Updated 4/11/2018
Here's my experience of installing the NVIDIA CUDA kit 9.0 on a fresh install of Ubuntu Desktop 16.04.4 LTS.
| Ninj0r admin, [Oct 20, 2017, 9:18:55 AM]: | |
| It's a three step process: | |
| 1) Start listening to the stream and buffering the messages | |
| 2) Get a depth snapshot | |
| 3) replay the buffered messages and the live messges. | |
| Depth updates have two variables, u and U | |
| U is the initial updateId, and u is the final updateId. There can be multiple updates "compressed" into a single update that comes out via the web socket stream. |
| import numpy as np | |
| from keras.models import Model | |
| from keras.layers import Dense, Activation, Lambda, Input | |
| import keras.backend as K | |
| from keras.utils import to_categorical | |
| # Model definition | |
| def foo(ip): | |
| a = ip[1] | |
| x = ip[0] |
Updated 4/11/2018
Here's my experience of installing the NVIDIA CUDA kit 9.0 on a fresh install of Ubuntu Desktop 16.04.4 LTS.
Most of programs will not accept an email using just @localhost as domain.
So, edit /etc/hosts file to make the domain localhost.com point to your machine, including this content to the file:
127.0.0.1 localhost.com