Compilation stops

Question

Compilation stops

Opened this issue a year ago · 16 comments

I have tried your compile script for compiling my yolov5 ti lite onnx, but compilation stops at the starting it self

Answer 1 · 2023-09-22T04:21:22.000Z

I'd guess you are running out of memory. How much RAM is available on this system?

Otherwise, it's probably a bad input. Do my_model_data/last.onnx and calibration_images exist? Can you show what's in the calibration directory?

Also try running it under gdb: gdb --args python3.6 .... What does it tell you about the crash?

Answer 2 · 2023-09-22T04:28:04.000Z

Brother can please tell me the clear overview to compile my last.onnx file You are really awesome brother, I have followed your steps in finetuning yolo ti lite model, it's just awesome, please help me in compiling my model. I am having only ×64 processer,If i want to use ×86 build pc also no problem

…

On Fri, 22 Sep, 2023, 9:51 am Kaelin Laundry, ***@***.***> wrote: I'd guess you are running out of memory. How much RAM is available on this system? Otherwise, it's probably a bad input. Do my_model_data/last.onnx and calibration_images exist? Can you show what's in the calibration directory? Also try running it under gdb: gdb --args python3.6 .... What does it tell you about the crash? — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BA3A26E7ISYMIRD6BXTMA6DX3UG4ZANCNFSM6AAAAAA5BZIA34> . You are receiving this because you authored the thread.Message ID: ***@***.***>

Answer 3 · 2023-09-22T04:29:23.000Z

I'd guess you are running out of memory. How much RAM is available on this system?

Otherwise, it's probably a bad input. Do my_model_data/last.onnx and calibration_images exist? Can you show what's in the calibration directory?

Also try running it under gdb: gdb --args python3.6 .... What does it tell you about the crash?

Brother can I have your contact details like instagram?
Ram : 8gb
64 bit processor

Answer 4 · 2023-09-22T04:36:23.000Z

I'd guess you are running out of memory. How much RAM is available on this system?

Otherwise, it's probably a bad input. Do my_model_data/last.onnx and calibration_images exist? Can you show what's in the calibration directory?

Also try running it under gdb: gdb --args python3.6 .... What does it tell you about the crash?

There is no error messages like directory not found like that, only thing is that it starts and ends with 10secs

Answer 5 · 2023-09-22T04:53:01.000Z

I'd guess you are running out of memory. How much RAM is available on this system?

Otherwise, it's probably a bad input. Do my_model_data/last.onnx and calibration_images exist? Can you show what's in the calibration directory?

Also try running it under gdb: gdb --args python3.6 .... What does it tell you about the crash?

I had 20 images from the training set in calibration directory

Answer 6 · 2023-09-23T00:52:58.000Z

TI's official documentation here: https://software-dl.ti.com/codegen/docs/tvm/tvm_tidl_users_guide/compiling.html

It was not very good when I was working on this initially, but it may be better now.

What have you done already, and what files do you have? Show me terminal output that describes what you have done and what files are present. If you trained a model you should have last.onnx.

There is no error messages like directory not found like that, only thing is that it starts and ends with 10secs

Please show the output.

Answer 7 · 2023-09-23T04:37:15.000Z

TI's official documentation here: https://software-dl.ti.com/codegen/docs/tvm/tvm_tidl_users_guide/compiling.html

It was not very good when I was working on this initially, but it may be better now.

What have you done already, and what files do you have? Show me terminal output that describes what you have done and what files are present. If you trained a model you should have last.onnx.

There is no error messages like directory not found like that, only thing is that it starts and ends with 10secs

Please show the output.

Answer 8 · 2023-09-23T04:57:12.000Z

Mugutech62 commented a year ago

Answer 9 · 2023-09-23T20:43:10.000Z

In the gdb prompt that it gave you, type "run" and hit enter. Then type "bt" and hit enter. Post the output.

Answer 10 · 2023-09-23T20:57:27.000Z

In the gdb prompt that it gave you, type "run" and hit enter. Then type "bt" and hit enter. Post the output.

Whether I had done the setup correctly?

Answer 11 · 2023-09-24T03:49:47.000Z

In the gdb prompt that it gave you, type "run" and hit enter. Then type "bt" and hit enter. Post the output.

Ok brother let me try and tell you.
Bro i am having a trained yolov8 model is there any way to convert them TI compatible model, I mean whether can I convert into TI models

Answer 12 · 2023-09-24T11:07:50.000Z

In the gdb prompt that it gave you, type "run" and hit enter. Then type "bt" and hit enter. Post the output.

Answer 13 · 2023-09-24T20:27:53.000Z

I am not aware of support for YOLOv8.

Your GDB screenshot does not include the command you ran. It is likely the incorrect command, as gdb is complaining of unused arguments. Did you include --args after gdb?

Answer 14 · 2023-09-25T02:19:55.000Z

I am not aware of support for YOLOv8.

Your GDB screenshot does not include the command you ran. It is likely the incorrect command, as gdb is complaining of unused arguments. Did you include --args after gdb?

Brother can please some examples with your screenshots, kindly please🙏 or else type in notepad and share exact commands

Answer 15 · 2023-09-28T18:34:00.000Z

I am not aware of support for YOLOv8.

Your GDB screenshot does not include the command you ran. It is likely the incorrect command, as gdb is complaining of unused arguments. Did you include --args after gdb?

Bro kindly please share your compile.py for reference

Answer 16 · 2023-10-31T06:25:36.000Z

I got a new error check out pls