We get the following error messages when attempting run Onbot Java:
Internal error: cannot create interpreter; Didn’t find op for builtin opcode ‘COV_2D’ version ‘5’
Didn’t find of for builtin opcode 'DEPTHWISE_2D VERSION ‘6’
Registration failed.
Sometimes an error message can provide enough context to determine what needs to be done to fix the problem. Other times we just need to see your code. This is a case of the latter.
Please post a simplified version of your code that still reproduces the problem.
We started with sample Opmode “ConceptTensorFlowObjectDetectionWebcam”. We were able to run this and identify the default rings from last years Ultimate Goal game.
We then changed a few line per the manual:
private static final String TFOD_MODEL_ASSET =
“/sdcard/FIRST/tflitemodels/OurModel.tflite”;
Change from “loadModelFromAsset” to “loadModelFromFile(TOFD_MODEL_ASSET, LABELS);"
private static final string[] LABELS = {
“OurModel”
};
Then we got the above error message.
Suspected Problem and Fix
This is most likely because you are running a version of the SDK that is 6.x. I assume this because you are using the TFOD models/opmodes from last year. FTC-ML produces TF2 Models. 6.x only has support for TF1 models. If you wish to use the models form FTC-ML please upgrade to 7.0. In addition in order to compete this year you need a version of the SDK that is 7.x ( <RS03>).
Upgrading Android Studios
The repository with the latest SDK can be found here. You will need to merge with your codebase or simply download it and transfer over your opmodes. This will be the easiest for you as long as you did not change anything other than your opmodes and do not use git.
Upgrading OBJ or Blocks
Consult the REV Hardware Client.
- Uday, Team 7350
1 Like
That fixed it, thanks!