search

bigcode-project / bigcode-evaluation-harness

A framework for the evaluation of autoregressive code generation language models.
Apache License 2.0
710 stars 183 forks source link

code #146

Closed Lzzzzx closed 8 months ago

Lzzzzx commented 9 months ago

When I ran Codellama-34b with two A100 cards, I was able to run it for the first time at first, but when I ran again later, I reported an error：ConnectionError: Tried to launch distributed communication on port 29500, but another process is utilizing it. Please specify a different port (such as using the ----main_process_port flag or specifying a different main_process_port in your config file) and rerun your script. To automatically use the next open port (on a single node), you can set this to 0.

loubnabnl commented 9 months ago

Hi, it seems like you have processes using the same GPUs, maybe run nvidia-smi to make sure they are free before launching the evaluation again