Back to Details

learn-deep-learning

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

深度学习学习路线

Deep Learning Learning Path

目标

Objectives

  • 给出可执行的学习计划（按周/按章节）。
  • 给出实践路线（从简单到复杂）。
  • Provide actionable learning plans (weekly / chapter-wise).
  • Provide a practical learning path (from simple to complex).

推荐资源

Recommended Resources

  • 书籍：Deep Learning（Goodfellow）、Dive into Deep Learning
  • 课程：CS231n、CS224n、DeepLearning.AI
  • 工具：PyTorch、Lightning、Weights & Biases
  • Books: Deep Learning (Goodfellow), Dive into Deep Learning
  • Courses: CS231n, CS224n, DeepLearning.AI
  • Tools: PyTorch, Lightning, Weights & Biases

实践项目建议

Practical Project Recommendations

  • 图像分类：CIFAR10 → ImageNet 子集
  • NLP：文本分类 → 微调小模型
  • 训练与评估：记录超参、对照实验、可复现
  • Image classification: CIFAR10 → ImageNet subset
  • NLP: text classification → fine-tuning small models
  • Training and evaluation: record hyperparameters, control experiments, reproducibility