bilylee

A simple modification to reproduce the results is scaling the outputs of the euclidean distance. That is, ```python feature_dims = 1600 # 1600 for miniimagenet, 64 for omniglot learnable_scale =...

报错上来看某些文件没有找到。请确认一下：模型文件、OTB数据集等是否存在？是否严格按照 README 的步骤在 terminal 中执行？

这是为了对训练图像做 +- 4 个 pixel 的平移。如果直接 `RandomCrop(255 - 2 * 8)` 会导致平移 +- 8 个 pixel，所以先CenterCrop，再 RandomCrop。相关问题 #19 #32 #56

Please follow exactly the same procedures listed in README to run the code. Don't skip step 2. And no `cd` thereafter. ``` # 1. Clone this repository to your disk...

Please update your matplotlib. It requires `matplotlib >= 2.0.0`

Did you have "SiamFC-TensorFlow/datasets/dataloader.py" in your directory? Moreover, Please note that the repo is developed under python2. I am not sure if it works under python3.

Hi, Sorry, this is out of my expertise.

num_step = examples_per_epoch * num_epoch / batch_size 请调节这三个参数来达到你的目的。 https://github.com/bilylee/SiamFC-TensorFlow/blob/c4f21d137257cb7685a8e3e74105f9be0d24afb4/configuration.py#L49-L50

Please take a look of #40 and see if it helps.

Hi, I use `tf.nn.sigmoid_cross_entropy_with_logits` for computing the loss while the original MatConvNet version use something like `tf.nn.softmax_cross_entropy_with_logits` to compute the loss. For `tf.nn.sigmoid_cross_entropy_with_logits`, the label is the probability of the...