-
-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathconda setup (for gpu).txt
43 lines (23 loc) · 1.41 KB
/
conda setup (for gpu).txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
-------
Windows
-------
First, remove conda-forge from your channel list (if you do not have that, it will return error, but no issue)
-------------------------------------------------------------------------------------------------------------
conda config --remove channels conda-forge
Second, clean all the previously downloaded packages (to make sure after that you have installed the right packages)
--------------------------------------------------------------------------------------------------------------------
conda clean --all
(you might want to run the above twice if you encounter issue with your installed environment)
Third, create the rtav environment
----------------------------------
conda create -n rtav python=3.6
Fourth, activate the rtav environment
-------------------------------------
conda activate rtav
Fifth, install opencv and some packages from conda-forge first
--------------------------------------------------------------
conda install opencv=4.1.0 imutils=0.5.2 imgaug librosa -c conda-forge
Finally, install the rest of the packages
-----------------------------------------
conda install tensorflow=1.13.1 tensorflow-gpu=1.13.1 cudatoolkit=9.0 cudnn=7.1.4 spyder=3.3.2 cython=0.29.2 pathlib=1.0.1 ipython=7.2.0 yaml pandas keras keras-gpu pydot graphviz
(Assume Cuda 9.0 is installed, and cudnn 7.1.4 installed, get cudnn from https://developer.nvidia.com/rdp/cudnn-archive)