PROCESSOR-SDK-AM69A: Issue when switching to using the tidlcpp library

Thai Hung Le

Part Number: PROCESSOR-SDK-AM69A
Other Parts Discussed in Thread: AM69A

Tool/software:

Hi Chris,

I am posting here on behalf of customer. Please help to taka look and comment. Thank you very much

Thanks for your previous guidance. I’ve verified that there’s no issue with the name_overlays DTBO file—I was able to successfully run inference of the ResNet-101 model on the device exactly as you described.
However, when I switch to using the tidlcpp library (TIDL C++ API) to load the model and perform inference, I encounter the following error:

root@am69-sk:/opt/edgeai-tidl-tools/bin/Release# ./tidlrt_resnet -f /opt/arrow/out -i face_aligned_1.jpg
Model Files names : /opt/arrow/out/tidl_net.bin,/opt/arrow/out/tidl_io_buff1.bin
APP: Init ... !!!
18753.543247 s: MEM: Init ... !!!
18753.543309 s: MEM: Initialized DMA HEAP (fd=5) !!!
18753.543457 s: MEM: Init ... Done !!!
18753.543479 s: IPC: Init ... !!!
18753.574032 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
18753.581335 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
18753.581420 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
18753.581432 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
18753.581450 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
18753.581986 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
18753.582115 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
18753.582219 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
18753.582339 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
18753.582351 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
18753.582364 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO
18753.749631 s: VX_ZONE_ERROR: [ownContextSendCmd:912] Command ack message returned failure cmd_status: -1
18753.749666 s: VX_ZONE_ERROR: [ownNodeKernelInit:604] Target kernel, TIVX_CMD_NODE_CREATE failed for node TIDLNode
18753.749673 s: VX_ZONE_ERROR: [ownNodeKernelInit:605] Please be sure the target callbacks have been registered for this core
18753.749684 s: VX_ZONE_ERROR: [ownNodeKernelInit:606] If the target callbacks have been registered, please ensure no errors are occurring within the create callback of this kernel
18753.749696 s: VX_ZONE_ERROR: [ownGraphNodeKernelInit:690] kernel init for node 0, kernel com.ti.tidl:1:1 ... failed !!!
18753.749742 s: VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Node kernel init failed
18753.749753 s: VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Graph verify failed
TIDL_RT_OVX: ERROR: Verifying TIDL graph ... Failed !!!
TIDL_RT_OVX: ERROR: Verify OpenVX graph failed
invoked
18753.810931 s: VX_ZONE_ERROR: [ownContextSendCmd:912] Command ack message returned failure cmd_status: -1
18753.810973 s: VX_ZONE_ERROR: [ownNodeKernelInit:604] Target kernel, TIVX_CMD_NODE_CREATE failed for node TIDLNode
18753.810985 s: VX_ZONE_ERROR: [ownNodeKernelInit:605] Please be sure the target callbacks have been registered for this core
18753.811000 s: VX_ZONE_ERROR: [ownNodeKernelInit:606] If the target callbacks have been registered, please ensure no errors are occurring within the create callback of this kernel
18753.811014 s: VX_ZONE_ERROR: [ownGraphNodeKernelInit:690] kernel init for node 0, kernel com.ti.tidl:1:1 ... failed !!!
18753.811032 s: VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Node kernel init failed
18753.811042 s: VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Graph verify failed
18753.811111 s: VX_ZONE_ERROR: [ownGraphScheduleGraphWrapper:944] graph is not in a state required to be scheduled
18753.811122 s: VX_ZONE_ERROR: [vxProcessGraph:868] schedule graph failed
18753.811133 s: VX_ZONE_ERROR: [vxProcessGraph:873] wait graph failed
ERROR: Running TIDL graph ... Failed !!!
average time: 49.384 ms
Feature vector statistics:
Dimension: 512
Min value: 0
Max value: 0
Mean value: 0
Feature vector saved to face_aligned_1.jpg.features.txt
APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
18753.828821 s: IPC: Deinit ... !!!
18753.829796 s: IPC: DeInit ... Done !!!
18753.829828 s: MEM: Deinit ... !!!
18753.829842 s: DDR_SHARED_MEM: Alloc's: 8 alloc's of 135423796 bytes
18753.829852 s: DDR_SHARED_MEM: Free's : 8 free's of 135423796 bytes
18753.829861 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
18753.829876 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!

