Ml Model Training

作者 secondsky88378361314fMIT227 個星標收錄於 2026年10月8日更新於 2026年10月8日儲存庫10 天前更新

Train ML models with scikit-learn, PyTorch, TensorFlow. Use for classification/regression, neural networks, hyperparameter tuning, or encountering overfitting, underfitting, convergence issues.

僅含說明Data & AnalyticsSoftware Development
AI 產生的概覽

指導使用 scikit-learn、PyTorch 和 TensorFlow 訓練機器學習模型，涵蓋資料準備、評估與常見問題。

功能
提供訓練機器學習模型的分步流程：資料準備、特徵工程、模型選擇、訓練與評估。內容包含 scikit-learn、PyTorch 和 TensorFlow/Keras 的程式碼範例、評估指標表、最佳實務，以及資料外洩、類別不平衡、過度擬合和可重現性等常見問題的解決方式。也指向兩份參考檔案，提供更完整的 PyTorch 與 TensorFlow/Keras 訓練範例。
適用情境
適用於建立分類或迴歸模型、神經網路，或進行超參數調校時。也用於診斷過度擬合、擬合不足或收斂問題等訓練故障。
執行需求
僅為說明文件，未附帶指令碼。範例假定使用 Python 及 pandas、scikit-learn、PyTorch、TensorFlow/Keras，可選用 imblearn 和 MLflow。

ML Model Training

Train machine learning models with proper data handling and evaluation.

Training Workflow

  1. Data Preparation → 2. Feature Engineering → 3. Model Selection → 4. Training → 5. Evaluation

Data Preparation

python
import pandas as pdfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScaler, LabelEncoder
# Load and clean datadf = pd.read_csv('data.csv')df = df.dropna()
# Encode categorical variablesle = LabelEncoder()df['category'] = le.fit_transform(df['category'])
# Split data (70/15/15)X = df.drop('target', axis=1)y = df['target']X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3)X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5)
# Scale featuresscaler = StandardScaler()X_train = scaler.fit_transform(X_train)X_val = scaler.transform(X_val)X_test = scaler.transform(X_test)

Scikit-learn Training

python
from sklearn.ensemble import RandomForestClassifierfrom sklearn.metrics import classification_report, accuracy_score
model = RandomForestClassifier(n_estimators=100, random_state=42)model.fit(X_train, y_train)
y_pred = model.predict(X_val)print(classification_report(y_val, y_pred))

PyTorch Training

python
import torchimport torch.nn as nn
class Model(nn.Module):    def __init__(self, input_dim):        super().__init__()        self.layers = nn.Sequential(            nn.Linear(input_dim, 64),            nn.ReLU(),            nn.Dropout(0.3),            nn.Linear(64, 32),            nn.ReLU(),            nn.Linear(32, 1),            nn.Sigmoid()        )
    def forward(self, x):        return self.layers(x)
model = Model(X_train.shape[1])optimizer = torch.optim.Adam(model.parameters(), lr=0.001)criterion = nn.BCELoss()
for epoch in range(100):    model.train()    optimizer.zero_grad()    output = model(X_train_tensor)    loss = criterion(output, y_train_tensor)    loss.backward()    optimizer.step()

Evaluation Metrics

TaskMetrics
ClassificationAccuracy, Precision, Recall, F1, AUC-ROC
RegressionMSE, RMSE, MAE, R²

Complete Framework Examples

  • PyTorch: See references/pytorch-training.md [blocked] for complete training with:

    • Custom model classes with BatchNorm and Dropout
    • Training/validation loops with early stopping
    • Learning rate scheduling
    • Model checkpointing
    • Full evaluation with classification report

  • TensorFlow/Keras: See references/tensorflow-keras.md [blocked] for:

    • Sequential model architecture
    • Callbacks (EarlyStopping, ReduceLROnPlateau, ModelCheckpoint, TensorBoard)
    • Training history visualization
    • TFLite conversion for mobile deployment
    • Custom training loops

Best Practices

Do:

  • Use cross-validation for robust evaluation
  • Track experiments with MLflow
  • Save model checkpoints regularly
  • Monitor for overfitting
  • Document hyperparameters
  • Use 70/15/15 train/val/test split

Don't:

  • Train without a validation set
  • Ignore class imbalance
  • Skip feature scaling
  • Use test set for hyperparameter tuning
  • Forget to set random seeds

Known Issues Prevention

1. Data Leakage

Problem: Scaling or transforming data before splitting leads to test set information leaking into training.

Solution: Always split data first, then fit transformers only on training data:

python
# ✅ Correct: Fit on train, transform train/val/testscaler = StandardScaler()X_train = scaler.fit_transform(X_train)X_val = scaler.transform(X_val)  # Only transformX_test = scaler.transform(X_test)  # Only transform
# ❌ Wrong: Fitting on all dataX_all = scaler.fit_transform(X)  # Leaks test info!

2. Class Imbalance Ignored

Problem: Training on imbalanced datasets (e.g., 95% class A, 5% class B) leads to models that predict only the majority class.

