把数据增强当作免费数据
通过变换扩充数据集
把数据增强当作免费数据 是 CoddyKit 上的免费 Deep Learning Academy 课时。 这是第 4 节课，共 4 节。 你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。 这是 Deep Learning Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Deep Learning Academy 课程共包含 4 节课。
本课时的部分内容尚未翻译，以英文显示。
Free Extra Data
Data augmentation creates new training examples by transforming the ones you already have, giving the model more variety for free. 🎁
Why It Fights Overfitting
Showing slightly changed versions each epoch stops the model from memorizing exact pixels and pushes it toward robust patterns.
Flips and Rotations
For images, simple flips and small rotations teach the model that a cat is still a cat when mirrored or tilted.
Crops and Resizes
Random crops and resizes shift the framing of an object, so the network learns to recognize it anywhere in the frame.
Color Jitter
Tweaking brightness, contrast, and hue with color jitter makes the model less fragile to lighting and camera differences.
Compose in torchvision
You chain transforms with transforms.Compose, and torchvision applies them on the fly as each image is loaded.
tf = transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.RandomCrop(32, padding=4)])Applied Per Batch
Augmentation runs inside the Dataset, so every epoch sees freshly transformed images and the dataset feels far larger.
Keep It Realistic
Only use transforms that preserve the label. Flipping a digit like 6 into 9 would corrupt your data instead of enriching it.
No Augmentation at Test
Apply augmentation to training only. For validation and test, use clean, fixed transforms so your scores stay comparable.
Beyond Images
Augmentation is not just for pictures. Text uses synonym swaps and audio uses time shifts and added noise to grow datasets.
Strong Mixing Tricks
Modern recipes blend whole images with Mixup and CutMix, mixing samples and labels for even stronger regularization.
Quick Check
Decide where data augmentation should and should not be applied.
Recap
You learned data augmentation: label-preserving transforms on training data that expand variety, fight overfitting, and cost nothing. 🌱
用 AI 导师学习 Python — 免费
在浏览器中编写并运行真实代码，获得全天候 AI 导师的即时帮助，并在网页或应用中继续学习。
- 课程
- 30
- 课程
- 120
常见问题解答
「把数据增强当作免费数据」课时是免费的吗？
是的 — 「把数据增强当作免费数据」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Deep Learning Academy 课程的其余内容，请升级到 CoddyKit PRO。 Deep Learning Academy 课程共包含 4 节课。
「把数据增强当作免费数据」这节课中我会学到什么？
通过变换扩充数据集 你通过在浏览器中直接运行的动手代码来练习 Deep Learning Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。
学习 Deep Learning Academy 需要有经验吗？
无需任何先前经验。CoddyKit 上的 Deep Learning Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。 这是第 4 节课，共 4 节。
「把数据增强当作免费数据」课时需要多长时间？
大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。
我能在这节 Deep Learning Academy 课中编写并运行代码吗？
能。每节 Deep Learning Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。
此课程中的所有课时
- 解读训练集与验证集之间的差距
- Dropout：随机丢弃神经元
- 批归一化与层归一化
- 把数据增强当作免费数据