Could you please share a minimal example or detailed steps showing how to:

Load the converted .onnx (or .tflite) model using tidlcpp
Initialize the TIDL context and allocate input/output tensors
Run inference and read back the output tensor data

Any sample code or pointers on integrating tidlcpp into my application would be hugely helpful.

Best regards,

Hung

4 months ago

0 Chris Tsongas 4 months ago

TI__Genius 14960 points

Hi Hung,

I am not sure exactly what tidlcpp is and why it is being used. If you want to minimally test a TIDL application, use TI_DEVICE_armv8_test_dl_algo_host_rt.out in /opt/tidl_test on the device. This will remove all variables.

I cannot just give a configuration for an unknown model that will work, but here is an example inference configuration.

inFileFormat = 0
debugTraceLevel = 0
writeTraceLevel = 0
netBinFile = "/opt/test_model/tidl_net.bin"
ioConfigFile = "/opt/test_model/tidl_io_buff1.bin"
outData = "/opt/test_model/test_model_out.bin"
inData = "/opt/test_model/image.bmp"
numFrames = 1

Assuming the above file is called infer_evm.txt in the /opt/tidl_test directory:

root@am69-sk:/opt/tidl_test# ./TI_DEVICE_armv8_test_dl_algo_host_rt.out s:infer_evm.txt

Processing config file #0 : infer_evm.txt
APP: Init ... !!!
525686.721685 s: MEM: Init ... !!!
525686.721735 s: MEM: Initialized DMA HEAP (fd=5) !!!
525686.721868 s: MEM: Init ... Done !!!
525686.721886 s: IPC: Init ... !!!
525686.757416 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
525686.765319 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
525686.765423 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
525686.765437 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
525686.765445 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
525686.766365 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
525686.766511 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
525686.766600 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
525686.766686 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
525686.766695 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
525686.766706 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO

# NETWORK_INIT_TIME = 62.54 (in ms, c7x @1GHz)
----------------------- TIDL Process with TARGET DATA FLOW ------------------------

# NETWORK_EXECUTION_TIME = 4.13 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) = 0.00, 0.00, 0.00 (in Mega Bytes/frame) ... .... .....APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
525686.837667 s: IPC: Deinit ... !!!
525686.838680 s: IPC: DeInit ... Done !!!
525686.838717 s: MEM: Deinit ... !!!
525686.838731 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 4592504 bytes
525686.838742 s: DDR_SHARED_MEM: Free's : 7 free's of 4592504 bytes
525686.838752 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
525686.838765 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!

Regards,

Chris

0 An Dao 4 months ago in reply to Chris Tsongas

Prodigy 170 points

Previously, I successfully tested the TI_DEVICE_armv8_test_dl_algo_host_rt target. Now I need to run the model using the TIDL Runtime from source code. I’m taking the following steps:

Cloned the repository:

git clone github.com/.../edgeai-tidl-tools.git

Checked out branch 10_01_00_02.
Added my code into /opt/edgeai-tidl-tools/examples/tidlrt_cpp to build an application that runs the CMC ResNet-101 model.

However, when I rebuild and execute the example, I encounter the error shown in the attached log. Could you please review the attached source files for any issues, rebuild the example, and verify that it runs correctly?

Thank you in advance for your assistance.

