search

ARM-software / armnn

Arm NN ML Software. The code here is a read-only mirror of https://review.mlplatform.org/admin/repos/ml/armnn
https://developer.arm.com/products/processors/machine-learning/arm-nn
MIT License
1.15k stars 307 forks source link

ARMNN v21.05: Checking intermediate output at various hidden layer #628

Closed supratimc239 closed 2 years ago

supratimc239 commented 2 years ago

Hi,

While executing my model, I am seeing some differences between prediction from Python and armnn. In order to debug this, I was planning to compare intermediate ouput at various hidden layers (I mean after various ops).

Could you please tell me how I can dump intermediate output at various ops level (for example my model consists of multiple conv2D and I want to dump output for each of the conv2D module)?

Kind regards Supratim

morgolock commented 2 years ago

HI @supratimc239

You can dump intermediate layers with the -p option root@acl_hikey_1:~/tmp/pabtel01/armnn# LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH ./ExecuteNetwork -f tflite-binary -m mobilenet_v2_1.0_224_quant.tflite -i input -o output -c CpuAcc -p

Hope this helps

supratimc239 commented 2 years ago

Hi @morgolock ,

I upgraded my ACL and ARM-NN to v22.02, but the ARM-NN build is failing. Could you please have a look at my latest comment in https://github.com/ARM-software/armnn/issues/607 and see why the ARM-NN build is failing.

Once I can execute my model on v22.02 will start checking your instruction on dumping intermediate layers.

Thanks

MikeJKelly commented 2 years ago

Hi I think this issue has been resolved and the linked ticket #607 has been closed so I'm going to close it now, if you need any more help with this issue then please open a new issue, thanks!

Best regards, Mike