第T6周 好莱坞明星识别

 - **🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/rnFa-IeY93EpjVu0yzzjkw) 中的学习记录博客**
- **🍖 原作者：[K同学啊](https://mtyjkh.blog.csdn.net/)**

语⾔环境：Python 3.9

开发⼯具： Jupyter Lab

深度学习环境：TensorFlow version 2.18.0

from tensorflow       import keras
from tensorflow.keras import layers,models
import os, PIL, pathlib
import matplotlib.pyplot as plt
import tensorflow        as tf
import numpy             as np

data_dir ="/Users/eleanorzhang/Downloads/48-data"
data_dir = pathlib.Path(data_dir)

image_count = len(list(data_dir.glob('*/*.jpg')))
print("图片总数为：",image_count)

图片总数为： 1800

roses = list(data_dir.glob('Jennifer Lawrence/*.jpg'))
PIL.Image.open(str(roses[0]))

batch_size = 32
img_height = 224
img_width = 224

train_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.1,
    subset="training",
    label_mode = "categorical",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

Found 1800 files belonging to 17 classes.
Using 1620 files for training.

val_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.1,
    subset="validation",
    label_mode = "categorical",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

Found 1800 files belonging to 17 classes.
Using 180 files for validation.

class_names = train_ds.class_names
print(class_names)

['Angelina Jolie', 'Brad Pitt', 'Denzel Washington', 'Hugh Jackman', 'Jennifer Lawrence', 'Johnny Depp', 'Kate Winslet', 'Leonardo DiCaprio', 'Megan Fox', 'Natalie Portman', 'Nicole Kidman', 'Robert Downey Jr', 'Sandra Bullock', 'Scarlett Johansson', 'Tom Cruise', 'Tom Hanks', 'Will Smith']

plt.figure(figsize=(20, 10))

for images, labels in train_ds.take(1):
    for i in range(20):
        ax = plt.subplot(5, 10, i + 1)

        plt.imshow(images[i].numpy().astype("uint8"))
        plt.title(class_names[np.argmax(labels[i])])
        
        plt.axis("off")

for image_batch, labels_batch in train_ds:
    print(image_batch.shape)
    print(labels_batch.shape)
    break

(32, 224, 224, 3)
(32, 17)

AUTOTUNE = tf.data.AUTOTUNE

train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)

model = models.Sequential([
    layers.Input(shape=(img_height, img_width, 3)),
    layers.Rescaling(1./255),
    
    layers.Conv2D(16, (3, 3), activation='relu'),  
    layers.AveragePooling2D((2, 2)),
    layers.Conv2D(32, (3, 3), activation='relu'),
    layers.AveragePooling2D((2, 2)),
    layers.Dropout(0.5),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.AveragePooling2D((2, 2)),
    layers.Dropout(0.5),
    layers.Conv2D(128, (3, 3), activation='relu'),
    layers.Dropout(0.5),

    layers.Flatten(),
    layers.Dense(128, activation='relu'),
    layers.Dense(len(class_names))
])
model.summary() 

Model: "sequential_2"

┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ rescaling_2 (Rescaling)         │ (None, 224, 224, 3)    │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ conv2d_8 (Conv2D)               │ (None, 222, 222, 16)   │           448 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ average_pooling2d_6             │ (None, 111, 111, 16)   │             0 │
│ (AveragePooling2D)              │                        │               │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ conv2d_9 (Conv2D)               │ (None, 109, 109, 32)   │         4,640 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ average_pooling2d_7             │ (None, 54, 54, 32)     │             0 │
│ (AveragePooling2D)              │                        │               │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dropout_6 (Dropout)             │ (None, 54, 54, 32)     │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ conv2d_10 (Conv2D)              │ (None, 52, 52, 64)     │        18,496 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ average_pooling2d_8             │ (None, 26, 26, 64)     │             0 │
│ (AveragePooling2D)              │                        │               │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dropout_7 (Dropout)             │ (None, 26, 26, 64)     │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ conv2d_11 (Conv2D)              │ (None, 24, 24, 128)    │        73,856 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dropout_8 (Dropout)             │ (None, 24, 24, 128)    │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ flatten_2 (Flatten)             │ (None, 73728)          │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_4 (Dense)                 │ (None, 128)            │     9,437,312 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_5 (Dense)                 │ (None, 17)             │         2,193 │
└─────────────────────────────────┴────────────────────────┴───────────────┘