Fullscreen log_eror (1).txt Download

root@am69-sk:/opt/edgeai-tidl-tools/bin/Release# ./tidlrt_resnet -f /opt/arrow/out -i face_aligned_1.jpg
Model Files names : /opt/arrow/out/tidl_net.bin,/opt/arrow/out/tidl_io_buff1.bin
APP: Init ... !!!
 18753.543247 s: MEM: Init ... !!!
 18753.543309 s: MEM: Initialized DMA HEAP (fd=5) !!!
 18753.543457 s: MEM: Init ... Done !!!
 18753.543479 s: IPC: Init ... !!!
 18753.574032 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
 18753.581335 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
 18753.581420 s:  VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
 18753.581432 s:  VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
 18753.581450 s:  VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
 18753.581986 s:  VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
 18753.582115 s:  VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
 18753.582219 s:  VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
 18753.582339 s:  VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
 18753.582351 s:  VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
 18753.582364 s:  VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO
 18753.749631 s:  VX_ZONE_ERROR: [ownContextSendCmd:912] Command ack message returned failure cmd_status: -1
 18753.749666 s:  VX_ZONE_ERROR: [ownNodeKernelInit:604] Target kernel, TIVX_CMD_NODE_CREATE failed for node TIDLNode
 18753.749673 s:  VX_ZONE_ERROR: [ownNodeKernelInit:605] Please be sure the target callbacks have been registered for this core
 18753.749684 s:  VX_ZONE_ERROR: [ownNodeKernelInit:606] If the target callbacks have been registered, please ensure no errors are occurring within the create callback of this kernel
 18753.749696 s:  VX_ZONE_ERROR: [ownGraphNodeKernelInit:690] kernel init for node 0, kernel com.ti.tidl:1:1 ... failed !!!
 18753.749742 s:  VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Node kernel init failed
 18753.749753 s:  VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Graph verify failed
TIDL_RT_OVX: ERROR: Verifying TIDL graph ... Failed !!!
TIDL_RT_OVX: ERROR: Verify OpenVX graph failed
invoked
 18753.810931 s:  VX_ZONE_ERROR: [ownContextSendCmd:912] Command ack message returned failure cmd_status: -1
 18753.810973 s:  VX_ZONE_ERROR: [ownNodeKernelInit:604] Target kernel, TIVX_CMD_NODE_CREATE failed for node TIDLNode
 18753.810985 s:  VX_ZONE_ERROR: [ownNodeKernelInit:605] Please be sure the target callbacks have been registered for this core
 18753.811000 s:  VX_ZONE_ERROR: [ownNodeKernelInit:606] If the target callbacks have been registered, please ensure no errors are occurring within the create callback of this kernel
 18753.811014 s:  VX_ZONE_ERROR: [ownGraphNodeKernelInit:690] kernel init for node 0, kernel com.ti.tidl:1:1 ... failed !!!
 18753.811032 s:  VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Node kernel init failed
 18753.811042 s:  VX_ZONE_ERROR: [ TIDL subgraph 1332 ] Graph verify failed
 18753.811111 s:  VX_ZONE_ERROR: [ownGraphScheduleGraphWrapper:944] graph is not in a state required to be scheduled
 18753.811122 s:  VX_ZONE_ERROR: [vxProcessGraph:868] schedule graph failed
 18753.811133 s:  VX_ZONE_ERROR: [vxProcessGraph:873] wait graph failed
ERROR: Running TIDL graph ... Failed !!!
average time: 49.384 ms
Feature vector statistics:
  Dimension: 512
  Min value: 0
  Max value: 0
  Mean value: 0
Feature vector saved to face_aligned_1.jpg.features.txt
APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
 18753.828821 s: IPC: Deinit ... !!!
 18753.829796 s: IPC: DeInit ... Done !!!
 18753.829828 s: MEM: Deinit ... !!!
 18753.829842 s: DDR_SHARED_MEM: Alloc's: 8 alloc's of 135423796 bytes
 18753.829852 s: DDR_SHARED_MEM: Free's : 8 free's  of 135423796 bytes
 18753.829861 s: DDR_SHARED_MEM: Open's : 0 allocs  of 0 bytes
 18753.829876 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!

4747.resnet.zip

0 Chris Tsongas 4 months ago in reply to An Dao

TI__Genius 14960 points

Hi Hung,

I rebuilt your application, and I replicated the same error. Which implies the problem is in the cpp application. This is probably a version/library/env problem. Can you please provide the env you are running? Also, did you build the tidlrt_resnet on the PC or EVM?

