-
设备为两台linux,每台2张A100 40G显卡:A100(40G) * 2
训练命令如下:主节点命令为CUDA_VISIBLE_DEVICES=0,1 NNODES=2 NODE_RANK=0 NPROC_PER_NODE=2 MASTER_ADDR=127.0.0.1 swift sft --model_type qwen1half-7b-chat --model_id_or_path /…
-
The current way to add new units is
1. annoying
2. verbose
3. littered all over
Instead we need to unify the unit construction. Instead of having:
- units defined as full name
- units defined …
-
Hello,
Just wanted to share my results on finetuning Llama3.1-8B-Instruct (4bit bnb, training took 1h30 on 2xA100 80GB, 32 epochs). Many thanks for the scripts, they worked very well, and I hope th…
-
使用flask搭建模型服务,模型从modelscope加载,报错RuntimeError: shape '[1, 279, 64, 128]' is invalid for input of size 285696
@app.route("/chat", methods=["POST"])
def chat():
content = request.json
print(c…
-
**Describe the bug**
Using the PEFT library with auto-gptq worked fine with 0.4.2, but now fails with v0.5.0
```
File "/root/miniconda3/envs/py3.10/lib/python3.10/site-packages/peft/mapping.py"…
-
While running
model, tokenizer = load_model(model_name, bnb_config)
I am getting the following error,
---------------------------------------------------------------------------
AttributeErro…
-
I try to run llava-v1.6-34b-hf-awq and sucessed, but how can I run the test for Llava-v1.5 ConditionalGeneration?
https://github.com/casper-hansen/AutoAWQ/pull/250
The bug of example likely :
1. ma…
-
### OS
Windows
### GPU Library
CUDA 12.x
### Python version
3.12
### Pytorch version
2.4.1+cu121
### Model
Qwen/Qwen2.5-72B-Instruct
### Describe the bug
Qwen 2.5 72B …
-
**Describe the bug**
A clear and concise description of what the bug is.
无法调用qwen-7b-chat-int4的模型,问题如下:
代码:
```
import time
import torch
from transformers import AutoModelForCausalLM, AutoToken…
-
# Todo
https://github.com/OCA/maintainer-tools/wiki/Migration-to-version-14.0
# Modules to migrate
- [x] delivery_package_default_shipping_weight - By @dzungtran89 - #793
- [x] procurement_…