Google Cloud Vision AutoML 和 TensorFlow 对象检测差异

Google Cloud Vision AutoML and TensorFlow Object Detection Differences

Google Cloud Vision Auto ML 和 Google Tensor Flow 有什么区别?

https://cloud.google.com/vision/automl/docs/train

https://www.tensorflow.org/hub/tutorials/object_detection

它们都是同一家公司生产的,进行API物体检测,允许用户训练图像模型,允许用户在python中编码。 Google Cloud AutoML 是否在后台使用 Google Tensorflow,仅在 Cloud 模型中,因此用户不需要下载软件?

只是想了解这两个组成部分。

Google 云自动机器学习

Google Cloud AutoML Vision runs on Google Cloud Platform and Google TensorFlow Object Detection 在 Google Colab 上运行。

几乎没有什么主要区别。我将尝试在此处介绍最重要的内容:

  1. Colab 实例会话在 where the maximum lifespan for a custom AutoML vision model is 18 months 后超时。

  2. Colab 在 Google Drive and its basic version is free where AutoML vision costs .15 per node hour.

    上运行

Note: "You can try AutoML Vision Object Detection for free by using 40 free node hours each for training and online prediction, and 1 free node hour for batch prediction, per billing account."

  1. Colab 易于身份验证,并且 pre-built connectors AutoML 与 Google Cloud Platform 完全集成。

  2. Colab 是一款没有任何企业支持的消费产品,AutML vision enterprise support 可以解决您的所有问题和疑问。


  • 对于最高级别的编程控制,最好使用 TensorFlow 创建自定义对象检测模型。可以找到一些示例 here. To reduce time to production, the customer can use transfer learning and use one of the available model architectures in TensorFlow Hub 或从其他来源下载模型权重。

  • 另一方面,AutoML Vision 不提供对模型构建的精细控制,因此您无法以编程方式调整生成的模型。 AutoML 使用神经 Architecture Search to design optimal neural network for the task at hand. This approach is faster since all that’s needed is training data and task to build a neural network for. You can find more details and examples on AutoML Vision.

  • 如果您想使用现成的 API(对模型本身进行零定制),他们可以直接利用 Vision API. The API serves predictions for common tasks (such as object detection, OCR, etc) and allows no control, without having to bring any data or train any ML model. The downside here is that the model behind the API is static and cannot be retrained like AutoML Vision. More details on the difference between Vision API and AutoML Vision can be found here.

  • 关于哪种产品在对象检测方面做得更好 - 您应该进行自己的实验,看看哪种产品最适合您的需求。

最后为增加程序化控制顺序,如下:

Vision API > AutoML Vision > TensorFlow (with transfer learning) > TensorFlow (custom model)