Search results

关于llava数据

Status: Closed (completed).

#205 In yuanzhoulvpi2017/zero_nlp;

· by zpc2002zpc was closed on Jan 13, 2025

To those who meet "AssertionError: no_sync context manager is incompatible with gradient partitioning logic of ZeRO stage 2"

Status: Closed (completed).

#204 In yuanzhoulvpi2017/zero_nlp;

· by Nyquist24 was closed on Feb 5, 2025

Train_Llava

llava

Status: Closed (completed).

#203 In yuanzhoulvpi2017/zero_nlp;

· by xupengfei-dr was closed on Dec 11, 2024

4 comments

code03_build_model_show.ipynb

Status: Closed (completed).

#201 In yuanzhoulvpi2017/zero_nlp;

· by Wfyikczy was closed on Nov 26, 2024

7 comments

Train_llava。 缺少preprocessor_config.json. 文件。

Status: Closed (completed).

#199 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Dec 15, 2024

3 comments

我发现有个issue 也是类似的错误：

Status: Closed (completed).

#198 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Oct 31, 2024

train_llava推理结果有问题。

Status: Closed (completed).

#197 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Jan 2, 2025

7 comments

train_llava关于保存权重文件的问题

Status: Closed (completed).

#196 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Jan 1, 2025

1 comment

关于训练LLava的问题

Status: Closed (completed).

#195 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Jan 1, 2025

1 comment

train_llava 如何实现第一阶段预训练和第二阶段微调

Status: Closed (completed).

#194 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Jan 1, 2025

2 comments

train_llava 关于数据集构建的问题

llava

Status: Closed (completed).

#192 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Nov 10, 2024

2 comments

train_llava 构建模型失败

llava

Status: Closed (completed).

#191 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Nov 7, 2024

3 comments

关于train_llava中processor的疑问。

llava

Status: Closed (completed).

#189 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Oct 30, 2024

1 comment

关于train_llava代码的TrainLLavaModelCollator的疑问

llava

Status: Closed (completed).

#188 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Oct 29, 2024

1 comment

train_llava 报错ValueError: Cannot use chat template functions because tokenizer.chat_template is not set and no template argument was passed!

llava

Status: Closed (completed).

#187 In yuanzhoulvpi2017/zero_nlp;

· by weiaicunzai was closed on Oct 27, 2024

2 comments

train_llava数据填充是否有问题？

Status: Closed (completed).

#186 In yuanzhoulvpi2017/zero_nlp;

· by AI-Study-Han was closed on Oct 23, 2024

1 comment

请问train_llava如何使用更大的model

Status: Closed (completed).

#183 In yuanzhoulvpi2017/zero_nlp;

· by 66246764 was closed on Aug 13, 2024

llava合并模型后加载推理报错

Status: Closed (completed).

#180 In yuanzhoulvpi2017/zero_nlp;

· by daihuidai was closed on Jul 1, 2024

2 comments

pretrain进行了设置仍然oom

Status: Closed (completed).

#179 In yuanzhoulvpi2017/zero_nlp;

· by TuuSiwei was closed on Jun 30, 2024

internlm-sft 单机多卡微调 GPU 利用率低

Status: Closed (completed).

#170 In yuanzhoulvpi2017/zero_nlp;

· by Shamepoo was closed on Mar 21, 2024

5 comments

能出一个ChatGLM

Status: Closed (completed).

#163 In yuanzhoulvpi2017/zero_nlp;

· by lzfeifei was closed on Dec 5, 2023

请问如果单纯使用zeroth-order向前优化少量batch（只要体现出一定的优化效果）的话要怎么实现

Status: Closed (completed).

#158 In yuanzhoulvpi2017/zero_nlp;

· by CharonsPluto was closed on Oct 23, 2023

2 comments

训练出错

Status: Closed (completed).

#154 In yuanzhoulvpi2017/zero_nlp;

· by loki1017 was closed on Apr 16, 2025

求助：chatglm2 lora训练error：RuntimeError: Expected is_sm80 to be true, but got false.

Status: Closed (completed).

#152 In yuanzhoulvpi2017/zero_nlp;

· by thirttyyy was closed on Aug 21, 2023

2 comments

有大佬遇到使用这个Lora训练，loss 不收敛的情况嘛

chatglm2

Status: Closed (completed).

#146 In yuanzhoulvpi2017/zero_nlp;

· by DuBaiSheng was closed on Aug 15, 2023

4 comments