用 ChatGPT 采用自有数据集训练问答机器人

type

Post

status

Published

date

Apr 17, 2023

slug

summary

chatGPT

安装Anaconda

清华大学的开源镜像站速度比较快，下载地址：https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D

选择最新日期的可执行文件进行下载。

下载后双击进行安装，安装时可以选择非C盘路径，安装完成后打开系统环境变量设置并将以下路径新建到Path下

安装完成后保存。

打开命令提示符或者Powershell输入conda 测试是否安装成功

安装完成后，可以使用conda进行python等环境安装

📝环境准备

安装python

To create a new environment named "chat" with Python 3.8 installed, use the following command:


conda create -n chat python=3.8

To activate the "chat" environment, use the following command:


conda activate chat

配置jupyter环境

To install the ipykernel package in the "chat" environment, use the following command:


conda install -n chat ipykernel

This package provides the IPython kernel for Jupyter, which will allow you to use the "chat" environment as a kernel for Jupyter notebooks.

After installing the ipykernel package in the "chat" environment, use the following command to register a new kernel:


python -m ipykernel install --user --name chat --display-name "chat"

安装pytorch

To install PyTorch, use the following command:


pip install torch torchvision torchaudio --extra-index-url <https://download.pytorch.org/whl/cu116>

This command installs PyTorch with support for CUDA 11.1. If you are using a different version of CUDA, please refer to the PyTorch documentation for the appropriate installation command.