I noticed that edge impulse uses a tflite file, and was wondering if its possible to run a pretrained one. I want to do this since I have tried and tried again and despite the fact that the web browser shows an accuracy of around 90%, the accuracy when its deployed is terrible. Everything is completely random and broken. I know that the pretrained tflite models work, so I want to try that. Could you let me know how this would be done?(if possible) Thanks!
Is there any ways:
- Can I import and Use my own model (tflite quantized model) for transfer learning in the case of image classification?
- I can visualize the model just like netron helps visualize .tflite model?
Is this a reply? Are you just restating what I said? I’m confused.