detection: numpy.AxisError #1
Description
I am trying to train you mothe to detect daphnia in a beaker. After training the following error happens in my terminal. What could be the reason?
For trial I just used a super simple video with only two daphnia in front of a grey background.
Epoch 24/25 71/71 [==============================] - 0s 4ms/step - loss: 0.5962 - accuracy: 0.7606 - val_loss: 0.5960 - val_accuracy: 0.7292 Epoch 25/25 71/71 [==============================] - 0s 4ms/step - loss: 0.5657 - accuracy: 0.7606 - val_loss: 0.6101 - val_accuracy: 0.7292 [UPDATING.....]1th frame detected and stored Exception in Tkinter callback Traceback (most recent call last): File "/Users/alpenblick/opt/miniconda3/envs/mothe/lib/python3.6/tkinter/__init__.py", line 1705, in __call__ return self.func(*args) File "mothe_gui.py", line 179, in click_run inst.detection(path, path1) File "/Users/alpenblick/opt/miniconda3/envs/mothe/lib/python3.6/site-packages/mothe/pipe.py", line 335, in detection Pclass = np.argmax(np.round(pred),axis=1) File "<__array_function__ internals>", line 6, in argmax File "/Users/alpenblick/opt/miniconda3/envs/mothe/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 1153, in argmax return _wrapfunc(a, 'argmax', axis=axis, out=out) File "/Users/alpenblick/opt/miniconda3/envs/mothe/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 61, in _wrapfunc return bound(*args, **kwds) numpy.AxisError: axis 1 is out of bounds for array of dimension 1
I only used around 30pictures for "yes" and 90 for "no". Could this be the error. I was just trying it out before I try to get the 10k pictures. :)
EDIT: It runs with bb_model_N12.h5py and Blackbuck4.MOV with no problem, so I guess my model is just not good enough and needs more data to train on? The detection seems not to good though... Lots of grass is detected as a BB
Thanks in advance and with best regards,
Sebastian