from sklearn.datasets import load_digits
from sklearn.svm import SVC
from sklearn.model_selection import GridSearchCV

# 加載數(shù)據(jù)集
digits = load_digits()
X, y = digits.data, digits.target

# 設(shè)置參數(shù)網(wǎng)格
param_grid = {'C': [0.1, 1, 10], 'gamma': [0.01, 0.001]}

# 創(chuàng)建 SVM 分類器
grid = GridSearchCV(SVC(), param_grid, cv=5)
grid.fit(X, y)

# 輸出最佳參數(shù)和測(cè)試集準(zhǔn)確率
print(f"最佳參數(shù): {grid.best_params_}")
print(f"測(cè)試集準(zhǔn)確率: {grid.best_score_:.2%}")

from sklearn.datasets import fetch_california_housing
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.inspection import permutation_importance
import matplotlib.pyplot as plt

# 加載數(shù)據(jù)集
housing = fetch_california_housing()
X, y = housing.data, housing.target

# 創(chuàng)建梯度提升回歸器
model = GradientBoostingRegressor(n_estimators=200)
model.fit(X, y)

# 特征重要性分析
result = permutation_importance(model, X, y, n_repeats=10)
sorted_idx = result.importances_mean.argsort()

# 可視化特征重要性
plt.barh(housing.feature_names[sorted_idx], result.importances_mean[sorted_idx])
plt.xlabel("Permutation Importance")
plt.show()

import torch
import torchvision
from torch import nn, optim

# 定義 CNN 模型
class CNN(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
        self.pool = nn.MaxPool2d(2, 2)
        self.fc1 = nn.Linear(16*112*112, 256)
        self.fc2 = nn.Linear(256, 10)

    def forward(self, x):
        x = self.pool(torch.relu(self.conv1(x)))
        x = x.view(-1, 16*112*112)
        x = torch.relu(self.fc1(x))
        return self.fc2(x)

# 數(shù)據(jù)加載
transform = torchvision.transforms.Compose([
    torchvision.transforms.Resize((224, 224)),
    torchvision.transforms.ToTensor()
])
dataset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform)

# 模型訓(xùn)練
model = CNN()
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)

for epoch in range(10):
    for i, (inputs, labels) in enumerate(dataset):
        optimizer.zero_grad()
        outputs = model(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()
    print(f"Epoch {epoch+1}, Loss: {loss.item()}")

from transformers import pipeline

# 加載情感分析器
classifier = pipeline("sentiment-analysis")

# 輸入文本
texts = [
    "I'm excited to learn AI with Python!",
    "This implementation is terribly confusing."
]

# 進(jìn)行情感分析
results = classifier(texts)

# 輸出結(jié)果
for result in results:
    print(f"文本: {result['label']}，置信度: {result['score']:.2f}")

from transformers import GPT2LMHeadModel, GPT2Tokenizer

# 加載預(yù)訓(xùn)練模型和分詞器
tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
model = GPT2LMHeadModel.from_pretrained("gpt2")

# 輸入文本
input_text = "Artificial intelligence is"
input_ids = tokenizer.encode(input_text, return_tensors='pt')

# 生成文本
output = model.generate(
    input_ids,
    max_length=100,
    num_return_sequences=3,
    temperature=0.7
)

# 輸出結(jié)果
for i, sample in enumerate(output):
    print(f"生成文本 {i+1}: {tokenizer.decode(sample)}")

import cv2
import dlib
from sklearn.svm import SVC
import numpy as np

# 人臉檢測(cè)
face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
landmark_predictor = dlib.shape_predictor('shape_predictor_68_face_landmarks.dat')

# 特征提取
def extract_features(image):
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    faces = face_detector.detectMultiScale(gray, 1.3, 5)
    for (x, y, w, h) in faces:
        face = gray[y:y+h, x:x+w]
        landmarks = landmark_predictor(face, dlib.rectangle(0, 0, w, h))
        features = np.array([landmarks.part(i).x for i in range(68)] + [landmarks.part(i).y for i in range(68)])
        return features
    return None

# 數(shù)據(jù)采集
images = [...]  # 假設(shè)已有圖像列表
labels = [...]  # 假設(shè)已有標(biāo)簽列表

features = [extract_features(img) for img in images]
features = np.array([f for f in features if f is not None])
labels = np.array([labels[i] for i in range(len(labels)) if features[i] is not None])

# 模型訓(xùn)練
model = SVC(kernel='linear')
model.fit(features, labels)

# 人臉識(shí)別
def recognize_face(image):
    features = extract_features(image)
    if features is not None:
        prediction = model.predict([features])
        return prediction[0]
    return None

# 測(cè)試
test_image = cv2.imread('test.jpg')
result = recognize_face(test_image)
print(f'Recognized person: {result}')