- Видео 19
- Просмотров 55 227
gravy
США
Добавлен 3 май 2016
I also want to be an interesting person, meeting interesting people, and doing interesting things
Introducing google's gemini to discord | building with gemini's api in node.js
gemini api documentation: ai.google.dev/docs
discord api platform: discord.com/developers/applications
github repository: github.com/clairelovesgravy/discord-bot-gemini
discord api platform: discord.com/developers/applications
github repository: github.com/clairelovesgravy/discord-bot-gemini
Просмотров: 651
Видео
A walkthrough of Openai's assistant api
Просмотров 426 месяцев назад
You can refer to the code here：github.com/clairelovesgravy/assistant-api-walkthrough
OpenAI leadership drama story line review | openai ceo transition | what happens to open ai?
Просмотров 207 месяцев назад
the first time i try to create a video with the help of ai tools, share with you!. news collection:me; script drawing: chatgpt; audio generationg: TTS github.com/coqui-ai/TTS; video caption generating: whisper ai github.com/openai/whisper; Video material synthesis：roop github.com/s0md3v/roop; video clip synthesis：me!
How to install and use roop on PC to swap face
Просмотров 3,6 тыс.9 месяцев назад
roop github link: github.com/s0md3v/roop try roop in cloud: replicate.com/okaris/roop installation note: enchanting-polonium-c95.notion.site/how-to-install-roop-on-pc-edf034a0c60148c4b0d0a7175de05c50?pvs=4
AI yourself | 一键换脸操作 ｜ midjourney换脸 ｜ roop换脸｜ 同时支持图片和视频
Просмотров 9089 месяцев назад
insight face swap邀请链接：discord.com/api/oauth2/authorize?client_id=1090660574196674713&permissions=274877945856&scope=bot roop github链接：github.com/s0md3v/roop roop replicate链接：replicate.com/okaris/roop
一键修复模糊老照片 ｜ 开源免费工具分享｜ 一鍵修復模糊老照片 ｜ 開源免費工具分享
Просмотров 69010 месяцев назад
codeformer github地址:github.com/sczhou/CodeFormer; gfpgan github 地址：github.com/TencentARC/GFPGAN; codeformer replicate地址:replicate.com/sczhou/codeformer; gfpgan replicate地址:replicate.com/tencentarc/gfpgan
怎样制作AI艺术写真｜colab在线训练Lora｜怎樣製作AI藝術寫真｜colab在線訓練Lora
Просмотров 72110 месяцев назад
暗香模型下载链接：civitai.com/models/94087 github项目地址链接：github.com/camenduru/stable-diffusion-webui-colab 素材仅做学习交流使用，担心版权问题，训练出来的Lora就不分享了。
揭秘大神們的配置技巧!.env文件,帶你從零入門Python/Node.js項目的環境變量管理｜ 把腳本里的API key 和 token藏起來 ｜ dotenv(.env)的配置和使用
Просмотров 11011 месяцев назад
起因是，我不小心把api key传到github里，然后open ai发邮件告诉我key泄露了。
Slack上也能有chatgpt!详解如何快速创建自己的专属AI聊天机器人 | 用langchain定制化聊天机器人｜ 从写代码开始
Просмотров 299Год назад
操作说明：enchanting-polonium-c95.notion.site/slack_bot-13fcff6bc03844cebe2bf4d515102b4b?pvs=4 github 项目地址：github.com/clairelovesgravy/slack_bot_demo
Photoshop ai 使用分享｜generative fill革新设计体验
Просмотров 300Год назад
PS官方下载地址：www.adobe.com/products/photoshop/generative-fill.html
忘掉SQL! 使用自然语言查询数据库 | 用Langchain连接chatgpt和数据库，实现人与数据库的无缝对话 | database search powered by langchain
Просмотров 1,5 тыс.Год назад
github项目地址：github.com/clairelovesgravy/langchain_database_search langchain地址：python.langchain.com/en/latest/index.html 数据库地址：data.sfgov.org/City-Management-and-Ethics/Employee-Compensation/88g8-5mnd data.sfgov.org/Housing-and-Buildings/Land-Use/us3s-fp9q
ml stable diffusion安装使用 | stable diffusion with core ml on apple silicon | 转化适配苹果神经网络的core ml模型
Просмотров 1,5 тыс.Год назад
项目地址：github.com/apple/ml-stable-diffusion annaconda官方地址：www.anaconda.com/ 0-1min57s:项目基本情况介绍； 1min58s-5min23s:安装annaconda,创建python环境 5min24s-7min41s:获取hugging face api token 7min42s-13min55s:将sd 模型转化成core ml格式 13min56s-21min03s:用新创建的core ml模型生成图片
利用Stable Diffusion打造AI电商模特图：操作分享
Просмотров 34 тыс.Год назад
stable diffusion 用的模型是chillout mix, 模型下载地址：civitai.com/models/6424/chilloutmix 模特prompt: positive: (8k, RAW photo, best quality, masterpiece:1.2), (realistic, photo-realistic:1.4), (depth of field), highres, absurdres, lora:taiwanDollLikeness_v10:0.4, ulzzang-6500-v1.1, 1girl, solo, nude, ((naked body)), realistic skin texture, skin pores, cute, smile, laughing, (grin), (brown hair , (eyelashes...
