-
Hi Phil, I tested it in my private project 2 days ago, and it seems to speed up learning quite significantly, not sure that final val/train losses are better, more like very similar to original but it…
-
Hi,
I've noticed that my BrainScript network trains **much** slower with version 2.4 than with version 2.3.
In CNTK 2.3 it trains with 25.4 samples/s and in CNTK 2.4 only with 11.7 samples/s. In P…
-
您好。打扰了。我想问下AssembledBlock是您自己浮现的还是AsConvSR: Fast and Lightweight Super-Resolution Network with Assembled Convolutions他们官方的代码？
-
Thank you so much for your reimplementation of ELIC. However, in ELIC's original paper, they reported that the decoding latency of ELIC is much less than 100ms. But when I test with ELIC from Compress…
-
Hi tianzhi,
I am Nic from NVIDIA, working on AI projects.
Thanks very much for your sharing here!
Now I am investigating how to optimize CTPN inference with faster convolutions.
Could you please…
-
With #26 fixed on my side, I was able to perform some benchmarks now. The libDNN generated convolutions are about 3x-4x faster than my naive kernel described in #26, which is very nice! But they are s…
romix updated
6 years ago
-
Currently we can have a fast result by focusing on small-field imaging. It's clearly described in the paper. But users often don't read the fine print and may get inaccurate results for wide-field ima…
-
We will need to look into that at some point, since at the moment everything is implemented in the most naive way possible :cry:. Some options worth thinking about.
### Use `np.convolve` or similar…
-
DenseNets are typically implemented with long strings of concatenations, which require O(n^2) time for O(n) convolutions just for memory copies. If the memory can be pre-allocated and then sliced, the…
-
## 🚀 Feature
Add option of Eigen Tensor library implementation of convolution on ARM CPU
## Motivation
While convolutions on Intel CPUs are quite fast, there are issues on ARM processors. I tri…