mirror of
https://github.com/KevinMidboe/immich.git
synced 2025-12-08 20:29:05 +00:00
feat(ml)!: customizable ML settings (#3891)
* consolidated endpoints, added live configuration * added ml settings to server * added settings dashboard * updated deps, fixed typos * simplified modelconfig updated tests * Added ml setting accordion for admin page updated tests * merge `clipText` and `clipVision` * added face distance setting clarified setting * add clip mode in request, dropdown for face models * polished ml settings updated descriptions * update clip field on error * removed unused import * add description for image classification threshold * pin safetensors for arm wheel updated poetry lock * moved dto * set model type only in ml repository * revert form-data package install use fetch instead of axios * added slotted description with link updated facial recognition description clarified effect of disabling tasks * validation before model load * removed unnecessary getconfig call * added migration * updated api updated api updated api --------- Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
This commit is contained in:
@@ -115,6 +115,7 @@ describe(FacialRecognitionService.name, () => {
|
||||
personMock = newPersonRepositoryMock();
|
||||
searchMock = newSearchRepositoryMock();
|
||||
storageMock = newStorageRepositoryMock();
|
||||
configMock = newSystemConfigRepositoryMock();
|
||||
|
||||
mediaMock.crop.mockResolvedValue(croppedFace);
|
||||
|
||||
@@ -179,9 +180,18 @@ describe(FacialRecognitionService.name, () => {
|
||||
machineLearningMock.detectFaces.mockResolvedValue([]);
|
||||
assetMock.getByIds.mockResolvedValue([assetStub.image]);
|
||||
await sut.handleRecognizeFaces({ id: assetStub.image.id });
|
||||
expect(machineLearningMock.detectFaces).toHaveBeenCalledWith('http://immich-machine-learning:3003', {
|
||||
imagePath: assetStub.image.resizePath,
|
||||
});
|
||||
expect(machineLearningMock.detectFaces).toHaveBeenCalledWith(
|
||||
'http://immich-machine-learning:3003',
|
||||
{
|
||||
imagePath: assetStub.image.resizePath,
|
||||
},
|
||||
{
|
||||
enabled: true,
|
||||
maxDistance: 0.6,
|
||||
minScore: 0.7,
|
||||
modelName: 'buffalo_l',
|
||||
},
|
||||
);
|
||||
expect(faceMock.create).not.toHaveBeenCalled();
|
||||
expect(jobMock.queue).not.toHaveBeenCalled();
|
||||
});
|
||||
|
||||
@@ -32,7 +32,7 @@ export class FacialRecognitionService {
|
||||
|
||||
async handleQueueRecognizeFaces({ force }: IBaseJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognitionEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognition.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -59,7 +59,7 @@ export class FacialRecognitionService {
|
||||
|
||||
async handleRecognizeFaces({ id }: IEntityJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognitionEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognition.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -68,7 +68,11 @@ export class FacialRecognitionService {
|
||||
return false;
|
||||
}
|
||||
|
||||
const faces = await this.machineLearning.detectFaces(machineLearning.url, { imagePath: asset.resizePath });
|
||||
const faces = await this.machineLearning.detectFaces(
|
||||
machineLearning.url,
|
||||
{ imagePath: asset.resizePath },
|
||||
machineLearning.facialRecognition,
|
||||
);
|
||||
|
||||
this.logger.debug(`${faces.length} faces detected in ${asset.resizePath}`);
|
||||
