0
0

【人工智能】Python在机器学习与人工智能中的应用

Python因其简洁易用、丰富的库支持以及强大的社区，被广泛应用于机器学习与人工智能（AI）领域。本教程通过实用的代码示例和讲解，带你从零开始掌握Python在机器学习与人工智能中的基本用法。

1. 机器学习与AI的Python生态系统

Python拥有多种支持机器学习和AI的库，以下是几个核心库：

  • NumPy：处理高效数组和矩阵运算。
  • Pandas：提供数据操作与分析工具。
  • Matplotlib/Seaborn：用于数据可视化。
  • Scikit-learn：机器学习的核心库，包含分类、回归、聚类等算法。
  • TensorFlow/PyTorch：深度学习框架，用于构建和训练神经网络。

安装：

  1. pip install numpy pandas matplotlib scikit-learn tensorflow

2. 数据预处理

加载数据
  1. import pandas as pd
  3. # 示例数据
  4. data = pd.DataFrame({
  5.     'Feature1': [1, 2, 3, 4, 5],
  6.     'Feature2': [5, 4, 3, 2, 1],
  7.     'Target': [1, 0, 1, 0, 1]
  8. })
  10. print(data)

输出：

  1.    Feature1  Feature2  Target
  2. 0         1         5       1
  3. 1         2         4       0
  4. 2         3         3       1
  5. 3         4         2       0
  6. 4         5         1       1
特征缩放

归一化或标准化数据有助于提升模型性能。

  1. import pandas as pd
  2. from sklearn.preprocessing import MinMaxScaler
  4. data = pd.DataFrame({
  5.     'Feature1': [1, 2, 3, 4, 5],
  6.     'Feature2': [5, 4, 3, 2, 1],
  7.     'Target': [1, 0, 1, 0, 1]
  8. })
  10. scaler = MinMaxScaler()
  11. scaled_features = scaler.fit_transform(data[['Feature1', 'Feature2']])
  12. print(scaled_features)

输出：

  1. [[0.   1.  ]
  2.  [0.25 0.75]
  3.  [0.5  0.5 ]
  4.  [0.75 0.25]
  5.  [1.   0.  ]]

3. 数据可视化

利用

  1. Matplotlib

和

  1. Seaborn

绘制数据分布图。

  1. import pandas as pd
  2. from sklearn.preprocessing import MinMaxScaler
  3. import matplotlib.pyplot as plt
  4. import seaborn as sns
  6. data = pd.DataFrame({
  7.     'Feature1': [1, 2, 3, 4, 5],
  8.     'Feature2': [5, 4, 3, 2, 1],
  9.     'Target': [1, 0, 1, 0, 1]
  10. })
  12. scaler = MinMaxScaler()
  13. scaled_features = scaler.fit_transform(data[['Feature1', 'Feature2']])
  14. print(scaled_features)
  16. # 散点图
  17. sns.scatterplot(x='Feature1', y='Feature2', hue='Target', data=data)
  18. plt.title('Feature Scatter Plot')
  19. plt.show()

4. 构建第一个机器学习模型

使用

  1. Scikit-learn

实现分类模型。

拆分数据
  1. import pandas as pd
  2. from sklearn.model_selection import train_test_split
  4. data = pd.DataFrame({
  5.     'Feature1': [1, 2, 3, 4, 5],
  6.     'Feature2': [5, 4, 3, 2, 1],
  7.     'Target': [1, 0, 1, 0, 1]
  8. })
  10. X = data[['Feature1', 'Feature2']]
  11. y = data['Target']
  13. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
  15. print('X_train:')
  16. print(X_train)
  17. print('X_test:')
  18. print(X_test)
  19. print('y_train:')
  20. print(y_train)
  21. print('y_test:')
  22. print(y_test)
  1. X_train:
  2.    Feature1  Feature2
  3. 4         5         1
  4. 2         3         3
  5. 0         1         5
  6. 3         4         2
  8. X_test:
  9.    Feature1  Feature2
  10. 1         2         4
  12. y_train:
  13. 4    1
  14. 2    1
  15. 0    1
  16. 3    0
  17. Name: Target, dtype: int64
  19. y_test:
  20. 1    0
  21. Name: Target, dtype: int64
训练模型
  1. import pandas as pd
  2. from sklearn.model_selection import train_test_split
  3. from sklearn.ensemble import RandomForestClassifier
  4. from sklearn.metrics import accuracy_score
  6. data = pd.DataFrame({
  7.     'Feature1': [1, 2, 3, 4, 5],
  8.     'Feature2': [5, 4, 3, 2, 1],
  9.     'Target': [1, 0, 1, 0, 1]
  10. })
  12. X = data[['Feature1', 'Feature2']]
  13. y = data['Target']
  15. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
  17. # 随机森林分类器
  18. model = RandomForestClassifier()
  19. model.fit(X_train, y_train)
  21. # 预测
  22. y_pred = model.predict(X_test)
  23. print("Accuracy:", accuracy_score(y_test, y_pred))
  1. Accuracy: 0.0