 Total params: 9,536,945 (36.38 MB)

 Trainable params: 9,536,945 (36.38 MB)

 Non-trainable params: 0 (0.00 B)

# 设置初始学习率
initial_learning_rate = 1e-4

lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(
        initial_learning_rate, 
        decay_steps=60,      
        decay_rate=0.96,     
        staircase=True)

# 将指数衰减学习率送入优化器
optimizer = tf.keras.optimizers.Adam(learning_rate=lr_schedule)

model.compile(optimizer=optimizer,
              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

history = model.fit(train_ds,
                    validation_data=val_ds,
                    epochs=epochs,
                    callbacks=[checkpointer, earlystopper])

Epoch 1/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 253ms/step - accuracy: 0.0708 - loss: 25.7558 
Epoch 1: val_accuracy improved from 0.06667 to 0.17778, saving model to best_model.weights.h5
51/51 ━━━━━━━━━━━━━━━━━━━━ 14s 264ms/step - accuracy: 0.0712 - loss: 25.5645 - val_accuracy: 0.1778 - val_loss: 4.2205
Epoch 2/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 248ms/step - accuracy: 0.1320 - loss: 4.2774 
Epoch 2: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 255ms/step - accuracy: 0.1319 - loss: 4.2648 - val_accuracy: 0.1389 - val_loss: 2.8320
Epoch 3/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 234ms/step - accuracy: 0.1083 - loss: 2.8260 
Epoch 3: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 241ms/step - accuracy: 0.1082 - loss: 2.8261 - val_accuracy: 0.0333 - val_loss: 2.8332
Epoch 4/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 232ms/step - accuracy: 0.1161 - loss: 2.8246 
Epoch 4: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 239ms/step - accuracy: 0.1160 - loss: 2.8246 - val_accuracy: 0.1389 - val_loss: 2.8225
Epoch 5/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 230ms/step - accuracy: 0.0929 - loss: 2.8204 
Epoch 5: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 237ms/step - accuracy: 0.0932 - loss: 2.8202 - val_accuracy: 0.1389 - val_loss: 2.7813
Epoch 6/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 229ms/step - accuracy: 0.1098 - loss: 2.7953 
Epoch 6: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 236ms/step - accuracy: 0.1098 - loss: 2.7952 - val_accuracy: 0.1389 - val_loss: 2.7722
Epoch 7/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 237ms/step - accuracy: 0.1115 - loss: 2.7761 
Epoch 7: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 245ms/step - accuracy: 0.1114 - loss: 2.7762 - val_accuracy: 0.1444 - val_loss: 2.7614
Epoch 8/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 249ms/step - accuracy: 0.1031 - loss: 2.7705 
Epoch 8: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 256ms/step - accuracy: 0.1032 - loss: 2.7703 - val_accuracy: 0.1444 - val_loss: 2.7321
Epoch 9/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 232ms/step - accuracy: 0.1091 - loss: 2.7378 
Epoch 9: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 239ms/step - accuracy: 0.1092 - loss: 2.7377 - val_accuracy: 0.1500 - val_loss: 2.7120
Epoch 10/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 230ms/step - accuracy: 0.1143 - loss: 2.7163 
Epoch 10: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 237ms/step - accuracy: 0.1146 - loss: 2.7159 - val_accuracy: 0.1611 - val_loss: 2.6781
Epoch 11/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 235ms/step - accuracy: 0.1492 - loss: 2.6592 
Epoch 11: val_accuracy did not improve from 0.17778
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 242ms/step - accuracy: 0.1491 - loss: 2.6591 - val_accuracy: 0.1500 - val_loss: 2.6439
Epoch 12/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 235ms/step - accuracy: 0.1626 - loss: 2.6155 
Epoch 12: val_accuracy improved from 0.17778 to 0.18889, saving model to best_model.weights.h5
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 245ms/step - accuracy: 0.1626 - loss: 2.6154 - val_accuracy: 0.1889 - val_loss: 2.6035
Epoch 13/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 231ms/step - accuracy: 0.1643 - loss: 2.5819 
Epoch 13: val_accuracy did not improve from 0.18889
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 238ms/step - accuracy: 0.1645 - loss: 2.5813 - val_accuracy: 0.1778 - val_loss: 2.5535