With the standard utility and an infer_evm.txt of:

inFileFormat = 0
debugTraceLevel = 0
writeTraceLevel = 0
netBinFile = "/opt/arrow/out/tidl_net.bin"
ioConfigFile = "/opt/arrow/out/tidl_io_buff1.bin"
outData = "/opt/arrow/out/test_model_out.bin"
inData = "/opt/test_model/jet.bmp"
numFrames = 1

./TI_DEVICE_armv8_test_dl_algo_host_rt.out s:infer_evm.txt

Processing config file #0 : infer_evm.txt
APP: Init ... !!!
795260.849938 s: MEM: Init ... !!!
795260.849988 s: MEM: Initialized DMA HEAP (fd=5) !!!
795260.850126 s: MEM: Init ... Done !!!
795260.850143 s: IPC: Init ... !!!
795260.880532 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
795260.887912 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
795260.887990 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
795260.888000 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
795260.888011 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
795260.888994 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
795260.889153 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
795260.889252 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
795260.889349 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
795260.889362 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
795260.889376 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO

# NETWORK_INIT_TIME = 668.89 (in ms, c7x @1GHz)
----------------------- TIDL Process with TARGET DATA FLOW ------------------------

# NETWORK_EXECUTION_TIME = 332.95 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) = 0.00, 0.00, 0.00 (in Mega Bytes/frame) ... .... .....APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
795261.862947 s: IPC: Deinit ... !!!
795261.863855 s: IPC: DeInit ... Done !!!
795261.863877 s: MEM: Deinit ... !!!
795261.863892 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
795261.863902 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes
795261.863910 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
795261.863923 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!
root@am69-sk:/opt/tidl_test#

0 An Dao 4 months ago in reply to Chris Tsongas

Prodigy 170 points

Hi Chris,

I’m using edgeai-tidl-tools version 10_01_00_02, and the build is running on the EVM.

Best regards,

0 Chris Tsongas 4 months ago in reply to An Dao

TI__Genius 14960 points

Hi An,

My question is where is the build performed (i.e., where did you compile the binary)? On the EVM or on the PC?

Regards,

Chris

0 An Dao 4 months ago in reply to Chris Tsongas

Prodigy 170 points

Hi Chris,

I compiled the binary directly on the EVM and I’m also running it there.

Best regards,
An

0 Chris Tsongas 3 months ago in reply to An Dao

TI__Genius 14960 points

Hi An,

I rebuilt your application on the device and compiled resnet101 model in a 10_01_00_02 environment and everything is working. It produces the out files in edgeai-tidl-tools/tidl_tools/out. The directory out/ must be created before the import. Please use the following import configuration file and scp -r out to <your_device_ip>/opt/arrow (example: scp -r out root@128.247.79.163:/opt/arrow) on the device.

Import:

./tidl_model_import.out import_resnet101.txt
========================= [Model Compilation Started] =========================

Model compilation will perform the following stages:
1. Parsing
2. Graph Optimization
3. Quantization & Calibration
4. Memory Planning

============================== [Version Summary] ==============================

-------------------------------------------------------------------------------
| TIDL Tools Version | 10_01_00_01 |
-------------------------------------------------------------------------------
| C7x Firmware Version | 10_00_02_00 |
-------------------------------------------------------------------------------

ONNX model (Proto) file : /shared/resnet101_simp.onnx
TIDL network file : out/tidl_net.bin
TIDL IO info file : out/tidl_io_buff
Current ONNX OpSet version : 11
============================ [Optimization started] ============================

[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_1's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_4's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_10's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_15's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_20's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_26's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_31's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_36's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_41's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_46's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_51's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_56's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_61's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_66's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_71's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_76's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_81's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_86's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_92's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_97's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_102's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_107's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_112's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_117's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_122's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_127's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_132's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_137's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_142's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_147's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_152's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_157's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_162's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_167's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_172's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_177's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_182's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_187's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_192's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_197's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_202's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_207's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_212's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_217's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_222's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_227's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_232's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_237's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_243's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_248's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: Batch Norm Layer BatchNormalization_254's coeff cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random

----------------------------- Optimization Summary -----------------------------
---------------------------------------------------------------------------------
| Layer | Nodes before optimization | Nodes after optimization |
---------------------------------------------------------------------------------
| TIDL_PReLULayer | 50 | 0 |
| TIDL_EltWiseLayer | 49 | 49 |
| TIDL_ConvolutionLayer | 103 | 103 |
| TIDL_FlattenLayer | 1 | 1 |
| TIDL_InnerProductLayer | 1 | 1 |
| TIDL_BatchNormLayer | 51 | 100 |
---------------------------------------------------------------------------------

