Clasificarea imaginilor

Modelarea imaginilor cu Keras

Ariel Rokem

Senior Data Scientist, University of Washington

Clasificarea imaginilor

Clasificarea imaginilor: antrenare

Clasificarea imaginilor: evaluare

Reprezentarea claselor: codificare one-hot

labels = ["shoe", "dress", "shoe", "t-shirt", 
          "shoe", "t-shirt", "shoe", "dress"]

Reprezentarea claselor: codificare one-hot

array([[0., 0., 1.],    <= shoe
       [0., 1., 0.],    <= dress
       [0., 0., 1.],    <= shoe
       [1., 0., 0.],    <= t-shirt
       [0., 0., 1.],    <= shoe
       [1., 0., 0.],    <= t-shirt
       [0., 0., 1.],    <= shoe
       [0., 1., 0.]])   <= dress

Codificare one-hot

categories = np.array(["t-shirt", "dress", "shoe"])

n_categories = 3
ohe_labels = np.zeros((len(labels), n_categories))

for ii in range(len(labels)):

    jj = np.where(categories == labels[ii])

    ohe_labels[ii, jj] = 1

Codificare one-hot: testarea predicțiilor

test

array([[0., 0., 1.], 
       [0., 1., 0.], 
       [0., 0., 1.], 
       [0., 1., 0.], 
       [0., 0., 1.],
       [0., 0., 1.],
       [0., 0., 1.],
       [0., 1., 0.]])

(test * prediction).sum()

6.0

prediction

array([[0., 0., 1.], 
       [0., 1., 0.], 
       [0., 0., 1.], 
       [1., 0., 0.], <= incorrect
       [0., 0., 1.],
       [1., 0., 0.], <= incorrect
       [0., 0., 1.], 
       [0., 1., 0.]])

Să exersăm!

Modelarea imaginilor cu Keras