this.logger.verbose(faces.map((face) => ({ ...face, embedding: `float[${face.embedding.length}]` })));
|
||||
@@ -80,7 +84,7 @@ export class FacialRecognitionService {
|
||||
|
||||
// try to find a matching face and link to the associated person
|
||||
// The closer to 0, the better the match. Range is from 0 to 2
|
||||
if (faceSearchResult.total && faceSearchResult.distances[0] < 0.6) {
|
||||
if (faceSearchResult.total && faceSearchResult.distances[0] <= machineLearning.facialRecognition.maxDistance) {
|
||||
this.logger.verbose(`Match face with distance ${faceSearchResult.distances[0]}`);
|
||||
personId = faceSearchResult.items[0].personId;
|
||||
}
|
||||
@@ -115,7 +119,7 @@ export class FacialRecognitionService {
|
||||
|
||||
async handleGenerateFaceThumbnail(data: IFaceThumbnailJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognitionEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.facialRecognition.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
|
||||
@@ -86,6 +86,7 @@ export interface ISearchRepository {
|
||||
deleteAssets(ids: string[]): Promise<void>;
|
||||
deleteFaces(ids: string[]): Promise<void>;
|
||||
deleteAllFaces(): Promise<number>;
|
||||
updateCLIPField(num_dim: number): Promise<void>;
|
||||
|
||||
searchAlbums(query: string, filters: SearchFilter): Promise<SearchResult<AlbumEntity>>;
|
||||
searchAssets(query: string, filters: SearchFilter): Promise<SearchResult<AssetEntity>>;
|
||||
|
||||
@@ -121,15 +121,18 @@ export class SearchService {
|
||||
await this.configCore.requireFeature(FeatureFlag.SEARCH);
|
||||
|
||||
const query = dto.q || dto.query || '*';
|
||||
const hasClip = machineLearning.enabled && machineLearning.clipEncodeEnabled;
|
||||
const hasClip = machineLearning.enabled && machineLearning.clip.enabled;
|
||||
const strategy = dto.clip && hasClip ? SearchStrategy.CLIP : SearchStrategy.TEXT;
|
||||
const filters = { userId: authUser.id, ...dto };
|
||||
|
||||
let assets: SearchResult<AssetEntity>;
|
||||
switch (strategy) {
|
||||
case SearchStrategy.CLIP:
|
||||
const clip = await this.machineLearning.encodeText(machineLearning.url, query);
|
||||
assets = await this.searchRepository.vectorSearch(clip, filters);
|
||||
const {
|
||||
machineLearning: { clip },
|
||||
} = await this.configCore.getConfig();
|
||||
const embedding = await this.machineLearning.encodeText(machineLearning.url, { text: query }, clip);
|
||||
assets = await this.searchRepository.vectorSearch(embedding, filters);
|
||||
break;
|
||||
case SearchStrategy.TEXT:
|
||||
default:
|
||||
|
||||
1
server/src/domain/smart-info/dto/index.ts
Normal file
1
server/src/domain/smart-info/dto/index.ts
Normal file
@@ -0,0 +1 @@
|
||||
export * from './model-config.dto';
|
||||
50
server/src/domain/smart-info/dto/model-config.dto.ts
Normal file
50
server/src/domain/smart-info/dto/model-config.dto.ts
Normal file
@@ -0,0 +1,50 @@
|
||||
import { ApiProperty } from '@nestjs/swagger';
|
||||
import { Type } from 'class-transformer';
|
||||
import { IsBoolean, IsEnum, IsNotEmpty, IsNumber, IsOptional, IsString, Max, Min } from 'class-validator';
|
||||
import { CLIPMode, ModelType } from '../machine-learning.interface';
|
||||
|
||||
export class ModelConfig {
|
||||
@IsBoolean()
|
||||
enabled!: boolean;
|
||||
|
||||
@IsString()
|
||||
@IsNotEmpty()
|
||||
modelName!: string;
|
||||
|
||||
@IsEnum(ModelType)
|
||||
@IsOptional()
|
||||
@ApiProperty({ enumName: 'ModelType', enum: ModelType })
|
||||