Claude人工智能助手,这么强大?!测试体验+完整教程!
Просмотров 2,3 тыс.Год назад
0.51-6.19:Claude功能测试； 6.20-7.48: Claude使用说明； slack地址：slack.com/ 或者直接APP store里搜
macos本地部署whisper ai,一键生成中英文视频字幕。
Просмотров 1,1 тыс.Год назад
github地址：github.com/openai/whisper python地址：www.python.org/downloads/ homebrew地址：brew.sh/ pytorch地址：pytorch.org/ 有问题可以加入discord讨论，discord.gg/cqMFcHTg 不能保证有求必应，但有空看到了会回复，看缘分。
build a google extension with chatgpt from scratch
Просмотров 60Год назад
build a google extension with chatgpt from scratch
MACOS本地部署stable diffusion + Controlnet插件安装
Просмотров 6 тыс.Год назад
MACOS本地部署stable diffusion Controlnet插件安装
电脑性能不够玩不了stable diffusion?没关系，带你白嫖google colab,同时支持controlnet,LoRA多种玩法哦！
Просмотров 1,1 тыс.Год назад
电脑性能不够玩不了stable diffusion?没关系，带你白嫖google colab,同时支持controlnet,LoRA多种玩法哦！
AI绘图哪家强！DALL.E, stable diffusion, midjourney大PK.
Просмотров 465Год назад
AI绘图哪家强！DALL.E, stable diffusion, midjourney大PK.
哈喽，想请教一下现在homebrew复制的链接到mac显示无法连接服务器，请问应该如何操作啊？
請問有辦法對不同Server (不同 IP 跟 Port )的資料庫收尋嗎?
博主小姐姐好，我的Mac在部署完之后，control net的插件位置给你的不一样呢。在你的controlnet下的model文件夹，在放入.pt模型前，只有一个txt文件。部署完之后，在webUI界面哦的文生图最下方，没有扩展插件“controlnet“，图像生成种子下面，直接就是脚本。请问该怎么办呢？
你怎么这么厉害！
一步一步按视频操作的，但是在安装ffmpeg这一步，whisper官方安装文件给的指令brew install ffmpeg，终端输入后却显示zsh：command not found：brew；明明第一步就已经安装过homebrew了呀，要怎么办，求助？感谢
你用的是mac吗？
Thank you for the tutorial, it was clear and concise!
需要
Thank you but can it send text more 2500 words in split as in discord on cannot send text bigger than 2000 words
I remember it has already support reply in multi times😂
use embeds if you want more then 2000
如果想把GTPs部署到SLACK上可行吗？
可行啊，承载平台的区别而已。gpts多了embedding, function-calling，这些通过api一样做啊
很好！我刚刚开始在SLACK上建设数字总部，希望能向你学习
The common PC is windows.
😂sry
第二次上传图片后无法换脸怎么回事
你得看终端报什么错
感谢分享😊
然后生成的mlpackage模型文件怎么用在webui或者ComfyUI里?
為什麼我安裝了openai但還是沒辦法import它呢?
这种基础的问题，我建议你问gemini,stackoverflow,chatgpt😂
非常感谢博主的分享，请教一个问题，为什么我用蒙版的时候，生成的图还是对蒙版只能的衣服颜色做了更改呀？？感觉蒙版不起作用，求解答，谢谢：）
事实上，我听不懂你的问题😂
Awesome tutorial! I read it carefully from start to finish and finally managed to complete my first Discord bot! To be honest, I was a total beginner without any knowledge at all. I had just downloaded VSCode. But by following the steps in the video one by one, I really did succeed in the end, which is amazing!😇😇😇
thank you! you give me some courage to keep it on😂
@@kailai7255 happy!looking forward to your latest tutorial videos!🥰
@@kailai7255 If possible, I would like to learn more in-depth about Discord bots, such as how to enable contextual memory for the bots to remember previous conversations, as well as questions about how to use the safety setting codes for the Gemini API.😵💫😵💫
@@nanakisahanakiyoshi1857 langchain helps you manage memory or you could use openai's thread in assistant api
@@kailai7255 thank you so much for the advice! I'll go look into it and give it a try to see how I can make use of it for my bot....😇
現在還有更新的coreml了嗎？
git仓库一直是同一个，我看到上个月还有更新
Also, if there is a photo of the outfit and a photo of another model, is there a way to have that model wear the outfit in the outfit photo?
