nv_ingest_api.internal.schemas.transform package#
Submodules#
nv_ingest_api.internal.schemas.transform.transform_image_caption_schema module#
- pydantic model nv_ingest_api.internal.schemas.transform.transform_image_caption_schema.ImageCaptionExtractionSchema[source]#
Bases:
BaseModel
Show JSON schema
{ "title": "ImageCaptionExtractionSchema", "type": "object", "properties": { "api_key": { "default": "", "title": "Api Key", "type": "string" }, "endpoint_url": { "default": "https://integrate.api.nvidia.com/v1/chat/completions", "title": "Endpoint Url", "type": "string" }, "prompt": { "default": "Caption the content of this image:", "title": "Prompt", "type": "string" }, "model_name": { "default": "nvidia/llama-3.1-nemotron-nano-vl-8b-v1", "title": "Model Name", "type": "string" }, "raise_on_failure": { "default": false, "title": "Raise On Failure", "type": "boolean" } }, "additionalProperties": false }
- Config:
extra: str = forbid
- Fields:
- Validators:
_coerce_api_key_none
»api_key
_coerce_none_to_empty
»all fields
- field api_key: str = ''#
- Validated by:
_coerce_api_key_none
_coerce_none_to_empty
- field endpoint_url: str = 'https://integrate.api.nvidia.com/v1/chat/completions'#
- Validated by:
_coerce_none_to_empty
- field model_name: str = 'nvidia/llama-3.1-nemotron-nano-vl-8b-v1'#
- Validated by:
_coerce_none_to_empty
- field prompt: str = 'Caption the content of this image:'#
- Validated by:
_coerce_none_to_empty
- field raise_on_failure: bool = False#
- Validated by:
_coerce_none_to_empty
nv_ingest_api.internal.schemas.transform.transform_image_filter_schema module#
- pydantic model nv_ingest_api.internal.schemas.transform.transform_image_filter_schema.ImageFilterSchema[source]#
Bases:
BaseModel
Show JSON schema
{ "title": "ImageFilterSchema", "type": "object", "properties": { "raise_on_failure": { "default": false, "title": "Raise On Failure", "type": "boolean" }, "cpu_only": { "default": false, "title": "Cpu Only", "type": "boolean" } }, "additionalProperties": false }
- Config:
extra: str = forbid
- Fields:
- field cpu_only: Annotated[bool, Strict(strict=True)] = False#
- Constraints:
strict = True
- field raise_on_failure: Annotated[bool, Strict(strict=True)] = False#
- Constraints:
strict = True
nv_ingest_api.internal.schemas.transform.transform_text_embedding_schema module#
- pydantic model nv_ingest_api.internal.schemas.transform.transform_text_embedding_schema.TextEmbeddingSchema[source]#
Bases:
BaseModel
Show JSON schema
{ "title": "TextEmbeddingSchema", "type": "object", "properties": { "api_key": { "default": "", "title": "Api Key", "type": "string" }, "batch_size": { "default": 4, "title": "Batch Size", "type": "integer" }, "embedding_model": { "default": "nvidia/llama-3.2-nv-embedqa-1b-v2", "title": "Embedding Model", "type": "string" }, "embedding_nim_endpoint": { "default": "http://embedding:8000/v1", "title": "Embedding Nim Endpoint", "type": "string" }, "encoding_format": { "default": "float", "title": "Encoding Format", "type": "string" }, "httpx_log_level": { "$ref": "#/$defs/LogLevel", "default": "WARNING" }, "input_type": { "default": "passage", "title": "Input Type", "type": "string" }, "raise_on_failure": { "default": false, "title": "Raise On Failure", "type": "boolean" }, "truncate": { "default": "END", "title": "Truncate", "type": "string" }, "text_elements_modality": { "default": "text", "title": "Text Elements Modality", "type": "string" }, "image_elements_modality": { "default": "text", "title": "Image Elements Modality", "type": "string" }, "structured_elements_modality": { "default": "text", "title": "Structured Elements Modality", "type": "string" }, "audio_elements_modality": { "default": "text", "title": "Audio Elements Modality", "type": "string" } }, "$defs": { "LogLevel": { "enum": [ "DEFAULT", "DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL" ], "title": "LogLevel", "type": "string" } }, "additionalProperties": false }
- Config:
extra: str = forbid
- Fields:
- Validators:
_coerce_api_key_none
»api_key
_coerce_none_to_empty
»all fields
- field api_key: str = ''#
- Validated by:
_coerce_api_key_none
_coerce_none_to_empty
- field audio_elements_modality: str = 'text'#
- Validated by:
_coerce_none_to_empty
- field batch_size: int = 4#
- Validated by:
_coerce_none_to_empty
- field embedding_model: str = 'nvidia/llama-3.2-nv-embedqa-1b-v2'#
- Validated by:
_coerce_none_to_empty
- field embedding_nim_endpoint: str = 'http://embedding:8000/v1'#
- Validated by:
_coerce_none_to_empty
- field encoding_format: str = 'float'#
- Validated by:
_coerce_none_to_empty
- field image_elements_modality: str = 'text'#
- Validated by:
_coerce_none_to_empty
- field input_type: str = 'passage'#
- Validated by:
_coerce_none_to_empty
- field raise_on_failure: bool = False#
- Validated by:
_coerce_none_to_empty
- field structured_elements_modality: str = 'text'#
- Validated by:
_coerce_none_to_empty
- field text_elements_modality: str = 'text'#
- Validated by:
_coerce_none_to_empty
- field truncate: str = 'END'#
- Validated by:
_coerce_none_to_empty
nv_ingest_api.internal.schemas.transform.transform_text_splitter_schema module#
- pydantic model nv_ingest_api.internal.schemas.transform.transform_text_splitter_schema.TextSplitterSchema[source]#
Bases:
BaseModel
Show JSON schema
{ "title": "TextSplitterSchema", "type": "object", "properties": { "tokenizer": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "title": "Tokenizer" }, "chunk_size": { "default": 1024, "exclusiveMinimum": 0, "title": "Chunk Size", "type": "integer" }, "chunk_overlap": { "default": 150, "minimum": 0, "title": "Chunk Overlap", "type": "integer" }, "raise_on_failure": { "default": false, "title": "Raise On Failure", "type": "boolean" } }, "additionalProperties": false }
- Config:
extra: str = forbid
- Fields:
- Validators:
- field chunk_overlap: int = 150#
- Constraints:
ge = 0
- Validated by:
- field chunk_size: int = 1024#
- Constraints:
gt = 0
- field raise_on_failure: bool = False#
- field tokenizer: str | None = None#
- validator check_chunk_overlap » chunk_overlap[source]#