5. 深度学习与神经网络

构建一个简单的神经网络进行分类任务。

安装TensorFlow
  1. conda install tensorflow

如果安装遇到Could not solve for environment spec错误，请先执行以下命令

  1. conda create -n tf_env python=3.8
  2. conda activate tf_env
构建模型
  1. import tensorflow as tf
  2. from tensorflow.keras.models import Sequential
  3. from tensorflow.keras.layers import Dense
  5. # 构建神经网络
  6. model = Sequential([
  7.     Dense(8, input_dim=2, activation='relu'),
  8.     Dense(4, activation='relu'),
  9.     Dense(1, activation='sigmoid')
  10. ])
编译与训练
  1. model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
  2. model.fit(X_train, y_train, epochs=50, batch_size=1, verbose=1)
评估模型
  1. loss, accuracy = model.evaluate(X_test, y_test)
  2. print("Loss:", loss)
  3. print("Accuracy:", accuracy)
完整代码
  1. import pandas as pd
  2. from sklearn.model_selection import train_test_split
  3. import tensorflow as tf
  4. from tensorflow.keras.models import Sequential
  5. from tensorflow.keras.layers import Dense
  7. data = pd.DataFrame({
  8.     'Feature1': [1, 2, 3, 4, 5],
  9.     'Feature2': [5, 4, 3, 2, 1],
  10.     'Target': [1, 0, 1, 0, 1]
  11. })
  13. X = data[['Feature1', 'Feature2']]
  14. y = data['Target']
  16. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
  18. # 构建神经网络
  19. model = Sequential([
  20.     Dense(8, input_dim=2, activation='relu'),
  21.     Dense(4, activation='relu'),
  22.     Dense(1, activation='sigmoid')
  23. ])
  25. model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
  26. model.fit(X_train, y_train, epochs=50, batch_size=1, verbose=1)
  28. loss, accuracy = model.evaluate(X_test, y_test)
  29. print("Loss:", loss)
  30. print("Accuracy:", accuracy)