Epoch 14/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 234ms/step - accuracy: 0.1871 - loss: 2.4949 
Epoch 14: val_accuracy improved from 0.18889 to 0.19444, saving model to best_model.weights.h5
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 243ms/step - accuracy: 0.1873 - loss: 2.4949 - val_accuracy: 0.1944 - val_loss: 2.5187
Epoch 15/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 232ms/step - accuracy: 0.2264 - loss: 2.4217 
Epoch 15: val_accuracy improved from 0.19444 to 0.22222, saving model to best_model.weights.h5
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 241ms/step - accuracy: 0.2265 - loss: 2.4217 - val_accuracy: 0.2222 - val_loss: 2.4918
Epoch 16/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 234ms/step - accuracy: 0.2466 - loss: 2.3612 
Epoch 16: val_accuracy did not improve from 0.22222
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 241ms/step - accuracy: 0.2464 - loss: 2.3616 - val_accuracy: 0.2056 - val_loss: 2.4734
Epoch 17/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 257ms/step - accuracy: 0.2434 - loss: 2.3478 
Epoch 17: val_accuracy did not improve from 0.22222
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 265ms/step - accuracy: 0.2433 - loss: 2.3475 - val_accuracy: 0.2167 - val_loss: 2.4563
Epoch 18/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 233ms/step - accuracy: 0.2577 - loss: 2.3213 
Epoch 18: val_accuracy did not improve from 0.22222
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 240ms/step - accuracy: 0.2579 - loss: 2.3210 - val_accuracy: 0.1944 - val_loss: 2.4675
Epoch 19/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 232ms/step - accuracy: 0.2850 - loss: 2.2352 
Epoch 19: val_accuracy did not improve from 0.22222
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 239ms/step - accuracy: 0.2847 - loss: 2.2357 - val_accuracy: 0.2056 - val_loss: 2.4397
Epoch 20/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 233ms/step - accuracy: 0.2604 - loss: 2.2457 
Epoch 20: val_accuracy improved from 0.22222 to 0.23889, saving model to best_model.weights.h5
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 241ms/step - accuracy: 0.2603 - loss: 2.2458 - val_accuracy: 0.2389 - val_loss: 2.4424
Epoch 21/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 253ms/step - accuracy: 0.2917 - loss: 2.1719 
Epoch 21: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 262ms/step - accuracy: 0.2917 - loss: 2.1724 - val_accuracy: 0.2389 - val_loss: 2.4397
Epoch 22/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 245ms/step - accuracy: 0.2938 - loss: 2.1678 
Epoch 22: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 252ms/step - accuracy: 0.2938 - loss: 2.1678 - val_accuracy: 0.2111 - val_loss: 2.4844
Epoch 23/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 237ms/step - accuracy: 0.3147 - loss: 2.0991 
Epoch 23: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 244ms/step - accuracy: 0.3146 - loss: 2.0997 - val_accuracy: 0.2056 - val_loss: 2.4475
Epoch 24/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 238ms/step - accuracy: 0.3112 - loss: 2.1336 
Epoch 24: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 246ms/step - accuracy: 0.3113 - loss: 2.1335 - val_accuracy: 0.2167 - val_loss: 2.4753
Epoch 25/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 250ms/step - accuracy: 0.3331 - loss: 2.0728 
Epoch 25: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 258ms/step - accuracy: 0.3330 - loss: 2.0731 - val_accuracy: 0.1944 - val_loss: 2.4913
Epoch 26/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 245ms/step - accuracy: 0.3269 - loss: 2.0983 
Epoch 26: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 253ms/step - accuracy: 0.3269 - loss: 2.0977 - val_accuracy: 0.2111 - val_loss: 2.4631
Epoch 27/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 248ms/step - accuracy: 0.3634 - loss: 2.0221 
Epoch 27: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 256ms/step - accuracy: 0.3632 - loss: 2.0223 - val_accuracy: 0.2111 - val_loss: 2.4825
Epoch 28/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 256ms/step - accuracy: 0.3508 - loss: 2.0024 
Epoch 28: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 263ms/step - accuracy: 0.3509 - loss: 2.0025 - val_accuracy: 0.2000 - val_loss: 2.4996
