can not reproduction acc with lfw dataset

[Ruanlt]

I try to reproduction acc with lfw dataset， but error occur.

Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/PIL/Image.py", line 2772, in fromarray
mode, rawmode = _fromarray_typemap[typekey]
KeyError: ((1, 1, 3), '<f4')

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "deepface_test.py", line 33, in
obj = DeepFace.verify(img1, img2, model_name = 'Dlib', model = dlib_model, enforce_detection=False, distance_metric = 'euclidean')
File "/usr/local/lib/python3.6/dist-packages/deepface/DeepFace.py", line 152, in verify
, detector_backend = detector_backend)
File "/usr/local/lib/python3.6/dist-packages/deepface/commons/functions.py", line 454, in preprocess_face
img = align_face(img = img, detector_backend = detector_backend)
File "/usr/local/lib/python3.6/dist-packages/deepface/commons/functions.py", line 437, in align_face
img = alignment_procedure(img, left_eye, right_eye)
File "/usr/local/lib/python3.6/dist-packages/deepface/commons/functions.py", line 360, in alignment_procedure
img = Image.fromarray(img)
File "/usr/local/lib/python3.6/dist-packages/PIL/Image.py", line 2774, in fromarray
raise TypeError("Cannot handle this data type: %s, %s" % typekey) from e
TypeError: Cannot handle this data type: (1, 1, 3), <f4

What shoud I do？

[serengil]

what exactly image pairs in your experiment? could you share your whole code?

you got same error when you test it for another model, e.g. Facenet?

[Ruanlt]

what exactly image pairs in your experiment? could you share your whole code?

you got same error when you test it for another model, e.g. Facenet?

I copy the code from here(https://github.com/serengil/tensorflow-101/blob/master/python/LFW.ipynb), and use Facenet got same error.

[serengil]

it seems mtcnn causes an error. You can skip the error if you set the detector backend to different to mtcnn.

DeepFace.verify(img1, img2, detector_backend = 'opencv', enforce_detection = False)

I will study why mtcnn causes this error

[serengil]

yes, mtcnn find facial area in a scale of 1, 1, 3 that's not true. It seems mtcnn related error. The best way is using different detector.