ws/tidl/int8_model/tidl_io_superpoint_net1.1_"
inWidth = 640
inHeight = 480
inNumChannels = 1
perfSimConfig = "/home/swl/ti-processor-sdk-rtos-j721e-evm-07_01_00_11/tidl_j7_01_03_00_11/ti_dl/test/testvecs/config/import/device_config.cfg"
debugTraceLevel = 3
writeTraceLevel = 0
inFileFormat = 1
inElementType = 0
rawDataInElementType = 0
postProcType = 0
numFrames = 1
netBinFile = "/home/swl/ws/tidl/int8_model/tidl_net_superpoint_net1.1.bin"
ioConfigFile = "/home/swl/ws/tidl/int8_model/tidl_io_superpoint_net1.1_1.bin"
inData = "/home/swl/ws/tidl/test_int8_in.y"
outData = "/home/swl/ws/tidl/int8_model/out/out.bin"
