Ignoring WARNING: Num classes mismatch

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
Jetson
• DeepStream Version
Latest version
• JetPack Version (valid for Jetson only)
Latest

I’m wondering if there’s a way to stop the “WARNING: Num classes mismatch” error that appears when setting the num_classes attribute in a config file lower than a network is configured for. I’m using an Object detection model to detect people, but don’t want to show the other classes so I’m intentionally only setting the config to look for the People class. The issue is the warning is shown in the console when I run the network. Is there a way to stop this message from showing?

3

Hey, What’s the postprocess parser are you using?

4

Where would I find this? I’m using the Python API if that’s any help

5

OK, which model are you using and can you share the model config with us?

6

I’m using YoloV3. I know the line that causes the error, since I intentionally set it to use less classes than its configured to run with. I’m just curious if I can stop the warning message from outputting. The file is below.

[property]
gpu-id=0
net-scale-factor=0.0039215697906911373
model-color-format=0
custom-network-config=/yolov3.cfg
model-file=/yolov3.weights
model-engine-file=/model_b1_gpu0_fp16.engine
labelfile-path=labels.txt
batch-size=1
network-mode=2
num-detected-classes=1
interval=2
gie-unique-id=1
process-mode=1
network-type=0
cluster-mode=4
maintain-aspect-ratio=0
parse-bbox-func-name=NvDsInferParseYolo
custom-lib-path=/nvdsinfer_custom_impl_Yolo/libnvdsinfer_custom_impl_Yolo.so
engine-create-func-name=NvDsInferYoloCudaEngineGet

[class-attrs-all]
pre-cluster-threshold=0.25
7

You need to disable it in this library

8

Thanks for the quick response, I’ll look at doing this