modelType?: ModelType;
|
||||
}
|
||||
|
||||
export class ClassificationConfig extends ModelConfig {
|
||||
@IsNumber()
|
||||
@Min(0)
|
||||
@Max(1)
|
||||
@Type(() => Number)
|
||||
@ApiProperty({ type: 'integer' })
|
||||
minScore!: number;
|
||||
}
|
||||
|
||||
export class CLIPConfig extends ModelConfig {
|
||||
@IsEnum(CLIPMode)
|
||||
@IsOptional()
|
||||
@ApiProperty({ enumName: 'CLIPMode', enum: CLIPMode })
|
||||
mode?: CLIPMode;
|
||||
}
|
||||
|
||||
export class RecognitionConfig extends ModelConfig {
|
||||
@IsNumber()
|
||||
@Min(0)
|
||||
@Max(1)
|
||||
@Type(() => Number)
|
||||
@ApiProperty({ type: 'integer' })
|
||||
minScore!: number;
|
||||
|
||||
@IsNumber()
|
||||
@Min(0)
|
||||
@Max(2)
|
||||
@Type(() => Number)
|
||||
@ApiProperty({ type: 'integer' })
|
||||
maxDistance!: number;
|
||||
}
|
||||
@@ -1,3 +1,4 @@
|
||||
export * from './dto';
|
||||
export * from './machine-learning.interface';
|
||||
export * from './smart-info.repository';
|
||||
export * from './smart-info.service';
|
||||
|
||||
@@ -1,9 +1,15 @@
|
||||
import { ClassificationConfig, CLIPConfig, RecognitionConfig } from './dto';
|
||||
|
||||
export const IMachineLearningRepository = 'IMachineLearningRepository';
|
||||
|
||||
export interface MachineLearningInput {
|
||||
export interface VisionModelInput {
|
||||
imagePath: string;
|
||||
}
|
||||
|
||||
export interface TextModelInput {
|
||||
text: string;
|
||||
}
|
||||
|
||||
export interface BoundingBox {
|
||||
x1: number;
|
||||
y1: number;
|
||||
@@ -19,9 +25,20 @@ export interface DetectFaceResult {
|
||||
embedding: number[];
|
||||
}
|
||||
|
||||
export interface IMachineLearningRepository {
|
||||
classifyImage(url: string, input: MachineLearningInput): Promise<string[]>;
|
||||
encodeImage(url: string, input: MachineLearningInput): Promise<number[]>;
|
||||
encodeText(url: string, input: string): Promise<number[]>;
|
||||
detectFaces(url: string, input: MachineLearningInput): Promise<DetectFaceResult[]>;
|
||||
export enum ModelType {
|
||||
IMAGE_CLASSIFICATION = 'image-classification',
|
||||
FACIAL_RECOGNITION = 'facial-recognition',
|
||||
CLIP = 'clip',
|
||||
}
|
||||
|
||||
export enum CLIPMode {
|
||||
VISION = 'vision',
|
||||
TEXT = 'text',
|
||||
}
|
||||
|
||||
export interface IMachineLearningRepository {
|
||||
classifyImage(url: string, input: VisionModelInput, config: ClassificationConfig): Promise<string[]>;
|
||||
encodeImage(url: string, input: VisionModelInput, config: CLIPConfig): Promise<number[]>;
|
||||
encodeText(url: string, input: TextModelInput, config: CLIPConfig): Promise<number[]>;
|
||||
detectFaces(url: string, input: VisionModelInput, config: RecognitionConfig): Promise<DetectFaceResult[]>;
|
||||
}
|
||||
|
||||
@@ -84,9 +84,13 @@ describe(SmartInfoService.name, () => {
|
||||
|
||||
await sut.handleClassifyImage({ id: asset.id });
|
||||
|
||||
expect(machineMock.classifyImage).toHaveBeenCalledWith('http://immich-machine-learning:3003', {
|
||||
imagePath: 'path/to/resize.ext',
|
||||
});
|
||||
expect(machineMock.classifyImage).toHaveBeenCalledWith(
|
||||
'http://immich-machine-learning:3003',
|
||||
{
|
||||
imagePath: 'path/to/resize.ext',
|
||||
},
|
||||
{ enabled: true, minScore: 0.9, modelName: 'microsoft/resnet-50' },
|
||||
);
|
||||
expect(smartMock.upsert).toHaveBeenCalledWith({
|
||||
assetId: 'asset-1',
|
||||
tags: ['tag1', 'tag2', 'tag3'],
|
||||
@@ -141,13 +145,16 @@ describe(SmartInfoService.name, () => {
