-
## 🚀 Feature
allow nnpack to install gracefully with pytorch on linux arm devices
## Motivation
pytorch+NNPack should be one of the fastest options for arm.
## Pitch
pytorch is already pr…
-
# Next paper candidates
Let's propose papers to study next! All papers mentioned in the comments of this issue will be listed in the next vote.
## Last session runner-up
[Revisiting Graph Neural …
-
## Description
I experience a performance regression for both CPU (mxnet-mkl) and GPU (mxnet-cu100) when updating from version 1.4.1 to 1.5.0. The speed regression is about 8% on CPU, if you're inter…
-
In the realm of image processing, Circular Convolution is common used because it is suitable to do FFT.
Circular Convolution means that firstly padding the tensor with circular boundary and then do …
ysnan updated
5 years ago
-
Hi there,
The cwt() code currently use a discrete convolution algorithm which is `~O( len(data) * scale`).
While this is usually fine for common values of data and scale < 100, it becomes an issue…
-
@mittagessen thank you for your hard work
- What benefits does ntcn_ocr have, what are you trying to achieve?
- Does it have similar concept to[ parallel wavenet](https://youtu.be/hzpxXZJQNFg?t=71…
mrocr updated
5 years ago
-
Hi
I wanted to run the pretrained frozen .pb models from [mobilenetv1](https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet_v1.md) and [mobilenetv2](https://github.com/tens…
-
This is the definition of my model:
```python
import torchvision
import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, in_dim,…
-
I am trying to build `libtensorflow_inference.so`, in order to run a TF model in an Android app. I found the following command in issue #6166, however the command is failing. I reported an earlier e…
-
Following https://github.com/huggingface/pytorch-pretrained-BERT/issues/126:
Parameters `predict_fp16`, `max_seq_length` and `predict_batch_size` should be tunable at predict time:
For train:
…