Epoch 29/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 234ms/step - accuracy: 0.3658 - loss: 1.9887 
Epoch 29: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 241ms/step - accuracy: 0.3657 - loss: 1.9887 - val_accuracy: 0.1833 - val_loss: 2.5364
Epoch 30/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 242ms/step - accuracy: 0.3817 - loss: 1.9563 
Epoch 30: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 250ms/step - accuracy: 0.3815 - loss: 1.9562 - val_accuracy: 0.2167 - val_loss: 2.5035
Epoch 31/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 242ms/step - accuracy: 0.3834 - loss: 1.8738 
Epoch 31: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 249ms/step - accuracy: 0.3832 - loss: 1.8748 - val_accuracy: 0.2222 - val_loss: 2.5464
Epoch 32/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 237ms/step - accuracy: 0.3772 - loss: 1.8888 
Epoch 32: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 12s 245ms/step - accuracy: 0.3774 - loss: 1.8889 - val_accuracy: 0.2000 - val_loss: 2.6030
Epoch 33/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 239ms/step - accuracy: 0.3808 - loss: 1.9160 
Epoch 33: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 246ms/step - accuracy: 0.3813 - loss: 1.9153 - val_accuracy: 0.2056 - val_loss: 2.5533
Epoch 34/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 239ms/step - accuracy: 0.4032 - loss: 1.8132 
Epoch 34: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 246ms/step - accuracy: 0.4032 - loss: 1.8135 - val_accuracy: 0.2000 - val_loss: 2.5865
Epoch 35/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 245ms/step - accuracy: 0.4223 - loss: 1.8315 
Epoch 35: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 252ms/step - accuracy: 0.4222 - loss: 1.8315 - val_accuracy: 0.2111 - val_loss: 2.5424
Epoch 36/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 249ms/step - accuracy: 0.4341 - loss: 1.8065 
Epoch 36: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 256ms/step - accuracy: 0.4342 - loss: 1.8060 - val_accuracy: 0.1667 - val_loss: 2.6543
Epoch 37/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 239ms/step - accuracy: 0.4138 - loss: 1.7475 
Epoch 37: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 247ms/step - accuracy: 0.4140 - loss: 1.7479 - val_accuracy: 0.2222 - val_loss: 2.6179
Epoch 38/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 238ms/step - accuracy: 0.4498 - loss: 1.7417 
Epoch 38: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 246ms/step - accuracy: 0.4495 - loss: 1.7420 - val_accuracy: 0.2278 - val_loss: 2.6206
Epoch 39/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 253ms/step - accuracy: 0.4541 - loss: 1.6876 
Epoch 39: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 261ms/step - accuracy: 0.4540 - loss: 1.6879 - val_accuracy: 0.2222 - val_loss: 2.6067
Epoch 40/100
51/51 ━━━━━━━━━━━━━━━━━━━━ 0s 240ms/step - accuracy: 0.4312 - loss: 1.7638 
Epoch 40: val_accuracy did not improve from 0.23889
51/51 ━━━━━━━━━━━━━━━━━━━━ 13s 247ms/step - accuracy: 0.4316 - loss: 1.7625 - val_accuracy: 0.2222 - val_loss: 2.6200
Epoch 40: early stopping

model.save_weights('best_model.weights.h5')

from datetime import datetime
current_time = datetime.now() # 获取当前时间

acc = history.history['accuracy']
val_acc = history.history['val_accuracy']

loss = history.history['loss']
val_loss = history.history['val_loss']

epochs_range = range(len(loss))

plt.figure(figsize=(12, 4))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, acc, label='Training Accuracy')
plt.plot(epochs_range, val_acc, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.xlabel(current_time) # 打卡请带上时间戳，否则代码截图无效

plt.subplot(1, 2, 2)
plt.plot(epochs_range, loss, label='Training Loss')
plt.plot(epochs_range, val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

from PIL import Image
import numpy as np

import os
path = os.path.join("/Users/eleanorzhang/Downloads/48-data", "Tom Cruise", "031_df3204d3.jpg")

 #这里选择你需要预测的图片
image = tf.image.resize(img, [img_height, img_width])

img_array = tf.expand_dims(image, 0) 

# 加载效果最好的模型权重
model.load_weights('best_model.weights.h5')
predictions = model.predict(img_array) # 这里选用你已经训练好的模型
print("预测结果为：",class_names[np.argmax(predictions)])

1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 123ms/step
预测结果为： Tom Cruise