-
Current implementation of fused winograd convolution uses very limited subset of underlying kernel implementation.
Current limitations are:
- only 2x3 version
- no dilation support
- no grouped …
-
Dear irvinxav:
Thanks for your code. In you paper writted “In our model, two convolutional layers (depthwise convolutional layer and pointwise convolutional layer) is replaced by one HetConv c…
-
Hi,
Can you please extend this to hidden/conv layers?
-
What hardware and operating system/distribution are you running?
Operating system: CentOS7
CUDA version: 8.0
CUDNN version: .6.0
openCV version: 3.0
BLAS: open
Python version: 3.6.3
boost 1.6…
-
https://github.com/ROCmSoftwarePlatform/MIOpen/blob/4e61a3ebdfe2b07b4d331cc46832eb4a6b49941c/src/solver.cpp#L264
Can we introduce some dedicated primitive (e.g. `Fused`) for fused convolutions and …
-
Dear author, in order to use redundant feature information, PConv is proposed in this paper. However, Partial convolution is so similar to Ghost convolution. Ghost convolution uses half of the channel…
-
This project classifies audio samples from urban environments into one of 10 classes. The dataset, known as UrbanSound8K, contains 8732 sound excerpts, each 4 seconds or shorter, representing urban so…
-
Here is my dataset example:
https://imgur.com/a/lYbNYBa
Im training on this dataset to detect alphanumeric on image, but after I trained a thousand steps, average loss increase from **1.7** to ver…
-
This is my command: ./darknet detector train cfg/person.data cfg/yolov3-person.cfg darknet53.conv.74 -gpus 0,1,2,3
Here is my cfg:
```
[net]
# Testing
# batch=1
subdivisions=1
# Training
bat…
-