第T3周：天气识别

于 2024-10-18 22:47:27 发布
阅读量756 收藏 29
点赞数 20
文章标签： python 机器学习
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.youkuaiyun.com/2405_87523719/article/details/143063765
版权

🚀我的环境：

  • 语言环境：python 3.12.6
  • 编译器：jupyter lab
  • 深度学习环境：TensorFlow 2.17.0

前期工作

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

data_dir = r"d:\Users\yxy\Desktop\weather_photos"


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： 1125

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

在这里插入图片描述

数据预处理

import tensorflow as tf

batch_size = 32
img_height = 180
img_width = 180

import os

if os.path.exists(data_dir):
    print("Path exists!")
else:
    print("Path does not exist.")


Path exists!

train_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.2,
    subset="training",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

Found 1125 files belonging to 4 classes.
Using 900 files for training.

val_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.2,
    subset="validation",
    seed=123,
    image_size=(img_height, img_width),
    batch_size=batch_size)

Found 1125 files belonging to 4 classes.
Using 225 files for validation.

class_names = train_ds.class_names
print(class_names)

['cloudy', 'rain', 'shine', 'sunrise']

训练集包含900张图片，验证集包含225张图片，所有图片被均匀地划分为4类

可视化数据

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[labels[i]])
        
        plt.axis("off")

在这里插入图片描述

再次检查数据

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

(32, 180, 180, 3)
(32,)

配置数据集

AUTOTUNE = tf.data.AUTOTUNE

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

构建CNN网络

num_classes = 4

model = models.Sequential([
    layers.Rescaling(1./255, input_shape=(img_height, img_width, 3)),
    
    layers.Conv2D(16, (3, 3), activation='relu', input_shape=(img_height, img_width, 3)), # 卷积层1，卷积核3*3  
    layers.AveragePooling2D((2, 2)),               # 池化层1，2*2采样
    layers.Conv2D(32, (3, 3), activation='relu'),  # 卷积层2，卷积核3*3
    layers.AveragePooling2D((2, 2)),               # 池化层2，2*2采样
    layers.Conv2D(64, (3, 3), activation='relu'),  # 卷积层3，卷积核3*3
    layers.Dropout(0.3),                           # 让神经元以一定的概率停止工作，防止过拟合，提高模型的泛化能力。
    
    layers.Flatten(),                       # Flatten层，连接卷积层与全连接层
    layers.Dense(128, activation='relu'),   # 全连接层，特征进一步提取
    layers.Dense(num_classes)               # 输出层，输出预期结果
])

model.summary()  # 打印网络结构

C:\Users\yxy\AppData\Local\Programs\Python\Python312\Lib\site-packages\keras\src\layers\preprocessing\tf_data_layer.py:19: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.
  super().__init__(**kwargs)
C:\Users\yxy\AppData\Local\Programs\Python\Python312\Lib\site-packages\keras\src\layers\convolutional\base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.
  super().__init__(activity_regularizer=activity_regularizer, **kwargs)

Model: "sequential"

┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓
┃ Layer (type)                         ┃ Output Shape                ┃         Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩
│ rescaling (Rescaling)                │ (None, 180, 180, 3)         │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ conv2d (Conv2D)                      │ (None, 178, 178, 16)        │             448 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ average_pooling2d (AveragePooling2D) │ (None, 89, 89, 16)          │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ conv2d_1 (Conv2D)                    │ (None, 87, 87, 32)          │           4,640 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ average_pooling2d_1                  │ (None, 43, 43, 32)          │               0 │
│ (AveragePooling2D)                   │                             │                 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ conv2d_2 (Conv2D)                    │ (None, 41, 41, 64)          │          18,496 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ dropout (Dropout)                    │ (None, 41, 41, 64)          │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ flatten (Flatten)                    │ (None, 107584)              │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ dense (Dense)                        │ (None, 128)                 │      13,770,880 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ dense_1 (Dense)                      │ (None, 4)                   │             516 │
└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘

 Total params: 13,794,980 (52.62 MB)

 Trainable params: 13,794,980 (52.62 MB)

 Non-trainable params: 0 (0.00 B)

编译

# 设置优化器
opt = tf.keras.optimizers.Adam(learning_rate=0.001)

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

训练模型

epochs = 10

history = model.fit(
  train_ds,
  validation_data=val_ds,
  epochs=epochs
)

Epoch 1/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m9s[0m 232ms/step - accuracy: 0.4405 - loss: 2.1117 - val_accuracy: 0.7644 - val_loss: 0.6185
Epoch 2/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 248ms/step - accuracy: 0.8258 - loss: 0.5146 - val_accuracy: 0.8089 - val_loss: 0.4748
Epoch 3/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m6s[0m 227ms/step - accuracy: 0.8760 - loss: 0.3491 - val_accuracy: 0.8267 - val_loss: 0.4637
Epoch 4/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m6s[0m 224ms/step - accuracy: 0.8636 - loss: 0.3294 - val_accuracy: 0.6533 - val_loss: 0.8600
Epoch 5/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 228ms/step - accuracy: 0.8303 - loss: 0.4439 - val_accuracy: 0.7867 - val_loss: 0.5367
Epoch 6/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m6s[0m 222ms/step - accuracy: 0.9181 - loss: 0.2308 - val_accuracy: 0.8356 - val_loss: 0.4584
Epoch 7/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 244ms/step - accuracy: 0.9236 - loss: 0.1989 - val_accuracy: 0.7956 - val_loss: 0.5130
Epoch 8/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 244ms/step - accuracy: 0.9231 - loss: 0.1858 - val_accuracy: 0.8489 - val_loss: 0.4298
Epoch 9/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 254ms/step - accuracy: 0.9721 - loss: 0.0921 - val_accuracy: 0.8578 - val_loss: 0.4897
Epoch 10/10
[1m29/29[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m7s[0m 244ms/step - accuracy: 0.9684 - loss: 0.0831 - val_accuracy: 0.8533 - val_loss: 0.4566

模型评估

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

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

epochs_range = range(epochs)

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.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()

在这里插入图片描述
从图中可以看出，模型在训练过程中，训练集的准确率和验证集的准确率都在逐渐提升，并且验证集的损失值逐渐收敛，说明模型的表现较为稳健。

总结：
本周我使用TensorFlow构建卷积神经网络（CNN），对天气图片进行分类。本周的过程包括数据集的加载、预处理、模型构建、训练以及评估。通过数据预处理与模型调优，我达到较高的训练准确率和验证准确率。

评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值