qqqsd

you can use image = tf.to_float(image) to solve this problem.

> @MBoaretto25我们必须在哪里使用此代码？哪个 .py 文件？请明确说明 写在train.tripletloss.py image.set_shape()之前

Running forward pass on sampled images: 165.140 Selecting suitable triplets for training (nrof_random_negs, nrof_triplets) = (17714, 17667): time=168.161 seconds Traceback (most recent call last): File "C:\Users\DELL\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\client\session.py", line 1375, in _do_call...

我想问一下你们解决训练精度一直是0的问题吗，我也遇到了