feat(ml): export clip models to ONNX and host models on Hugging Face (#4700)

* export clip models

* export to hf

refactored export code

* export mclip, general refactoring

cleanup

* updated conda deps

* do transforms with pillow and numpy, add tokenization config to export, general refactoring

* moved conda dockerfile, re-added poetry

* minor fixes

* updated link

* updated tests

* removed `requirements.txt` from workflow

* fixed mimalloc path

* removed torchvision

* cleaner np typing

* review suggestions

* update default model name

* update test

server/src/domain/smart-info/smart-info.service.spec.ts

@@ -193,7 +193,7 @@ describe(SmartInfoService.name, () => {
       expect(machineMock.encodeImage).toHaveBeenCalledWith(
         'http://immich-machine-learning:3003',
         { imagePath: 'path/to/resize.ext' },
-        { enabled: true, modelName: 'ViT-B-32::openai' },
+        { enabled: true, modelName: 'ViT-B-32__openai' },
       );
       expect(smartMock.upsert).toHaveBeenCalledWith({
         assetId: 'asset-1',

server/src/domain/system-config/system-config.core.ts

@@ -67,7 +67,7 @@ export const defaults = Object.freeze<SystemConfig>({
     },
     clip: {
       enabled: true,
-      modelName: 'ViT-B-32::openai',
+      modelName: 'ViT-B-32__openai',
     },
     facialRecognition: {
       enabled: true,

server/src/domain/system-config/system-config.service.spec.ts

@@ -68,7 +68,7 @@ const updatedConfig = Object.freeze<SystemConfig>({
     },
     clip: {
       enabled: true,
-      modelName: 'ViT-B-32::openai',
+      modelName: 'ViT-B-32__openai',
     },
     facialRecognition: {
       enabled: true,