1 |
пн, 10 янв. 2022 г. в 01:29, gevisz <gevisz@×××××.com>: |
2 |
> |
3 |
> вс, 9 янв. 2022 г. в 16:52, Mark Knecht <markknecht@×××××.com>: |
4 |
> > |
5 |
> > On Sun, Jan 9, 2022 at 3:59 AM gevisz <gevisz@×××××.com> wrote: |
6 |
> > > |
7 |
> > > вс, 21 нояб. 2021 г. в 17:12, Mark Knecht <markknecht@×××××.com>: |
8 |
> > > > |
9 |
> > > > Congrats! |
10 |
> > > |
11 |
> > > Thank you. However, it was not for long. On 30-12-2021 recompilation |
12 |
> > > of the same tensorflow-2.7.0 because of some changed dependencies |
13 |
> > > failed with the same f**ng "Bazel failed" error as before. |
14 |
> > > |
15 |
> > > So, I am currently going to degrade my Gentoo system to the state it |
16 |
> > > was in on 12-12-2021, when its last update was successful and froze it |
17 |
> > > forever. |
18 |
> > > |
19 |
> > > The problem is that I do not know how to do it but I am going to post |
20 |
> > > this question as a separate thread. (I am using webrsync method.) |
21 |
> > > |
22 |
> > |
23 |
> > Sorry for the problems. I saw your other thread about downgrading |
24 |
> > Gentoo. I agree with the other responses you got there that Gentoo, in |
25 |
> > general, does not support standing still much less going backward. |
26 |
> > |
27 |
> > I will offer what will probably not be a popular comment but my |
28 |
> > opinion is Gentoo is exactly the wrong sort of distribution for doing |
29 |
> > work in tensorflow. With all of it's updates, limited testing of |
30 |
> > packages, extreme amounts of code building, and not being a distro |
31 |
> > that the official tensorflow folks even verify on, it's just too hard. |
32 |
> > (And a contributing factor to how I moved away in the beginning.) |
33 |
> > |
34 |
> > My thought is that you might create a 20.04 LTS Ubuntu VM (or possibly |
35 |
> > an LXC container) running whatever your desktop flavor of Gentoo is - |
36 |
> > I run Kubuntu - and just run tensorflow in the VM. You won't easily |
37 |
> > get GPU support unless you deal with passthrough, but the software |
38 |
> > will just work and you won't spend time dealing with building code |
39 |
> > which can be spent coding tensorflow. |
40 |
> > |
41 |
> > If you insist on running in Gentoo consider the LXD container running |
42 |
> > an older rev of Gentoo. (If you can find one) Get it working, if you |
43 |
> > can, and then never update it. |
44 |
> > |
45 |
> > lxc image list images: gentoo |
46 |
> > |
47 |
> > There are openrc and systemd versions available, but a Kubuntu stable |
48 |
> > container would more likely to 'just work' IMO. |
49 |
> |
50 |
> Thank you for your reply, Mark. |
51 |
> |
52 |
> Unfortunately, you missed my previous message in this thread |
53 |
> where I wrote that I do have Ubuntu 20.04 on the same computer. |
54 |
|
55 |
Correction: |
56 |
|
57 |
> However, tensorflow fails to run on it because it is compiled |
58 |
> to be inconsistent with my videocard. So, Gentoo is my only option |
59 |
> for this hardware. |