In order to show the real wearing effect of the picture at close range, the clothing needs to be photographed wearing it. You can find an ordinary person to put on clothes and take pictures, or put the clothes on a mannequin to take pictures, and then use roop to change the face and replace it with the character of the model you need to specify.
Thank you for always. We are trying to generate only people, mannequins, backgrounds, and clothes without shadows. However, it is not easy to create just clothes. Mannequins appear, and people keep appearing. Is there any way to solve this?
acctually i dont know your intention. why generate clothes img with ai? in this video, the intention was to displaying real clothes in real people, not just clothes imgs.
如果表名列名不是标准的英文翻译，如果是拼音啥的就废了
额，应该不会有公司的数据这么构建吧。。。
生成的服装边缘有原图的颜色 怎么融合呢
inpaint 那邊可以設置邊界要從蒙版擴大幾個 pixel, 預設是 4，如果蒙版不精確，可以調大一點；另外新版本 inpaint 可以勾選 soft inpaint, control net 也多了 inpaint 模式，可以勾選全局融合，inpaint 的時候也讓他畫整張圖後再拼起來，會比只畫局部要好。
姐，我是看你4年前支持王局来看看你的。
Sorry, but I don't know who he is🙃
你好，麻烦你想问一下，roop 的 replicate 网址链接的上不去了 还有其他解法不？
你可以把roop装到本地
🎉
搞了一個晚上的colab，最後被妳一個影片給接上線解決了，妳真的是太棒了。
小姐姐厉害，点个赞👍
你这是采用默认的模型，我指定大模型之后好慢
把你要用的大模型预下载好，存到model文件夹里
我是这样，苹果M1 pro的，效果很差，需要很久@@kailai7255
Dona video for windows
很不错，跟着你的视频，训练了几个LORA了，对于我没有本地部署的来说，很方便
Could you show how to do with windows? maybe there is a difference. Can it be installed in the d drive instead of c or must everything be in the same drive? maybe that was my mistake?
I had installed on windows and finally get the Atribute Error NoneType object has no attribute Configure .
学到了，但是效果没有封面那张那么有细节
啊 水平太差，我就分享个框架😂
RuntimeError: "upsample_nearest2d_channels_last" not implemented for 'Half' 请问这个怎么解决呀
启动webui的时候，在命令行增加“--no-half”试试，出处在这里github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Command-Line-Arguments-and-Settings
@@kailai7255 感谢！
AttributeError: 'NoneType' object has no attribute 'shape' after I clicked the start.
can you tell me what is displayed in your terminal?
@@kailai7255 pasteboard. co/yb1zJAHOEClX.png
@@kailai7255 I fix this problem. My jpeg file name contains turkish latters such as " -ö,-ü ", so I looked a little and tried to figure out and I have just tried to rename the jpeg file with englisch characters. It worked. Thanks a lot.
老妹 能不能出个安装这个Stable Diffusion的教程？网上的都过期了。你现在出个这种教程，播放量肯定破50万。
我以前出过一期，即使迭代了，基本的部署安装都大同小异，看看github里的说明也就能装好了，新视频么等我有心情了再说吧😂
Server join 不了啊
我刚试了没失效啊😂你网络环境ok吗？discord能正常使用吗
包含NSFW内容的大模型因为包含太多与工作不相关的内容会使得效率很低下，这种模型如果你的商家不是做这种内衣和情色有点关系的电商的话，会难以实现效率因为大部分姿势穿在普通模特上都太奇怪了，而且在打光真实性方面还有待研究，但是流程是可以借鉴的，支持一下
教程简单实用，谢谢！
我下载了mac whisper是不是可以不用这个了？
对！
我现在就是要给大量的英文视频做双语字幕，可有好方法@@kailai7255
已經有兩個UI App可以用了，Diffusers and Mochi
是的，这个库本来就是面向os开发者的
controlnet最新版不能在Mac上正常使用了
讲解的非常棒，我很喜欢
請問手部要修改的話如何處理？
参考下这个项目，是一种解决思路github.com/jexom/sd-webui-depth-lib。当然你也可以线下自己摆拍，拍好了以后用controlnet提取姿势。
可以留个联系方式么 有没有交流群 想和你多学点
python不需要安装，系统本身自带python3
请问 如果关闭终端就无法打开了 ,要想再打开怎么运行终端呢
还是打开终端，cd切换项目目录，再运行一遍命令./webui. sh 还是运行一遍同样的命令（我已经很久没玩了，具体命令你再看下说明）
@@kailai7255 谢谢谢谢
你这个提示词是认真的吗，我都不跑NSFW 图
要生成的是模特啊！如果是穿衣服的模特，那要展示的衣服怎么办
🙂
Expecting value: line 1 column 1 (char 0) 总是这个错误,是怎么回事
安装完了,就是启动后界面总是报这个错误
可以一键贴提示词和参数，点生成下面的那个小箭头