=========================== [Optimization completed] ===========================

-------- Running Calibration in Float Mode to Collect Tensor Statistics --------
[=============================================================================] 100 %

------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------
[=============================================================================] 100 %

==================== [Quantization & Calibration Completed] ====================

========================== [Memory Planning Started] ==========================

------------------------- Network Compiler Traces ------------------------------
Successful Memory Allocation
Successful Workload Creation

========================= [Memory Planning Completed] =========================

[TIDL Import] WARNING: Size larger than (2048*2048) in [Gemm_253] is not optimal
======================== Subgraph Compiled Successfully ========================

Fullscreen import_resnet101.txt Download

modelType           = 2
numParamBits            = 8
numFeatureBits          = 8


inputNetFile        = "/shared/resnet101_simp.onnx"
outputNetFile = "out/tidl_net.bin"
outputParamsFile = "out/tidl_io_buff"


addDataConvertToNet = 0
foldPreBnConv2D         = 0
rawDataInElementType = 0
inElementType = 0
outElementType = 0


tidlStatsTool = "./PC_dsp_test_dl_algo.out"
perfSimTool = "./ti_cnnperfsim.out" 
graphVizTool = "./tidl_graphVisualiser.out"

inWidth             = 112
inHeight            = 112
inNumChannels       = 3
numFrames           = 1
inFileFormat        = 0
inData  =   "./airshow.jpg"
perfSimConfig = "device_config.cfg"
debugTraceLevel         = 0

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/791/resnet101.onnx

root@am69-sk:/home/root/edgeai-tidl-tools/bin/Release# ./tidlrt_resnet -f /opt/arrow/out --image face_aligned_1.jpg
Model Files names : /opt/arrow/out/tidl_net.bin,/opt/arrow/out/tidl_io_buff1.bin
APP: Init ... !!!
81266.466307 s: MEM: Init ... !!!
81266.466366 s: MEM: Initialized DMA HEAP (fd=5) !!!
81266.466511 s: MEM: Init ... Done !!!
81266.466533 s: IPC: Init ... !!!
81266.501083 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
81266.509069 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
81266.509174 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
81266.509187 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
81266.509197 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
81266.510039 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
81266.510170 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
81266.510263 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
81266.510369 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
81266.510381 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
81266.510394 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO
invoked
average time: 332.367 ms
Feature vector statistics:
Dimension: 512
Min value: -3.00667e+23
Max value: 2.8864e+23
Mean value: 1.70299e+21
Feature vector saved to face_aligned_1.jpg.features.txt
APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
81267.488617 s: IPC: Deinit ... !!!
81267.489629 s: IPC: DeInit ... Done !!!
81267.489661 s: MEM: Deinit ... !!!
81267.489672 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
81267.489680 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes
81267.489686 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
81267.489696 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!

Regards,

Chris

0 Le Tung 3 months ago in reply to Chris Tsongas

Prodigy 120 points

Hi Chris,

After following your instructions, we were able to run our custom-trained model. However, there is still one issue: why do we have to reboot the device before the custom-trained model can run? If we don't reboot, the same error as before still occurs.

Thanks and best regards,

0 Khoa Nam Nguyen 3 months ago in reply to Chris Tsongas

Prodigy 70 points

Hi Chris,

1. After following your instructions, we were able to run our custom-trained model. However, there is still one issue: why do we have to reboot the device before the custom-trained model can run? If we don't reboot, the same error as before still occurs. Could you help us analyze where the issue lies?

2. The average time: 332.367 ms is too high — our target is approximately 10 ms. On the QCM6490, this same model achieves 10.22 ms inference time.
We tried adjusting msmcSizeKB, ENABLE_PERSIST_WT_ALLOC, and numBatches as suggested in the following guide, but the result didn’t improve. Could you suggest how we might optimize further?
https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_00_00_12/exports/docs/tidl_j7_08_00_00_10/ti_dl/docs/user_guide_html/md_tidl_fsg_steps_to_debug_performance.html

3. Does this log mean that MSMC SRAM is overflowing and using DDR? Is there any way to fix this?:
81267.489672 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
81267.489680 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes

0 Chris Tsongas 3 months ago in reply to Khoa Nam Nguyen

TI__Genius 14960 points

Hi,

1. If it fails after the 1st run, please look into your custom driver program. There is a bug in it and it is not cleaning up (releasing resources) after execution.