输出：

  1. Epoch 1/50
  2. 4/4 [==============================] - 1s 1ms/step - loss: 0.6867 - accuracy: 0.5000
  3. Epoch 2/50
  4. 4/4 [==============================] - 0s 997us/step - loss: 0.6493 - accuracy: 0.5000
  5. Epoch 3/50
  6. 4/4 [==============================] - 0s 997us/step - loss: 0.6183 - accuracy: 0.5000
  7. Epoch 4/50
  8. 4/4 [==============================] - 0s 665us/step - loss: 0.5920 - accuracy: 0.5000
  9. Epoch 5/50
  10. 4/4 [==============================] - 0s 1ms/step - loss: 0.5702 - accuracy: 0.5000
  11. Epoch 6/50
  12. 4/4 [==============================] - 0s 997us/step - loss: 0.5612 - accuracy: 0.7500
  13. Epoch 7/50
  14. 4/4 [==============================] - 0s 998us/step - loss: 0.5405 - accuracy: 0.7500
  15. Epoch 8/50
  16. 4/4 [==============================] - 0s 665us/step - loss: 0.5223 - accuracy: 0.7500
  17. Epoch 9/50
  18. 4/4 [==============================] - 0s 1ms/step - loss: 0.5047 - accuracy: 0.7500
  19. Epoch 10/50
  20. 4/4 [==============================] - 0s 665us/step - loss: 0.4971 - accuracy: 0.7500
  21. Epoch 11/50
  22. 4/4 [==============================] - 0s 997us/step - loss: 0.4846 - accuracy: 0.7500
  23. Epoch 12/50
  24. 4/4 [==============================] - 0s 997us/step - loss: 0.4762 - accuracy: 0.7500
  25. Epoch 13/50
  26. 4/4 [==============================] - 0s 665us/step - loss: 0.4753 - accuracy: 0.7500
  27. Epoch 14/50
  28. 4/4 [==============================] - 0s 997us/step - loss: 0.4623 - accuracy: 1.0000
  29. Epoch 15/50
  30. 4/4 [==============================] - 0s 998us/step - loss: 0.4563 - accuracy: 1.0000
  31. Epoch 16/50
  32. 4/4 [==============================] - 0s 998us/step - loss: 0.4530 - accuracy: 1.0000
  33. Epoch 17/50
  34. 4/4 [==============================] - 0s 997us/step - loss: 0.4469 - accuracy: 1.0000
  35. Epoch 18/50
  36. 4/4 [==============================] - 0s 997us/step - loss: 0.4446 - accuracy: 0.7500
  37. Epoch 19/50
  38. 4/4 [==============================] - 0s 665us/step - loss: 0.4385 - accuracy: 0.7500
  39. Epoch 20/50
  40. 4/4 [==============================] - 0s 998us/step - loss: 0.4355 - accuracy: 0.7500
  41. Epoch 21/50
  42. 4/4 [==============================] - 0s 997us/step - loss: 0.4349 - accuracy: 0.7500
  43. Epoch 22/50
  44. 4/4 [==============================] - 0s 665us/step - loss: 0.4290 - accuracy: 0.7500
  45. Epoch 23/50
  46. 4/4 [==============================] - 0s 997us/step - loss: 0.4270 - accuracy: 0.7500
  47. Epoch 24/50
  48. 4/4 [==============================] - 0s 997us/step - loss: 0.4250 - accuracy: 0.7500
  49. Epoch 25/50
  50. 4/4 [==============================] - 0s 665us/step - loss: 0.4218 - accuracy: 0.7500
  51. Epoch 26/50
  52. 4/4 [==============================] - 0s 997us/step - loss: 0.4192 - accuracy: 0.7500
  53. Epoch 27/50
  54. 4/4 [==============================] - 0s 997us/step - loss: 0.4184 - accuracy: 0.7500
  55. Epoch 28/50
  56. 4/4 [==============================] - 0s 665us/step - loss: 0.4152 - accuracy: 0.7500
  57. Epoch 29/50
  58. 4/4 [==============================] - 0s 997us/step - loss: 0.4129 - accuracy: 0.7500
  59. Epoch 30/50
  60. 4/4 [==============================] - 0s 997us/step - loss: 0.4111 - accuracy: 0.7500
  61. Epoch 31/50
  62. 4/4 [==============================] - 0s 997us/step - loss: 0.4095 - accuracy: 0.7500
  63. Epoch 32/50
  64. 4/4 [==============================] - 0s 997us/step - loss: 0.4070 - accuracy: 0.7500
  65. Epoch 33/50
  66. 4/4 [==============================] - 0s 997us/step - loss: 0.4053 - accuracy: 0.7500
  67. Epoch 34/50
  68. 4/4 [==============================] - 0s 997us/step - loss: 0.4033 - accuracy: 0.7500
  69. Epoch 35/50
  70. 4/4 [==============================] - 0s 998us/step - loss: 0.4028 - accuracy: 0.7500
  71. Epoch 36/50
  72. 4/4 [==============================] - 0s 997us/step - loss: 0.3998 - accuracy: 0.7500
  73. Epoch 37/50
  74. 4/4 [==============================] - 0s 1ms/step - loss: 0.3978 - accuracy: 0.7500
  75. Epoch 38/50
  76. 4/4 [==============================] - 0s 997us/step - loss: 0.3966 - accuracy: 0.7500
  77. Epoch 39/50
  78. 4/4 [==============================] - 0s 665us/step - loss: 0.3946 - accuracy: 0.7500
  79. Epoch 40/50
  80. 4/4 [==============================] - 0s 997us/step - loss: 0.3926 - accuracy: 0.7500
  81. Epoch 41/50
  82. 4/4 [==============================] - 0s 997us/step - loss: 0.3918 - accuracy: 0.7500
  83. Epoch 42/50
  84. 4/4 [==============================] - 0s 997us/step - loss: 0.3898 - accuracy: 0.7500
  85. Epoch 43/50
  86. 4/4 [==============================] - 0s 997us/step - loss: 0.3877 - accuracy: 0.7500
  87. Epoch 44/50
  88. 4/4 [==============================] - 0s 997us/step - loss: 0.3861 - accuracy: 0.7500
  89. Epoch 45/50
  90. 4/4 [==============================] - 0s 665us/step - loss: 0.3842 - accuracy: 0.7500
  91. Epoch 46/50
  92. 4/4 [==============================] - 0s 665us/step - loss: 0.3830 - accuracy: 0.7500
  93. Epoch 47/50
  94. 4/4 [==============================] - 0s 997us/step - loss: 0.3815 - accuracy: 0.7500
  95. Epoch 48/50
  96. 4/4 [==============================] - 0s 665us/step - loss: 0.3790 - accuracy: 0.7500
  97. Epoch 49/50
  98. 4/4 [==============================] - 0s 665us/step - loss: 0.3778 - accuracy: 0.7500
  99. Epoch 50/50
  100. 4/4 [==============================] - 0s 997us/step - loss: 0.3768 - accuracy: 0.7500
  101. 1/1 [==============================] - 0s 277ms/step - loss: 2.8638 - accuracy: 0.0000e+00
  102. Loss: 2.863826274871826
  103. Accuracy: 0.0

