镜头工作室的SnAPML模型问题
我构建了一个自定义模型,用于使用TensorFlow和Keras对汽车的图像进行分类,以构建由机器学习提供动力的快速镜头。镜头工作室仅接受量化的模型;该模型必须使用TFLITE模块进行量化过程。
但是,问题在于传递到镜头工作室的模型无法正常运行。它仅在启动后首次显示分类结果;然后,尽管图像/视频更改,结果(甚至分类背后的概率数)仍然是静态的。
关于如何解决此问题的任何提示将不胜感激。输入图像设置的配置与SNAP提供的模板相同。
I built a custom model for classifying images of cars using Tensorflow and Keras, to use it for building a Snap lens powered by machine learning. Lens Studio only accepts quantized models; the model had to go through the quantization process using the TFLite module.
However, the problem is that the model passed into Lens Studio is unable to function properly. It only displays classification results for the first time after its initiation; then the results (and even the probability numbers behind classification) remain static despite image/video changes.
Any tips on how to solve this issue would be appreciated. The configurations for input image setups remain identical as the provided template by Snap.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(1)
如果您共享代码以调查问题,那将是有益的。
但是,根据我在手册的
如果我们希望对一系列图像或视频进行培训,那么我们需要在它们上创建所有输入和 itterate 的列表/词典,并在其中制作一个计时器。
您可以在实验中做类似的事情:
我还希望您阅读此文章以更好地理解。
It would have been beneficial had you shared the code to probe into the issue.
However, as per what I have seen in the manual of SnapML, the code is available to train 1 model at a time.
If we want a series of images or videos to be trained then we need to create a list/dictionary of all the inputs and Iterate over them as well as put a timer to it.
You can do something like this in your experiment :
I also would like you to read this article for better understanding.