2. This will take some optimization in model and execution and should be in another E2E.

3. Probably, but your import file has compileConstraintsFlag. Look for a csv file in the output directory to get a clear idea where the time is taken up and where most of the execution time is taken.

Regards,

Chris

0 Khoa Nam Nguyen 3 months ago in reply to Chris Tsongas

Prodigy 70 points

Kính thưa Chris Tsongas ,

Tôi đã tạo một chủ đề mới: PROCESSOR-SDK-AM69A: Optimization iResnet101

Cảm ơn!

0 Christina Kuruvilla 3 months ago in reply to Khoa Nam Nguyen

TI__Expert 5680 points

Hi Khoa,

I have assigned the new E2e to a person on our team. I will go ahead and close this thread since it seems as if it has been answered. Please reopen if not.

Warm regards,

Christina

0 Khoa Nam Nguyen 3 months ago in reply to Christina Kuruvilla

Prodigy 70 points

Sorry I mistakenly ticked Resolved so I will create a new topic: PROCESSOR-SDK-AM69A: Optimization Resnet101

0 Le Tung 3 months ago in reply to Chris Tsongas

Prodigy 120 points

Hi Chris ,

When I flashed the release image from TI's website using SDK version 10_01 and ran the test for the first time, I encountered the same error. However, after rebooting the device, the error was gone

0 Khoa Nam Nguyen 3 months ago in reply to Le Tung

Prodigy 70 points

Chris Tsongas Hi Chris, We are waiting for your response.

0 Chris Tsongas 3 months ago in reply to Khoa Nam Nguyen

TI__Genius 14960 points

Hi Khoa,

There is something wrong in the CPP program you provided and it will run the first time and fails subsequent times. Also, we know that this works with TIDL driver and the same model. This means the app is not cleaning up after itself and we cannot help with a custom application.

Regards,

Chris

0 Le Tung 3 months ago in reply to Chris Tsongas

Prodigy 120 points

Hi Chris

When we use TI_DEVICE_armv8_test_dl_algo_host_rt.out in /opt/tidl_test on the device, we encounter the same error — it's not caused by adding our own C++ code.

thanks and best regard,

0 Chris Tsongas 2 months ago in reply to Le Tung

TI__Genius 14960 points

Hi Le Tung,

I just ran the resnet101_simp model 3 times consecutively on an AM69A EVM running 10_01_00_02 without any problem.

./TI_DEVICE_armv8_test_dl_algo_host_rt.out s:infer_evm.txt

Processing config file #0 : infer_evm.txt
APP: Init ... !!!
140.281651 s: MEM: Init ... !!!
140.281709 s: MEM: Initialized DMA HEAP (fd=5) !!!
140.281877 s: MEM: Init ... Done !!!
140.281896 s: IPC: Init ... !!!
140.315890 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
140.325556 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
140.327855 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
140.327882 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
140.327889 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
140.330293 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
140.330441 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
140.330525 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
140.330606 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
140.330619 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
140.330633 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO

# NETWORK_INIT_TIME = 693.86 (in ms, c7x @1GHz)
----------------------- TIDL Process with TARGET DATA FLOW ------------------------

# NETWORK_EXECUTION_TIME = 332.61 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) = 0.00, 0.00, 0.00 (in Mega Bytes/frame) ... .... .....APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
141.328831 s: IPC: Deinit ... !!!
141.329626 s: IPC: DeInit ... Done !!!
141.329650 s: MEM: Deinit ... !!!
141.329664 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
141.329673 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes
141.329681 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
141.329692 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!
root@am69-sk:/opt/tidl_test# ./TI_DEVICE_armv8_test_dl_algo_host_rt.out s:infer_evm.txt

Processing config file #0 : infer_evm.txt
APP: Init ... !!!
146.284267 s: MEM: Init ... !!!
146.284320 s: MEM: Initialized DMA HEAP (fd=5) !!!
146.284463 s: MEM: Init ... Done !!!
146.284481 s: IPC: Init ... !!!
146.316466 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
146.324037 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
146.324118 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
146.324130 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
146.324145 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
146.324851 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
146.325002 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
146.325099 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
146.325197 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
146.325215 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
146.325226 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO

