Unable to deploy model

Question

Hi,

I am in the early stages of building a CV demo using Qualcomm chipset and Greengrass V2.
My app contains the following components components.

![Enter image description here](/media/postImages/original/IMSBZMNHSET3KEC1_Oa6cxDQ)

My system is unable to do any inference though it is able to send empty MQTT packages to the cloud.

I am seeing the following error when I am deploying a model using the variant.DLR

'''2022-10-10T20:21:22.232Z [WARN] (Copier) variant.DLR: stderr. ++ PATH=/greengrass/v2/work/variant.DLR/greengrass_ml/greengrass_ml_dlr_venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.232Z [WARN] (Copier) variant.DLR: stderr. ++ export PATH. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.232Z [WARN] (Copier) variant.DLR: stderr. ++ '[' -n '' ']'. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.233Z [WARN] (Copier) variant.DLR: stderr. ++ '[' -z '' ']'. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.233Z [WARN] (Copier) variant.DLR: stderr. ++ _OLD_VIRTUAL_PS1=. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.233Z [WARN] (Copier) variant.DLR: stderr. ++ '[' 'x(greengrass_ml_dlr_venv) ' '!=' x ']'. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.234Z [WARN] (Copier) variant.DLR: stderr. ++ PS1='(greengrass_ml_dlr_venv) '. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.234Z [WARN] (Copier) variant.DLR: stderr. ++ export PS1. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.234Z [WARN] (Copier) variant.DLR: stderr. ++ '[' -n /bin/bash -o -n '' ']'. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.235Z [WARN] (Copier) variant.DLR: stderr. ++ hash -r. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.235Z [WARN] (Copier) variant.DLR: stderr. ++ python3 -c 'try:. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.236Z [WARN] (Copier) variant.DLR: stderr. import dlr. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.236Z [WARN] (Copier) variant.DLR: stderr. print(dlr.__version__). {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.236Z [WARN] (Copier) variant.DLR: stderr. except Exception as e:. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.237Z [WARN] (Copier) variant.DLR: stderr. print(e)'. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.647Z [WARN] (Copier) variant.DLR: stderr. + get_dlr_version=1.6.0. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.647Z [INFO] (Copier) variant.DLR: stdout. Skipping DLR installation as it already exists.. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.648Z [WARN] (Copier) variant.DLR: stderr. + [[ 1.6.0 == \1\.\6\.\0 ]]. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
2022-10-10T20:21:22.649Z [WARN] (Copier) variant.DLR: stderr. + return 0. {scriptName=services.variant.DLR.lifecycle.install.script, serviceName=variant.DLR, currentState=NEW}
'''

Any help would be appreciated.

Thanks,

Answer

The error you're seeing in the logs appears to be related to the installation of DLR (Deep Learning Runtime) during the deployment of your model using variant.DLR. From the logs, it seems that DLR version 1.6.0 is already installed, and the installation process is being skipped.

While the error message itself doesn't indicate a problem, it's important to verify that DLR is installed correctly and that it's compatible with your model and environment.

Here are some steps you can take to troubleshoot and resolve the issue:

Confirm that DLR is installed correctly on your system and that it's functioning as expected. You can check the installation status and version by running dlr --version or python3 -c "import dlr; print(dlr.__version__)".

Ensure that your model is compatible with the version of DLR installed on your system. Some models may require specific versions of DLR or have compatibility issues with certain versions.

: Double-check your deployment configuration to ensure that the correct version of DLR is specified, and that any dependencies or requirements are met.

Inspect Environment: Investigate any changes or updates to your environment that may have impacted the deployment process. This could include changes to system libraries, dependencies, or permissions.

Let me know if this helps

Unable to deploy model

관련 콘텐츠