Trainer issues

Add support for Python 3.12

1

Seems trivial, all tests pass locally.

[Bug] ValueError: not allowed to raise maximum limit (rlimit)

1

### Describe the bug Error while training:- - I tried with sudo same error - I am using docker image nvidia/cuda:11.7.0-base-ubuntu22.04 - The default value of the docker container for...

iamkhalidbashir

bug

[Feature request] Multi-node training

2

Hi, I have two questions: 1. Can it be used in multi-node training? 2. When will trainer support deepspeed? I have noticed that integrating deepspeed is in to-do list, but...

yjzhong111

feature request

[Bug] trainer.distribute not working

1

### Describe the bug For single GPU training, I am using train_yourtts.py. When I switch to multi-gpu, the program could run but didn't show acceleration. I checked the code in...

xyyimian

bug

### Describe the bug Hi everybody, Trainer is great, but still uses [APEX,](https://github.com/NVIDIA/apex) which is deprecated and tends to cause problems. Could you remove and/or replace with Nvidia/amp? Best regrads...

mueller91