# NETWORK_INIT_TIME = 675.25 (in ms, c7x @1GHz)
----------------------- TIDL Process with TARGET DATA FLOW ------------------------

# NETWORK_EXECUTION_TIME = 332.60 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) = 0.00, 0.00, 0.00 (in Mega Bytes/frame) ... .... .....APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
147.303141 s: IPC: Deinit ... !!!
147.304152 s: IPC: DeInit ... Done !!!
147.304184 s: MEM: Deinit ... !!!
147.304197 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
147.304206 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes
147.304213 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
147.304227 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!
root@am69-sk:/opt/tidl_test# ./TI_DEVICE_armv8_test_dl_algo_host_rt.out s:infer_evm.txt

Processing config file #0 : infer_evm.txt
APP: Init ... !!!
150.114551 s: MEM: Init ... !!!
150.114603 s: MEM: Initialized DMA HEAP (fd=5) !!!
150.114744 s: MEM: Init ... Done !!!
150.114762 s: IPC: Init ... !!!
150.147036 s: IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
150.153659 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
150.153729 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR
150.153740 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING
150.153751 s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO
150.154320 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-0
150.154421 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-1
150.154501 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-2
150.154578 s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:134] Added target MPU-3
150.154590 s: VX_ZONE_INFO: [tivxInitLocal:126] Initialization Done !!!
150.154602 s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO

# NETWORK_INIT_TIME = 676.16 (in ms, c7x @1GHz)
----------------------- TIDL Process with TARGET DATA FLOW ------------------------

# NETWORK_EXECUTION_TIME = 332.56 (in ms, c7x @1GHz) with DDR_BANDWIDTH (Read + Write) = 0.00, 0.00, 0.00 (in Mega Bytes/frame) ... .... .....APP: Deinit ... !!!
REMOTE_SERVICE: Deinit ... !!!
REMOTE_SERVICE: Deinit ... Done !!!
151.134547 s: IPC: Deinit ... !!!
151.135497 s: IPC: DeInit ... Done !!!
151.135543 s: MEM: Deinit ... !!!
151.135557 s: DDR_SHARED_MEM: Alloc's: 7 alloc's of 68044312 bytes
151.135567 s: DDR_SHARED_MEM: Free's : 7 free's of 68044312 bytes
151.135578 s: DDR_SHARED_MEM: Open's : 0 allocs of 0 bytes
151.135592 s: MEM: Deinit ... Done !!!
APP: Deinit ... Done !!!

I have included the model and import log. Please try the model I have included.

Regards,

Chris

Fullscreen resnet101.log Download

./tidl_model_import.out import_resnet101    
========================= [Model Compilation Started] =========================

Model compilation will perform the following stages:
1. Parsing
2. Graph Optimization
3. Quantization & Calibration
4. Memory Planning

============================== [Version Summary] ==============================

-------------------------------------------------------------------------------
|          TIDL Tools Version          |              10_01_00_01             |
-------------------------------------------------------------------------------
|         C7x Firmware Version         |              10_00_02_00             |
-------------------------------------------------------------------------------

ONNX model (Proto) file      : /shared/resnet101_simp.onnx  
TIDL network file            : out/tidl_net.bin  
TIDL IO info file            : out/tidl_io_buff  
Current ONNX OpSet version   : 11  
============================ [Optimization started] ============================

[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_1's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_4's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_10's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_15's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_20's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_26's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_31's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_36's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_41's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_46's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_51's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_56's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_61's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_66's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_71's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_76's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_81's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_86's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_92's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_97's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_102's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_107's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_112's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_117's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_122's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_127's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_132's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_137's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_142's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_147's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_152's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_157's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_162's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_167's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_172's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_177's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_182's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_187's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_192's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_197's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_202's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_207's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_212's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_217's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_222's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_227's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_232's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_237's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_243's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: PReLU Layer PRelu_248's bias cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random
[TIDL Import] [PARSER] WARNING: Batch Norm Layer BatchNormalization_254's coeff cannot be found(or not match) in coeff file, Random bias will be generated only for evaluation usage. Results are all random