Solution: Use class weights or resampling:

python
from sklearn.utils.class_weight import compute_class_weight
# Compute class weightsclass_weights = compute_class_weight('balanced', classes=np.unique(y_train), y=y_train)model = RandomForestClassifier(class_weight='balanced')
# Or use SMOTE for oversampling minority classfrom imblearn.over_sampling import SMOTEsmote = SMOTE()X_resampled, y_resampled = smote.fit_resample(X_train, y_train)

3. Overfitting Due to No Regularization

Problem: Complex models memorize training data, perform poorly on validation/test sets.

Solution: Add regularization techniques:

python
# Dropout in PyTorchnn.Dropout(0.3)
# L2 regularization in scikit-learnRandomForestClassifier(max_depth=10, min_samples_split=20)
# Early stopping in Kerasfrom tensorflow.keras.callbacks import EarlyStoppingearly_stop = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)model.fit(X_train, y_train, validation_data=(X_val, y_val), callbacks=[early_stop])

4. Not Setting Random Seeds

Problem: Results are not reproducible across runs, making debugging and comparison impossible.

Solution: Set all random seeds:

python
import randomimport numpy as npimport torch
random.seed(42)np.random.seed(42)torch.manual_seed(42)if torch.cuda.is_available():    torch.cuda.manual_seed_all(42)

5. Using Test Set for Hyperparameter Tuning

Problem: Optimizing hyperparameters on test set leads to overfitting to test data.

Solution: Use validation set for tuning, test set only for final evaluation:

python
from sklearn.model_selection import GridSearchCV
# ✅ Correct: Tune on train+val, evaluate on testparam_grid = {'n_estimators': [50, 100, 200], 'max_depth': [5, 10, 15]}grid_search = GridSearchCV(RandomForestClassifier(), param_grid, cv=5)grid_search.fit(X_train, y_train)  # Cross-validation on training setbest_model = grid_search.best_estimator_
# Final evaluation on held-out test setfinal_score = best_model.score(X_test, y_test)

When to Load References

Load reference files when you need:

  • PyTorch implementation details: Load references/pytorch-training.md for complete training loops with early stopping, learning rate scheduling, and checkpointing
  • TensorFlow/Keras patterns: Load references/tensorflow-keras.md for callback usage, custom training loops, and mobile deployment with TFLite

來源與署名

來源：secondsky/claude-skills位於plugins/ml-model-training/skills/ml-model-training提交8837836

授權條款： MIT

內容歸原作者所有。SourceWeft 從公開儲存庫中收錄這些內容。

檢舉或申請下架

更多來自 secondsky/claude-skills 的技能

Vulnerability Scanning

secondsky

指導使用 Trivy、Snyk、npm audit 與 Bandit 對相依套件、程式碼和容器進行自動化漏洞掃描。

Security22710 天前更新

Web Performance Optimization

secondsky

透過程式碼分割、延遲載入、快取、壓縮與 Core Web Vitals 監控來指導網頁效能最佳化。

Software Development22710 天前更新

Web Performance Audit

secondsky

依據 Core Web Vitals、Lighthouse、最佳化策略與效能預算指導網頁效能稽核。

Software Development22710 天前更新

Threejs

secondsky

用於以 Three.js 打造 3D 網頁體驗的參考知識庫，涵蓋十個核心領域並提供程式碼範例。

Software Development22710 天前更新

Tailwind V4 Shadcn

secondsky

設定 Tailwind CSS v4 與 shadcn/ui、Vite 和 React，涵蓋主題、深色模式與移轉指引。

包含腳本
Software Development22710 天前更新

Technical Specification

secondsky

指導撰寫軟體專案技術規格書，涵蓋需求、架構、API、測試與風險。

Software Development22710 天前更新

更多Data & Analytics技能

Swap Curve Strategy

anthropics

精選

運用 MCP 定價工具分析利率交換曲線、交換利差、實質利率與曲線交易機會。

Data & Analytics39K2 週前更新

Unit Economics

anthropics

精選

為私募股權標的分析單位經濟：ARR 世代、LTV/CAC、淨留存、回收期與利潤瀑布。

Data & Analytics39K2 週前更新

Option Vol Analysis

anthropics

精選

結合波動率曲面、含希臘字母的選擇權定價與歷史價格，比較隱含波動率與已實現波動率。

Data & Analytics39K2 週前更新

Macro Rates Monitor

anthropics

精選

運用 MCP 的 GDP、通膨、殖利率曲線、損益兩平通膨與交換利率資料，建立總體與利率儀表板。

Data & Analytics39K2 週前更新

Clean Data Xls

anthropics

精選

清理雜亂的試算表資料：修剪空白、統一大寫、轉換文字數字、標準化日期並移除重複項。

Data & Analytics39K2 週前更新

Fixed Income Portfolio

anthropics

精選

透過債券定價、彙總風險指標、預測現金流量並執行利率情境，審查固定收益投資組合。

Data & Analytics39K2 週前更新