بتل کد Battle Codeدارد باز می‌شود…

جواب «Step 6: Mistakes and Your Own Digit» چالش 6 از جلسه‌ی 47

سؤال

قدم ۶: الف) هر تصویر آزمون که شبکه اشتباه حدس زده را با print("true", ..., "predicted", ...) و بعد show چاپ کن؛ چهار اشتباه است و اولی true 9 predicted 8 است. ب) تابع read_drawing(rows) را بنویس که پنج رشته‌ی # و . را به آرایه‌ای با شکل (1, 25) تبدیل کند، و حدس شبکه را چاپ کن: my digit: 3 (97% sure).

راهنما

اشتباه یعنی guesses[i] != yt[i]. برای رسم دستی، "".join(rows) همه‌ی ۲۵ نویسه را پشت هم می‌دهد؛ np.array([pixels]) شکل (1, 25) می‌سازد.

یک جواب درست

import random
import numpy as np

SHAPES = '''
.###. ..#.. .###. ####. #..#. ##### .###. ##### .###. .###.
#...# .##.. #...# ....# #..#. #.... #.... ....# #...# #...#
#...# ..#.. ..##. .###. ##### ####. ####. ...#. .###. .####
#...# ..#.. .#... ....# ...#. ....# #...# ..#.. #...# ....#
.###. .###. ##### ####. ...#. ####. .###. ..#.. .###. .###.
'''


def make_digits(copies, noise, seed):
    random.seed(seed)
    rows = SHAPES.strip().splitlines()
    X, y = [], []
    for c in range(copies):
        for d in range(10):
            pixels = [1.0 if ch == "#" else 0.0 for ch in "".join(r.split()[d] for r in rows)]
            for i in random.sample(range(25), noise):
                pixels[i] = 1.0 - pixels[i]
            X.append(pixels)
            y.append(d)
    return np.array(X), np.array(y)

X, y = make_digits(20, 2, 1)     # 200 training digits
Xt, yt = make_digits(10, 2, 2)   # 100 test digits



def show(pixels):
    for r in range(5):
        row = pixels[r * 5:r * 5 + 5]
        print("".join("#" if p > 0.5 else "." for p in row))


print(X.shape, Xt.shape)
print("label:", int(y[3]))
show(X[3])


random.seed(0)


def rand_matrix(rows, cols):
    return np.array([[random.uniform(-0.5, 0.5) for c in range(cols)] for r in range(rows)])


net = {"W1": rand_matrix(25, 16), "b1": np.zeros(16),
       "W2": rand_matrix(16, 10), "b2": np.zeros(10)}


def softmax(z):
    e = np.exp(z - z.max(axis=1, keepdims=True))
    return e / e.sum(axis=1, keepdims=True)


def forward(x):
    h = np.maximum(0, x @ net["W1"] + net["b1"])
    p = softmax(h @ net["W2"] + net["b2"])
    return h, p


h, p = forward(X[:3])
print(p.shape)
print(np.round(p[0], 2))
print(round(float(p[0].sum()), 3))



def accuracy(Xs, ys):
    h, p = forward(Xs)
    guesses = p.argmax(axis=1)
    return float((guesses == ys).mean())


print(f"before training: train {round(accuracy(X, y) * 100)}% test {round(accuracy(Xt, yt) * 100)}%")



def train_step(xb, yb, lr):
    h, p = forward(xb)
    dz2 = (p - np.eye(10)[yb]) / len(xb)
    dh = dz2 @ net["W2"].T * (h > 0)
    net["W2"] -= lr * (h.T @ dz2)
    net["b2"] -= lr * dz2.sum(axis=0)
    net["W1"] -= lr * (xb.T @ dh)
    net["b1"] -= lr * dh.sum(axis=0)


def train(epochs, batch, lr):
    order = list(range(len(X)))
    for epoch in range(1, epochs + 1):
        random.shuffle(order)
        for start in range(0, len(order), batch):
            idx = order[start:start + batch]
            train_step(X[idx], y[idx], lr)
        print(f"epoch {epoch} train {round(accuracy(X, y) * 100)}%")


train(10, 20, 0.5)


# step 5: the accuracy of every digit on the test set, then the total
for d in range(10):
    mask = yt == d
    print(f"digit {d}: {round(accuracy(Xt[mask], yt[mask]) * 100)}%")
test = accuracy(Xt, yt)
print(f"test accuracy: {round(test * 100)}%")
if test > 0.9:
    print("goal reached: above 90%")


# step 6a: list every test mistake (true and predicted digit) and draw it
h, p = forward(Xt)
guesses = p.argmax(axis=1)
for i in range(len(yt)):
    if guesses[i] != yt[i]:
        print("true", int(yt[i]), "predicted", int(guesses[i]))
        show(Xt[i])

# step 6b: recognise a digit drawn by hand
drawing = ["####.",
           "...#.",
           ".###.",
           "...#.",
           "####."]


def read_drawing(rows):
    pixels = [1.0 if ch == "#" else 0.0 for ch in "".join(rows)]
    return np.array([pixels])


h, p = forward(read_drawing(drawing))
print(f"my digit: {int(p.argmax())} ({round(float(p.max()) * 100)}% sure)")

اول خودت امتحان کن؛ جواب‌های دیگر هم قبول است. برگرد و خودت حلش کن