magictut

AI修改后的脚本，依然无法运行 bash #!/bin/bash NUM_WORKERS=1 NUM_GPUS_PER_WORKER=1 MP_SIZE=1 script_path=$(realpath $0) script_dir=$(dirname $script_path) main_dir=$(dirname $script_dir) MODEL_TYPE="visualglm-6b" MODEL_ARGS="--max_source_length 64 \ --max_target_length 256 \ --lora_rank 10\ --pre_seq_len 4" OPTIONS_SAT="SAT_HOME=$1" # OPTIONS_SAT="SAT_HOME=/raid/dm/sat_models" OPTIONS_NCCL="NCCL_DEBUG=info NCCL_IB_DISABLE=0 NCCL_NET_GDR_LEVEL=2" HOST_FILE_PATH="hostfile...

> 你是在windows运行的吗？为什么会有\r呢 V100 linux上运行 (visualGLM) root@iZbp1ewp3ew1qt4u8bdh0iZ:~/ai/VisualGLM-6B# ssh -V OpenSSH_8.9p1 Ubuntu-3ubuntu0.1, OpenSSL 3.0.2 15 Mar 2022

> 你是在windows运行的吗？为什么会有\r呢 原脚本VisualGLM-6B/finetune/finetune_visualglm.sh #! /bin/bash NUM_WORKERS=1 NUM_GPUS_PER_WORKER=1 MP_SIZE=1 script_path=$(realpath $0) script_dir=$(dirname $script_path) main_dir=$(dirname $script_dir) MODEL_TYPE="visualglm-6b" MODEL_ARGS="--max_source_length 64 \ --max_target_length 256 \ --lora_rank 10\ --pre_seq_len 4" # OPTIONS_SAT="SAT_HOME=$1" #"SAT_HOME=/raid/dm/sat_models" OPTIONS_NCCL="NCCL_DEBUG=info NCCL_IB_DISABLE=0...

> > > 你是在windows运行的吗？为什么会有\r呢 > > > > > > 原脚本VisualGLM-6B/finetune/finetune_visualglm.sh > > #! /bin/bash NUM_WORKERS=1 NUM_GPUS_PER_WORKER=1 MP_SIZE=1 > > script_path=$(realpath 0)scriptdir=(dirname scriptpath)maindir=(dirname $script_dir) MODEL_TYPE="visualglm-6b" MODEL_ARGS="--max_source_length 64 --max_target_length 256 --lora_rank...

> 更新deepspeed版本 怎么更新deepspeed版本？可以告知一下命令么

> @magicwang1111 `pip install --upgrade deepspeed` 谢谢

> I am not entirely sure if the configuration of this project can be completed on a Win system. We trained and inferred using Ubuntu and Python 3.8 environment. thx,i...

> Can you post your workflow as is? Will look into this. I would suggest removing your negative prompt and reducing CFG scale - the text guidance is probably overriding...

(flux) [wangxi@v100-4 SimpleTuner]$ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2022 NVIDIA Corporation Built on Wed_Sep_21_10:33:58_PDT_2022 Cuda compilation tools, release 11.8, V11.8.89 Build cuda_11.8.r11.8/compiler.31833905_0 (flux) [wangxi@v100-4 SimpleTuner]$

(flux) [wangxi@v100-4 SimpleTuner]$ pip show bitsandbytes Name: bitsandbytes Version: 0.35.0 Summary: 8-bit optimizers and matrix multiplication routines. Home-page: https://github.com/TimDettmers/bitsandbytes Author: Tim Dettmers Author-email: [email protected] License: MIT Location: /mnt/data/wangxi/miniconda3/envs/flux/lib/python3.10/site-packages Requires: Required-by:...