python - append numpy arrays using pandas

Question

i have the following dataset as files,

      image          voice
0   -64VH3Apae0.npy -64VH3Apae0.npy
1   -6zgNLiO0hw.npy -6zgNLiO0hw.npy
2   -8LOU7c__hc.npy -8LOU7c__hc.npy
3   -9vlXXK491c.npy -9vlXXK491c.npy
4   -f2d0VPU3Zg.npy -f2d0VPU3Zg.npy
... ... ...
1230    _S_OkdQlox8.npy _S_OkdQlox8.npy
1231    _ua2HGzbvn8.npy _ua2HGzbvn8.npy
1232    _XexneTqik4.npy _XexneTqik4.npy
1233    _xg8VLVfYSI.npy _xg8VLVfYSI.npy
1234    _YAclFXGGpM.npy _YAclFXGGpM.npy

i want to load each file from image tag and append to array called im_features, i tried to do the following,

def createImageFeatures(table):
    file_im_feature = np.load(image_features + str(table.image), allow_pickle=True)
    im_features = np.append(im_features, file_im_feature[0], axis=1) 

dataset.apply(createImageFeatures, axis=1)

but got

UnboundLocalError: local variable 'im_features' referenced before assignment example:

im_features1 = np.load(image_features + '-64VH3Apae0.npy', allow_pickle=True)
im_features2 = np.load(image_features + '4QgpaBUYeI4.npy', allow_pickle=True)
print(im_features1[0])
[[0.28235294]
 [0.28627451]
 [0.32156863]
 ...
 [0.17647059]
 [0.17647059]
 [0.18431373]]
print(im_features2[0])
[[1.        ]
 [1.        ]
 [1.        ]
 ...
 [0.16862745]
 [0.15294118]
 [0.14117647]]
np.append(im_features1[0], im_features2[0], axis=1)
im_features = np.append(im_features1[0], im_features2[0], axis=1)
print(im_features)
[[0.28235294 1.        ]
 [0.28627451 1.        ]
 [0.32156863 1.        ]
 ...
 [0.17647059 0.16862745]
 [0.17647059 0.15294118]
 [0.18431373 0.14117647]]

how to achieve such thing.

Answer 1

等待大神答复