FTE-BTE Food Experiment

[amyvanee]

My issue is about running the FTE-BTE Food Experiment

Reproducing code example:

dict_x = {}
for k in range(10):
    # Initialize data_x* with the first image in the first class, then concatenate to acquire all images from the first class
    food_class = os.listdir(os.path.join(data_dir,foods_sorted[10*k]))
    data_xk = [plt.imread(os.path.join(data_dir, foods_sorted[10*k], food_class[0]))]

    for i in range(1,1000):
            data_xk = np.concatenate([data_xk, [(plt.imread(os.path.join(data_dir, foods_sorted[10*k], food_class[i])))]])

    # Add to the initialized data_x* array until it contains all images from the 10 classes
    # Concatenating more than 10000 images per batch increases the run time by a lot
    for j in range(((k*10)+1),(10*(k+1))):
        food_class = os.listdir(os.path.join(data_dir,foods_sorted[j]))
        for i in range(0,1000):
            data_xk = np.concatenate([data_xk, [(plt.imread(os.path.join(data_dir, foods_sorted[j], food_class[i])))]])
            
    dict_x['data_x' + str(k+1)] = data_xk

Error message

ValueError: all the input array dimensions for the concatenation axis must match exactly, but along dimension 2, the array at index 0 has size 308 and the array at index 1 has size 512
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-7-8ad3416dbc2e> in <module>
      6 
      7     for i in range(1,1000):
----> 8             data_xk = np.concatenate([data_xk, [(plt.imread(os.path.join(data_dir, foods_sorted[10*k], food_class[i])))]])
      9 
     10     # Add to the initialized data_x* array until it contains all images from the 10 classes

<__array_function__ internals> in concatenate(*args, **kwargs)

ValueError: all the input array dimensions for the concatenation axis must match exactly, but along dimension 2, the array at index 0 has size 308 and the array at index 1 has size 512

Version information

• OS: Windows
• Python Version: 3.8.2
• Package Version: 0.0.1

Thank you!