-
I have trained 130400 iterations with batch_size 8 on flying chairs dataset. The predicted flows on test set are still not good. Should I continue or switched to FlownetCS?
-
My device is :
> Ubuntu16.04
> SSD 850EVO
> GPU: 2 x TITANX Pascal
The forward&backward cost for each iteration is 1.5s
It seems that your method will continue for **10,000** epoch (each ep…
-
1. Using **--pretrained** to continue training on previously trained data set seems to overwrite results?
2. Is it meant to be used with **--start-epoch** to manually specify epoch to be restarted on…
bkvie updated
6 years ago
-
Have been coming on and off on Homebrewery, writing bits here and there for DnD campaign. However I have tried using it today, and yesterday, and some pages are missing, and things are in places they …
-
-
per @marcoscaceres comment in #44
-
Add mimic to mineshaft
-
```
using System;
using System.Collections.Generic;
using System.Linq;
using System.Net.Http;
using System.Text;
using System.Threading.Tasks;
using Newtonsoft.Json;
namespace Matterhook.N…
-
We propose that WebGPU accept HLSL shaders as input.
1. HLSL is by far the most common source language for shaders. There is a huge amount of existing content as well as an extremely large, stable …
grorg updated
6 years ago
-
# Working Drafts
- [Gratipay](https://docs.google.com/document/d/1O7PPVnnaK2ML6eNVgUwYiTbaNiqn92Gb3CZMufY1VqY/edit?usp=sharing)
- [Chad](https://docs.google.com/document/d/12brjiQzVNETdYrE-6r1p7-Ey_…