----------------------------- Optimization Summary -----------------------------
---------------------------------------------------------------------------------
|          Layer         | Nodes before optimization | Nodes after optimization |
---------------------------------------------------------------------------------
| TIDL_PReLULayer        |                        50 |                        0 |
| TIDL_EltWiseLayer      |                        49 |                       49 |
| TIDL_ConvolutionLayer  |                       103 |                      103 |
| TIDL_FlattenLayer      |                         1 |                        1 |
| TIDL_InnerProductLayer |                         1 |                        1 |
| TIDL_BatchNormLayer    |                        51 |                      100 |
---------------------------------------------------------------------------------

=========================== [Optimization completed] ===========================


-------- Running Calibration in Float Mode to Collect Tensor Statistics --------
[=============================================================================] 100 %

------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------
[=============================================================================] 100 %

==================== [Quantization & Calibration Completed] ====================

========================== [Memory Planning Started] ==========================


------------------------- Network Compiler Traces ------------------------------
Successful Memory Allocation
Successful Workload Creation

========================= [Memory Planning Completed] =========================

[TIDL Import]  WARNING: Size larger than (2048*2048) in [Gemm_253] is not optimal
======================== Subgraph Compiled Successfully ========================


root@9bfe94da4395:/home/root/tidl_tools# ./PC_dsp_test_dl_algo.out s:infer
infer_surav.txt                inference_config_merlinv6      inference_hirain.txt           inference_test
infer_surav_evm.txt            inference_config_new           inference_pc.txt               
inference_chris                inference_config_qdq_victoria  inference_resnet101            
inference_config_merlinv3      inference_gather.txt           inference_sri                  
root@9bfe94da4395:/home/root/tidl_tools# ./PC_dsp_test_dl_algo.out s:inference_resnet101 

Processing config file #0 : inference_resnet101 
 ----------------------- TIDL Process with REF_ONLY FLOW------------------------ 

#    0 . .. T    7968.34  .... ..... ... .... .....root@9bfe94da4395:/home/root/tidl_tools# 
root@9bfe94da4395:/home/root/tidl_tools# 
root@9bfe94da4395:/home/root/tidl_tools# scp -r out root@128.247.79.154:/opt/tidl_test/
The authenticity of host '128.247.79.154 (128.247.79.154)' can't be established.
RSA key fingerprint is SHA256:M6yDy2N/OCgV0dV0Telnqp4FQiDvhJlIyRQWfkAYXoQ.
This key is not known by any other names
Are you sure you want to continue connecting (yes/no/[fingerprint])? yes
Warning: Permanently added '128.247.79.154' (RSA) to the list of known hosts.
import_resnet101.qunat_stats_config.txt                                                                100%  539   247.3KB/s   00:00    
tidl_net.bin.layer_info.txt                                                                            100% 3342     1.5MB/s   00:00    
tidl_net.bin_netLog.txt                                                                                100%   70KB   6.1MB/s   00:00    
tidl_net.bin_paramDebug.csv                                                                            100%  986KB   6.0MB/s   00:00    
perfSimInfo.bin                                                                                        100%   29MB   8.0MB/s   00:03    
wlinfolog_0.txt                                                                                        100%  651KB   8.2MB/s   00:00    
bufinfolog_0.txt                                                                                       100%  252KB   3.4MB/s   00:00    
bufinfolog_0.csv                                                                                       100%   65KB   1.3MB/s   00:00    
tidl_net.bin                                                                                           100%   64MB   8.0MB/s   00:07    
import_resnet101_stats_tool_out.bin                                                                    100%  512   255.7KB/s   00:00    
tidl_out.bin                                                                                           100%  512   278.4KB/s   00:00    
import_resnet101.perf_sim_config.txt                                                                   100%  781   431.4KB/s   00:00    
tidl_io_buff1.bin                                                                                      100%  370KB   4.3MB/s   00:00    
tidl_net.bin.svg                                                                                       100% 2891KB   7.4MB/s   00:00

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/791/resnet101_5F00_simp.onnx

Processors

Processors forum

PROCESSOR-SDK-AM69A: Issue when switching to using the tidlcpp library