Class: ClipEmbedding
Hierarchy
-
↳
ClipEmbedding
Constructors
constructor
• new ClipEmbedding(): ClipEmbedding
Returns
Inherited from
MultiModalEmbedding.constructor
Properties
embedBatchSize
• embedBatchSize: number = DEFAULT_EMBED_BATCH_SIZE
Inherited from
MultiModalEmbedding.embedBatchSize
Defined in
packages/core/src/embeddings/types.ts:8
modelType
• modelType: ClipEmbeddingModelType = ClipEmbeddingModelType.XENOVA_CLIP_VIT_BASE_PATCH16
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:11
processor
• Private processor: any
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:15
textModel
• Private textModel: any
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:17
tokenizer
• Private tokenizer: any
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:14
visionModel
• Private visionModel: any
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:16
Methods
getImageEmbedding
▸ getImageEmbedding(image): Promise<number[]>
Parameters
| Name | Type |
|---|---|
image | ImageType |
Returns
Promise<number[]>
Overrides
MultiModalEmbedding.getImageEmbedding
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:61
getImageEmbeddings
▸ getImageEmbeddings(images): Promise<number[][]>
Parameters
| Name | Type |
|---|---|
images | ImageType[] |
Returns
Promise<number[][]>
Inherited from
MultiModalEmbedding.getImageEmbeddings
Defined in
packages/core/src/embeddings/MultiModalEmbedding.ts:11
getProcessor
▸ getProcessor(): Promise<any>
Returns
Promise<any>
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:27
getQueryEmbedding
▸ getQueryEmbedding(query): Promise<number[]>
Parameters
| Name | Type |
|---|---|
query | string |
Returns
Promise<number[]>
Overrides
MultiModalEmbedding.getQueryEmbedding
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:76
getTextEmbedding
▸ getTextEmbedding(text): Promise<number[]>
Parameters
| Name | Type |
|---|---|
text | string |
Returns
Promise<number[]>
Overrides
MultiModalEmbedding.getTextEmbedding
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:68
getTextEmbeddings
▸ getTextEmbeddings(texts): Promise<number[][]>
Optionally override this method to retrieve multiple embeddings in a single request
Parameters
| Name | Type |
|---|---|
texts | string[] |
Returns
Promise<number[][]>
Inherited from
MultiModalEmbedding.getTextEmbeddings
Defined in
packages/core/src/embeddings/types.ts:25
getTextEmbeddingsBatch
▸ getTextEmbeddingsBatch(texts, options?): Promise<number[][]>
Get embeddings for a batch of texts
Parameters
| Name | Type |
|---|---|
texts | string[] |
options? | Object |
options.logProgress? | boolean |
Returns
Promise<number[][]>
Inherited from
MultiModalEmbedding.getTextEmbeddingsBatch
Defined in
packages/core/src/embeddings/types.ts:41
getTextModel
▸ getTextModel(): Promise<any>
Returns
Promise<any>
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:48
getTokenizer
▸ getTokenizer(): Promise<any>
Returns
Promise<any>
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:19
getVisionModel
▸ getVisionModel(): Promise<any>
Returns
Promise<any>
Defined in
packages/core/src/embeddings/ClipEmbedding.ts:35
similarity
▸ similarity(embedding1, embedding2, mode?): number
Parameters
| Name | Type | Default value |
|---|---|---|
embedding1 | number[] | undefined |
embedding2 | number[] | undefined |
mode | SimilarityType | SimilarityType.DEFAULT |
Returns
number
Inherited from
MultiModalEmbedding.similarity
Defined in
packages/core/src/embeddings/types.ts:10
transform
▸ transform(nodes, _options?): Promise<BaseNode<Metadata>[]>
Parameters
| Name | Type |
|---|---|
nodes | BaseNode<Metadata>[] |
_options? | any |