Failed to load classification error

#1

Good day! New to Edge Impulse. Regardless if I attempt to classify on mobile phone or my computer I get Failed to load classifcation error -3. Any assistance with this will be appreciated.

mike

#2

Hi @pendulous,

Could you share your project ID for this specific project?

Thanks,
Aurelien

#3

Hi @pendulous, during training there’s an error and because of this the arena size of your model cannot be calculated correctly which then leads to an initialization error when running the model.

IndexError: index 186 is out of bounds for axis 0 with size 186
Traceback (most recent call last):

  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/ops/script_ops.py", line 244, in __call__
    ret = func(*args)

  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/autograph/impl/api.py", line 302, in wrapper
    return func(*args, **kwargs)

  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 827, in generator_py_func
    values = next(generator_state.get_iterator(iterator_id))

  File "./resources/libraries/ei_tensorflow/training.py", line 110, in gen
    yield X_values[ix], Y_values[ix]

  File "/usr/local/lib/python3.7/site-packages/numpy/core/memmap.py", line 331, in __getitem__
    res = super(memmap, self).__getitem__(index)

IndexError: index 186 is out of bounds for axis 0 with size 186

I’ve asked @dansitu to take a look and see what the underlying bug is!

#4

Hi, looked like some corruption between generating features and training the network that lead to the features and labels having a different length… I’ve regenerated the features on your project, and you should be good to go now. Sorry for the inconvenience!

Note that I do see a large quantization error on your model, I think due to class imbalance, so you might want to play around a bit with balancing out your dataset better.

#5

thank you so much for getting back to me. I really appreciate it. Your product is amazing!

1 Like
#6

It was sorted out. Thank you for responding! Great platform!

2 Likes