blib-la/lorado

LoRAdo is a UI that allows easy creation of LoRAs for stable diffussion

lorado's Issues

[BUG]: Error during install

Describe the bug
Error while npm install

Expected behavior
No error

Screenshots

C:\Users\new\LoRAdo>npm install
npm WARN deprecated @types/[email protected]: This is a stub types definition. unsplash-js provides its own type definitions, so you do not need this installed.
npm ERR! code 1
npm ERR! path C:\Users\new\LoRAdo\node_modules\@tensorflow\tfjs-node
npm ERR! command failed
npm ERR! command C:\Windows\system32\cmd.exe /d /s /c node scripts/install.js
npm ERR! CPU-windows-4.10.0.zip
npm ERR! https://storage.googleapis.com/tensorflow/libtensorflow/libtensorflow-cpu-windows-x86_64-2.9.1.zip
npm ERR! * Downloading libtensorflow
npm ERR!
npm ERR! * Building TensorFlow Node.js bindings
npm ERR! symlink ./lib/napi-v9 failed:  Error: Command failed: node scripts/deps-stage.js symlink ./lib/napi-v9
npm ERR!   * Symlink of lib\napi-v9\tensorflow.dll failed, creating a copy on disk.
npm ERR! node:internal/process/promises:288
npm ERR!             triggerUncaughtException(err, true /* fromPromise */);
npm ERR!             ^
npm ERR!
npm ERR! [Error: ENOENT: no such file or directory, copyfile 'C:\Users\new\LoRAdo\node_modules\@tensorflow\tfjs-node\deps\lib\tensorflow.dll' -> 'C:\Users\new\LoRAdo\node_modules\@tensorflow\tfjs-node\lib\napi-v9\tensorflow.dll'] {
npm ERR!   errno: -4058,
npm ERR!   code: 'ENOENT',
npm ERR!   syscall: 'copyfile',
npm ERR!   path: 'C:\\Users\\new\\LoRAdo\\node_modules\\@tensorflow\\tfjs-node\\deps\\lib\\tensorflow.dll',
npm ERR!   dest: 'C:\\Users\\new\\LoRAdo\\node_modules\\@tensorflow\\tfjs-node\\lib\\napi-v9\\tensorflow.dll'
npm ERR! }
npm ERR!
npm ERR! Node.js v18.17.0
npm ERR!
npm ERR!     at ChildProcess.exithandler (node:child_process:419:12)
npm ERR!     at ChildProcess.emit (node:events:514:28)
npm ERR!     at maybeClose (node:internal/child_process:1091:16)
npm ERR!     at ChildProcess._handle.onexit (node:internal/child_process:302:5) {
npm ERR!   code: 1,
npm ERR!   killed: false,
npm ERR!   signal: null,
npm ERR!   cmd: 'node scripts/deps-stage.js symlink ./lib/napi-v9'
npm ERR! }

npm ERR! A complete log of this run can be found in: C:\Users\new\AppData\Local\npm-cache\_logs\2023-08-15T12_22_27_806Z-debug-0.log

Versions (please complete the following information):

• Node.js: [e.g. v18.17]
• OS: [e.g. windows 10]
• LoRAdo [e.g. 0.1.0]

Additional context
Log file in attachment
2023-08-15T12_22_27_806Z-debug-0.log

Better Captioning

Is your feature request related to a problem?
FIrst I try for making my first lora, i don't any experience before and the result from dataset and config here it's amazing, but when i try write captioning for 1-6 images anythings fine but when 7 image or more, text box getting slower

Describe the solution you'd like
Maybe just turn off word checking (if any) and make this regular text box

Describe alternatives you've considered A clear and concise description of any alternative
Sometime, i still lazy to manual caption, so i go to a1111 or comfy to make img2txt for caption, and add stuff i want, maybe option auto captioning or manual captioning will be great.

