Comments (3)
Morizeyao
commented on May 18, 2024
从经验上看，数据量越大需要训练的epoch越少。
数据如果比较小的话，可以训练数十个epoch。
因为代码里包含了每个epoch保存模型的功能，所以可以对于每个节点分别进行生成，观察并比较生成的效果。
如有1G以上的数据，训练一个epoch其实就可以观察到比较可观的生成效果了。当然，对大数据来说，训练的越久越好。
from gpt2-chinese.
ZhaoyueSun
commented on May 18, 2024
非常感谢！
from gpt2-chinese.
fengzuo97
commented on May 18, 2024
试验5个epoch基本可以收敛，英伟达训练megatron gpt2也是5个epoch收敛
from gpt2-chinese.
Related Issues (20)
- 小说的模型开放了吗？
- 我的语料文本很多，能不能分开多个train.json HOT 3
- 能不能用gpt3再创造个中文ai,gpt3比gpt2好很多
- 语料库700M，一跑就报This script is extremely slow，请教700M语料大约需要多大的配置
- Mark
- Mark
- GPU9GB能跑起来吗请问
- 是否支持多机多卡分布式训练？ HOT 2
- 如何修改訓練結構?
- finetune需要的语料量+output上限 HOT 2
- 求助 HOT 2
- 请问如何实现问答任务 HOT 2
- 语料预处理
- 下载的对联模型使用时为什么生成的是乱码？ HOT 1
- 你好，很期待和您沟通，请查收gmail 邮件。
- 请问GPT2-Chinese 参数量有多大呀，跟原版gpt2参数量一样吗？
- BertTokenizer的__init__报NotImplementedError HOT 3
- Using bos_token, but it is not set yet. Using eos_token, but it is not set yet. HOT 1
- train. py encountered an error during runtime. Value Error: invalid literal for int() with base 10: '[SEP]' How to solve it? HOT 1
- 求助求助 HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from gpt2-chinese.