|
||||
});
|
||||
|
||||
it('should save the returned objects', async () => {
|
||||
smartMock.upsert.mockResolvedValue();
|
||||
machineMock.encodeImage.mockResolvedValue([0.01, 0.02, 0.03]);
|
||||
|
||||
await sut.handleEncodeClip({ id: asset.id });
|
||||
|
||||
expect(machineMock.encodeImage).toHaveBeenCalledWith('http://immich-machine-learning:3003', {
|
||||
imagePath: 'path/to/resize.ext',
|
||||
});
|
||||
expect(machineMock.encodeImage).toHaveBeenCalledWith(
|
||||
'http://immich-machine-learning:3003',
|
||||
{ imagePath: 'path/to/resize.ext' },
|
||||
{ enabled: true, modelName: 'ViT-B-32::openai' },
|
||||
);
|
||||
expect(smartMock.upsert).toHaveBeenCalledWith({
|
||||
assetId: 'asset-1',
|
||||
clipEmbedding: [0.01, 0.02, 0.03],
|
||||
|
||||
@@ -22,7 +22,7 @@ export class SmartInfoService {
|
||||
|
||||
async handleQueueObjectTagging({ force }: IBaseJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.tagImageEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.classification.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -43,7 +43,7 @@ export class SmartInfoService {
|
||||
|
||||
async handleClassifyImage({ id }: IEntityJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.tagImageEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.classification.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -52,7 +52,11 @@ export class SmartInfoService {
|
||||
return false;
|
||||
}
|
||||
|
||||
const tags = await this.machineLearning.classifyImage(machineLearning.url, { imagePath: asset.resizePath });
|
||||
const tags = await this.machineLearning.classifyImage(
|
||||
machineLearning.url,
|
||||
{ imagePath: asset.resizePath },
|
||||
machineLearning.classification,
|
||||
);
|
||||
await this.repository.upsert({ assetId: asset.id, tags });
|
||||
|
||||
return true;
|
||||
@@ -60,7 +64,7 @@ export class SmartInfoService {
|
||||
|
||||
async handleQueueEncodeClip({ force }: IBaseJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.clipEncodeEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.clip.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -81,7 +85,7 @@ export class SmartInfoService {
|
||||
|
||||
async handleEncodeClip({ id }: IEntityJob) {
|
||||
const { machineLearning } = await this.configCore.getConfig();
|
||||
if (!machineLearning.enabled || !machineLearning.clipEncodeEnabled) {
|
||||
if (!machineLearning.enabled || !machineLearning.clip.enabled) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@@ -90,7 +94,12 @@ export class SmartInfoService {
|
||||
return false;
|
||||
}
|
||||
|
||||
const clipEmbedding = await this.machineLearning.encodeImage(machineLearning.url, { imagePath: asset.resizePath });
|
||||
const clipEmbedding = await this.machineLearning.encodeImage(
|
||||
machineLearning.url,
|
||||
{ imagePath: asset.resizePath },
|
||||
machineLearning.clip,
|
||||
);
|
||||
|
||||
await this.repository.upsert({ assetId: asset.id, clipEmbedding: clipEmbedding });
|
||||
|
||||
return true;
|
||||
|
||||
@@ -1,4 +1,6 @@
|
||||
import { IsBoolean, IsUrl, ValidateIf } from 'class-validator';
|
||||
import { ClassificationConfig, CLIPConfig, RecognitionConfig } from '@app/domain';
|
||||
import { Type } from 'class-transformer';
|
||||
import { IsBoolean, IsObject, IsUrl, ValidateIf, ValidateNested } from 'class-validator';
|
||||
|
||||
export class SystemConfigMachineLearningDto {
|
||||
@IsBoolean()
|
||||
