Open winstonjthinker opened 3 weeks ago
same error, happening to me as well Didn't find op for builtin opcode 'FULLY_CONNECTED' version '11'. An older version of this builtin might be supported. Are you using Didn't find op for builtin opcode 'FULLY_CONNECTED' version '11' an old TFLite binary with a newer model?, but the interpreter at collab and kaggle process it without any problem., I have implemented all cures stated in online including builtin ops while converting, etc but none resolved it if managed to resolve please let me know,
Since this is an Arduino question, I have to ask: are you somehow using Flutter on the Arduino Nano? I haven't heard of that being possible, so any documentation you have on that would be awesome. Otherwise I'd expect you'd use TFLite for Microcontrollers, which is a whole other library/project unrelated to this one.
No, Sorry for posting in ardunio section, I am also facing the same issue in flutter android side, but in collab and kaggle i am able to successfully load and run inference to the tflite model, but here i am getting this error Didn't find op for builtin opcode 'FULLY_CONNECTED' version '11'. An older version of this builtin might be supported. Are you using Didn't find op for builtin opcode 'FULLY_CONNECTED' version '11' an old TFLite binary with a newer model?, the version used to convert the model is tf2.17
Didn't find op for builtin opcode 'FULLY_CONNECTED' version '1'. An older version of this builtin might be supported. Are you using Didn't find op for builtin opcode 'FULLY_CONNECTED' version '1' an old TFLite binary with a newer model?
Failed to get registration from op code FULLY_CONNECTED
Failed to allocate tensors. 07:31:13.216 -> Didn't find op for builtin opcode 'FULLY_CONNECTED' version '1'. An older version of this builtin might be supported. Are you using an old TFLite binary with a newer model? 07:31:13.216 -> 07:31:13.216 -> Failed to get registration from op code FULLY_CONNECTED 07:31:13.216 ->
07:31:13.216 -> Failed to allocate tensors.