6. 数据聚类

实现一个K-Means聚类模型：

  1. from sklearn.cluster import KMeans
  3. # 数据
  4. data_points = [[1, 2], [2, 3], [3, 4], [8, 7], [9, 8], [10, 9]]
  6. # K-Means
  7. kmeans = KMeans(n_clusters=2)
  8. kmeans.fit(data_points)
  10. # 输出聚类中心
  11. print("Cluster Centers:", kmeans.cluster_centers_)

输出：

  1. Cluster Centers: [[9. 8.]
  2.  [2. 3.]]

7. 自然语言处理 (NLP)

使用

  1. NLTK

处理文本数据：

  1. pip install nltk
文本分词
  1. import nltk
  3. nltk.download('punkt_tab')
  4. nltk.download('punkt')
  6. from nltk.tokenize import word_tokenize
  8. text = "Machine learning is amazing!"
  9. tokens = word_tokenize(text)
  10. print(tokens)

输出： 

  1. ['Machine', 'learning', 'is', 'amazing', '!']
词袋模型
  1. from sklearn.feature_extraction.text import CountVectorizer
  3. texts = ["I love Python", "Python is great for AI"]
  4. vectorizer = CountVectorizer()
  5. X = vectorizer.fit_transform(texts)
  7. print(X.toarray())

输出： 

  1. [[0 0 0 0 1 1]
  2.  [1 1 1 1 0 1]]

8. 实用案例：房价预测

  1. from sklearn.datasets import fetch_california_housing
  2. from sklearn.model_selection import train_test_split
  3. from sklearn.linear_model import LinearRegression
  4. from sklearn.metrics import mean_squared_error
  6. # 加载数据集
  7. data = fetch_california_housing(as_frame=True)
  8. X = data.data
  9. y = data.target
  11. # 数据拆分
  12. X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
  14. # 模型训练
  15. model = LinearRegression()
  16. model.fit(X_train, y_train)
  18. # 预测
  19. y_pred = model.predict(X_test)
  20. print("Model Coefficients:", model.coef_)
  22. # 评估
  23. mse = mean_squared_error(y_test, y_pred)
  24. print(f"Mean Squared Error: {mse}")

输出： 

  1. Model Coefficients: [ 4.48674910e-01  9.72425752e-03 -1.23323343e-01  7.83144907e-01
  2.  -2.02962058e-06 -3.52631849e-03 -4.19792487e-01 -4.33708065e-01]
  3. Mean Squared Error: 0.5558915986952442

总结

本教程涵盖了Python在机器学习和人工智能领域的基础应用，从数据预处理、可视化到模型构建和评估，再到深度学习的基本实现。通过这些示例，你可以逐步掌握如何使用Python进行机器学习和AI项目开发。

标签： 人工智能 python 开发语言
本文转载自: https://blog.csdn.net/IT_ORACLE/article/details/143933949
版权归原作者 IT古董 所有， 如有侵权，请联系我们删除。
登录后发布评论

“【人工智能】Python在机器学习与人工智能中的应用”的评论:

还没有评论
关于作者
...
overfit同步小助手
文章同步
相关阅读

网络安全法-网络运行安全

使用selenium/drissionpage时如何阻止chrome自动跳转http到https

docker desktop 里部署的Open WebUI 管理员密码忘记了的处理方法

在ubuntu20.04中搭建onsite比赛运行环境

利用开源的低代码表单设计器FcDesigner高效管理和渲染复杂表单结构

Kafka学习笔记

【前端】浏览器输入url到页面呈现发生了什么？

文章导航
没有找到文章内容分段。。。