@@ -8,12 +10,18 @@ export class SystemConfigMachineLearningDto {
|
||||
@ValidateIf((dto) => dto.enabled)
|
||||
url!: string;
|
||||
|
||||
@IsBoolean()
|
||||
clipEncodeEnabled!: boolean;
|
||||
@Type(() => ClassificationConfig)
|
||||
@ValidateNested()
|
||||
@IsObject()
|
||||
classification!: ClassificationConfig;
|
||||
|
||||
@IsBoolean()
|
||||
facialRecognitionEnabled!: boolean;
|
||||
@Type(() => CLIPConfig)
|
||||
@ValidateNested()
|
||||
@IsObject()
|
||||
clip!: CLIPConfig;
|
||||
|
||||
@IsBoolean()
|
||||
tagImageEnabled!: boolean;
|
||||
@Type(() => RecognitionConfig)
|
||||
@ValidateNested()
|
||||
@IsObject()
|
||||
facialRecognition!: RecognitionConfig;
|
||||
}
|
||||
|
||||
@@ -47,12 +47,25 @@ export const defaults = Object.freeze<SystemConfig>({
|
||||
[QueueName.THUMBNAIL_GENERATION]: { concurrency: 5 },
|
||||
[QueueName.VIDEO_CONVERSION]: { concurrency: 1 },
|
||||
},
|
||||
|
||||
machineLearning: {
|
||||
enabled: process.env.IMMICH_MACHINE_LEARNING_ENABLED !== 'false',
|
||||
url: process.env.IMMICH_MACHINE_LEARNING_URL || 'http://immich-machine-learning:3003',
|
||||
facialRecognitionEnabled: true,
|
||||
tagImageEnabled: true,
|
||||
clipEncodeEnabled: true,
|
||||
classification: {
|
||||
enabled: true,
|
||||
modelName: 'microsoft/resnet-50',
|
||||
minScore: 0.9,
|
||||
},
|
||||
clip: {
|
||||
enabled: true,
|
||||
modelName: 'ViT-B-32::openai',
|
||||
},
|
||||
facialRecognition: {
|
||||
enabled: true,
|
||||
modelName: 'buffalo_l',
|
||||
minScore: 0.7,
|
||||
maxDistance: 0.6,
|
||||
},
|
||||
},
|
||||
oauth: {
|
||||
enabled: false,
|
||||
@@ -143,9 +156,9 @@ export class SystemConfigCore {
|
||||
const mlEnabled = config.machineLearning.enabled;
|
||||
|
||||
return {
|
||||
[FeatureFlag.CLIP_ENCODE]: mlEnabled && config.machineLearning.clipEncodeEnabled,
|
||||
[FeatureFlag.FACIAL_RECOGNITION]: mlEnabled && config.machineLearning.facialRecognitionEnabled,
|
||||
[FeatureFlag.TAG_IMAGE]: mlEnabled && config.machineLearning.tagImageEnabled,
|
||||
[FeatureFlag.CLIP_ENCODE]: mlEnabled && config.machineLearning.clip.enabled,
|
||||
[FeatureFlag.FACIAL_RECOGNITION]: mlEnabled && config.machineLearning.facialRecognition.enabled,
|
||||
[FeatureFlag.TAG_IMAGE]: mlEnabled && config.machineLearning.classification.enabled,
|
||||
[FeatureFlag.SIDECAR]: true,
|
||||
[FeatureFlag.SEARCH]: process.env.TYPESENSE_ENABLED !== 'false',
|
||||
|
||||
@@ -230,7 +243,7 @@ export class SystemConfigCore {
|
||||
_.set(config, key, value);
|
||||
}
|
||||
|
||||
return _.defaultsDeep(config, defaults) as SystemConfig;
|
||||
return plainToClass(SystemConfigDto, _.defaultsDeep(config, defaults));
|
||||
}
|
||||
|
||||
private async loadFromFile(filepath: string, force = false) {
|
||||
|
||||
@@ -49,9 +49,21 @@ const updatedConfig = Object.freeze<SystemConfig>({
|
||||
machineLearning: {
|
||||
enabled: true,
|
||||
url: 'http://immich-machine-learning:3003',
|
||||
facialRecognitionEnabled: true,
|
||||
tagImageEnabled: true,
|
||||
clipEncodeEnabled: true,
|
||||
classification: {
|
||||
enabled: true,
|
||||
modelName: 'microsoft/resnet-50',
|
||||
minScore: 0.9,
|
||||
},
|
||||
clip: {
|
||||
enabled: true,
|
||||
modelName: 'ViT-B-32::openai',
|
||||
},
|
||||
facialRecognition: {
|
||||
enabled: true,
|
||||
modelName: 'buffalo_l',
|
||||
minScore: 0.7,
|
||||
maxDistance: 0.6,
|
||||
},
|
||||
},
|
||||
oauth: {
|
||||
autoLaunch: true,
|
||||
|
||||
Reference in New Issue
Block a user