Additional context Add any other context or screenshots about the feature request here.
I know it's lot work, i think we don't need clip interrogator like pharma did or a1111 just for describe image like kohya will be enough

Thank you

[BUG]: EXDEV: cross-device link not permitted

Describe the bug
I have LoRAdo installed on drive D: on Windows. I leave all default configurations, uploaded 5 images, added captions and clicked on "prepare". Then it takes some time, nothing is happening in the UI besides the "loading" animation. In the CLI I see the following error instantly:

[Error: EXDEV: cross-device link not permitted, rename 'C:\Users\NERDDI~1\AppData\Local\Temp\34cc4814dea94a20066614202' -> 'D:\dev\js\failfast\LoRAdo\training\n3rdd1sc0\dataset\n3rdd1sc0--0003.jpg'] {
  errno: -4037,
  code: 'EXDEV',
  syscall: 'rename',
  path: 'C:\\Users\\NERDDI~1\\AppData\\Local\\Temp\\34cc4814dea94a20066614202',
  dest: 'D:\\dev\\js\\failfast\\LoRAdo\\training\\n3rdd1sc0\\dataset\\n3rdd1sc0--0003.jpg'
}

Then after some more time, the "loading"-animation on the "prepare"-button is gone and I get this error in the CLI:

ypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:11413:11)
    at async invokeRequest (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\server\lib\server-ipc\invoke-request.js:17:12)
    at async invokeRender (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\server\lib\router-server.js:253:29)
    at async handleRequest (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\server\lib\router-server.js:446:24)
    at async requestHandler (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\server\lib\router-server.js:463:13)
    at async Server.<anonymous> (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\server\lib\start-server.js:117:13) {
  cause: HeadersTimeoutError: Headers Timeout Error
      at Timeout.onParserTimeout [as callback] (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\compiled\undici\index.js:1:62578)
      at Timeout.onTimeout [as _onTimeout] (D:\dev\js\failfast\LoRAdo\node_modules\next\dist\compiled\undici\index.js:2:269659)
      at listOnTimeout (node:internal/timers:569:17)
      at process.processTimers (node:internal/timers:512:7) {
    code: 'UND_ERR_HEADERS_TIMEOUT'
  }
}

The dataset is created, I also see the captions and the config.json. Just the images are not copied.

Expected behavior
I can install LoRAdo on any drive I want and clicking on "prepare" will actually copy the images that I uploaded into the generated dataset.

Versions (please complete the following information):

• Browser: Chrome 115.0.5790.170 (Official Build) (arm64)
• Node.js: v18.15.0
• OS: Windows 11
• LoRAdo: https://github.com/failfa-st/LoRAdo/releases/tag/v0.1.0-alpha.3

Support for SD 1.5

Feature Request:
It would be amazing if LoRAdo could support SD 1.5.

I'm not sure what changes would need to be made and which features wouldn't be supported, but it would let more people use the tool and make their training easier! A healthier life for everyone please!

Create actual releases that are checking if there are build errors

Is your feature request related to a problem? Please describe.
Currently, we only use git tags for a new release, but we need to do proper releases, that also run the build script, so we can be sure, that there are no build errors.

[BUG]: Many items cause slow rendering

Describe the bug

When I upload more that 10 images, the caption editing is noticably slower. At 20 images it is extremely laggy

Expected behavior

editing captions should be fast/immediate

Screenshots

Versions (please complete the following information):

• Browser: [Chrome 115.0.5790.171]
• Node.js: [v18.17.1]
• OS: [windows 11]
• LoRAdo [e.g. 0.1.0-alpha-3]

Additional context

as reported in #3

Define the Node versions that are actually working in README

Is your feature request related to a problem? Please describe.
It's annoying that during installation with npm install some node versions are not working, see #1. It would be nice if we would define the working Node-Versions in the README, so that people can check if they have a supported version and are not running into problems and have to ask us.

Describe the solution you'd like

• Test the supported versions of Node
• Write down the supported versions in the README, so that everyone can see them instantly

Describe alternatives you've considered

• Go on Discord to ask a question
• Search on GitHub issues to find the closed issue, that is actually not resolved, as I still don't know which versions are working