Tsun-Yi Yang
Tsun-Yi Yang
Since the limited computer resource, I only have a CPU-only computer. However, some pre-trained model is based on GPU. It seems that the torch installation will not contain cudnn when...
Hi, I am using torch openface for testing the performance of lfw dataset. Since your model is caffe based, I use loadcaffe library for torch to convert the model into...
請問一下蘇大，由於您的capsule_test裏的前一層採用了： Reshape((-1,128))應該是對應於primary capsule， 這是否是對應於128 capsule的dim而非數量？ 因為我稍微看了一下原本Hinton的paper和下面這個實現 https://github.com/XifengGuo/CapsNet-Keras 似乎都把8-dim 當作primary capsule的特徵維度。 雖然說dim和num就算反過來也完全可以做運算， 只是想問一下跟原paper之間的關係是否是想選擇128當作capsule num？ P.S. Blog真的解釋得非常好，真的獲益良多，感謝
I use the following setting, as suggested in the github. **L=40,k=12, no bottleneck** However, the parameter number is not **1M**, it's **0.6M**. This problem also happen when I turn **bottelneck**...
Hello. Very impressive project. Just want to ask, is it possible to provide some example data (e.g. 1 or 2 files) for the inference and visualization for 1) cad estimation...
您好，非常喜歡大大對於保障勞工和開源的精神， 由於身為一個學術研究的學生，我覺得對於paper的評論系統跟工作一樣是需要的， 然而因為本身不是做這方面的，比較想知道的是大大使用AWS當做這個work的伺服器嗎？ 如果我想做類似的事情是否一定需要一個server? 大大如果你是無償讓大家用天眼通但是AWS不是要付錢的嗎？這樣也太佛了吧？ 這幾個疑問煩請指教。謝謝
**model1:add(cudnn.SpatialConvolution(32, 64, 6, 6))** I replace the upper one with the lower three lines **model1:add(nn.LateralConvolution(32, 64)) model1:add(nn.VerticalConvolution(64, 64, 6)) model1:add(nn.HorizontalConvolution(64, 64, 6))** However, the training time is **10 times slower**...
Hi, The reason is because in the `FormattedLLMCall`, there is string .format to replace some part. This might fail due to '{', '}' exist in a more loosing format such...
Hi first of all super cool library, I find it very useful and inspiring. I am playing with it and found there is no control sometimes in terms of temperature....
Bedrock
Do you have plan to support model in bedrock? I think that would open up even more possibilities. Thanks :)