Fashion mnist 10 toifadagi 70 000 ta kul rangdagi kiyim turlarini o'z ichiga olgan ma'lumotlar to'plami. Rasmlar kichik oʻlchamdagi (28x28 piksel) kiyimlarni koʻrsatadi

) va rasmlarni tekshirish 
(
test_images
) , tekshirish javoblari (
test_labels
) yuklanadi (
load_data
) .
clothe_names
=
[
'Futbolka'
,
'Shim'
,
'Sviter'
,
'Qizlar 
Koylagi'
,
'Palto'
,
'Sandalet'
,
'Bolalar 
Koylagi'
,
'Krossovka'
,
'Sumka'
,
'Etik'
]
Malumotlarni suratlarga mos tarzda kiritamiz. Kiritgan malumotlarimiz rasmga mosligi 
tekshiriladi va toifalarga ajratiladi.
a
=
np.random.randint(
1
,
10000
)
plt.imshow(train_images[a])
plt.xlabel(clothe_names[train_labels[a]])
Text(0.5, 0, 'Etik')

---

Tarmoq qurildi . Malumotlar to'g'ri kiritilganini va o'rgatishga tayyor ekanligini tekshirib 
ko'ramiz. Buning uchun o'rgatishi to'plami (
traning_set
)dan 1 ta rasmni chiqaramiz va rasm 
ostidagi toifaga mosligini tekshiramiz. Rasmni chiqarish Random funksiyasi orqali amalga 
oshiriladi.
train_images
=
train_images
/
255.0
test_images
=
test_images
/
255.0
plt.figure(figsize
=
(
15
,
10
))
for
i 
in
range(
50
):
plt.subplot(
5
,
10
,i
+
1
)
plt.xticks([])
plt.yticks([])
plt.grid(
False
)
plt.imshow(train_images[i],cmap
=
plt.cm.binary)
plt.xlabel(clothe_names[train_labels[i]])
plt.show()

---

Keyingi qadamda o'rgatish to'plami (
traning_set
)dagi 50 ta rasmni va rasmlarga mos 
toifalarni ko'rsatishni amalga oshiramiz.
model
=
tf.keras.Sequential([
tf.keras.layers.Flatten(input_shape
=
(
28
,
28
)),
tf.keras.layers.Dense(
128
,activation
=
'relu'
),
tf.keras.layers.Dense(
10
)
])
model.compile(
optimizer
=
'adam'
,
loss
=
tf.keras.losses.SparseCategoricalCrossentropy(from_logits
=
True
),
metrics
=
[
'accuracy'
]
)
model.fit(train_images, train_labels, epochs
=
10
)
Epoch 1/10
1875/1875 [==============================] - 5s 2ms/step - loss: 
1.0826 - accuracy: 0.6629
Epoch 2/10
1875/1875 [==============================] - 5s 2ms/step - loss: 
0.6391 - accuracy: 0.7702
Epoch 3/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.5656 - accuracy: 0.7976

---

Epoch 4/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.5240 - accuracy: 0.8146
Epoch 5/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4968 - accuracy: 0.8247
Epoch 6/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4776 - accuracy: 0.8320
Epoch 7/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4632 - accuracy: 0.8362
Epoch 8/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4523 - accuracy: 0.8408
Epoch 9/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4425 - accuracy: 0.8442
Epoch 10/10
1875/1875 [==============================] - 4s 2ms/step - loss: 
0.4345 - accuracy: 0.8476