I use the TensorFlowLite_ESP32 and train a model to judge garbage
I try a long time, now it can upload but I have new problem
serial monitor display the
13:19:52.961 -> Didn't find op for builtin opcode 'HARD_SWISH' version '1'. An older version of this builtin might be supported. Are you using an old TFLite binary with a newer model?
13:19:52.961 ->
13:19:52.961 -> Failed to get registration from op code HARD_SWISH
13:19:53.007 ->
13:19:53.007 -> ❌ 張量內存分配失敗!(fail)
the GPT say the psarm need enable ,I confirm the arduino tools
and type a code display the psram have work,
I try over and over but still can't fix
Is my model have error? or the something have crush
please have me my brain gonna broken HAHA
I konw but my model so large I already Compressed archive to 1MB ,
use built-in RAM will over the size,I use the code show the psram have work but when I upload my code that's dosen't work
Sorry, I have no idea what you are saying. I deal in source code and error logs all contained in code tags.
You may benefit from reading the pinned post re 'How to get the most from the forum'