SwinV23D

pydantic model vision_architectures.nets.swinv2_3d.SwinV23DPatchMergingConfig

Bases: CustomBaseModel

Show JSON schema

{
   "title": "SwinV23DPatchMergingConfig",
   "type": "object",
   "properties": {
      "out_dim_ratio": {
         "title": "Out Dim Ratio",
         "type": "integer"
      },
      "merge_window_size": {
         "maxItems": 3,
         "minItems": 3,
         "prefixItems": [
            {
               "type": "integer"
            },
            {
               "type": "integer"
            },
            {
               "type": "integer"
            }
         ],
         "title": "Merge Window Size",
         "type": "array"
      }
   },
   "required": [
      "out_dim_ratio",
      "merge_window_size"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• merge_window_size (tuple[int, int, int])

• out_dim_ratio (int)

Validators:

• validate_before » all fields

field out_dim_ratio: int [Required]

Validated by:

• validate

• validate_before

field merge_window_size: tuple[int, int, int] [Required]

Validated by:

• validate

• validate_before

validator validate_before  »  all fields

Base class method for validating data before creating the model.

pydantic model vision_architectures.nets.swinv2_3d.SwinV23DPatchSplittingConfig

Bases: CustomBaseModel

Show JSON schema

{
   "title": "SwinV23DPatchSplittingConfig",
   "type": "object",
   "properties": {
      "out_dim_ratio": {
         "title": "Out Dim Ratio",
         "type": "integer"
      },
      "final_window_size": {
         "maxItems": 3,
         "minItems": 3,
         "prefixItems": [
            {
               "type": "integer"
            },
            {
               "type": "integer"
            },
            {
               "type": "integer"
            }
         ],
         "title": "Final Window Size",
         "type": "array"
      }
   },
   "required": [
      "out_dim_ratio",
      "final_window_size"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• final_window_size (tuple[int, int, int])

• out_dim_ratio (int)

Validators:

• validate_before » all fields

field out_dim_ratio: int [Required]

Validated by:

• validate

• validate_before

field final_window_size: tuple[int, int, int] [Required]

Validated by:

• validate

• validate_before

validator validate_before  »  all fields

Base class method for validating data before creating the model.

pydantic model vision_architectures.nets.swinv2_3d.SwinV23DStageConfig

Bases: Attention3DWithMLPConfig

Show JSON schema

{
   "title": "SwinV23DStageConfig",
   "type": "object",
   "properties": {
      "dim": {
         "default": 0,
         "description": "dim at which attention is performed",
         "title": "Dim",
         "type": "integer"
      },
      "mlp_ratio": {
         "default": 4,
         "title": "Mlp Ratio",
         "type": "integer"
      },
      "activation": {
         "default": "gelu",
         "title": "Activation",
         "type": "string"
      },
      "mlp_drop_prob": {
         "default": 0.0,
         "title": "Mlp Drop Prob",
         "type": "number"
      },
      "num_heads": {
         "description": "Number of query heads",
         "title": "Num Heads",
         "type": "integer"
      },
      "ratio_q_to_kv_heads": {
         "default": 1,
         "title": "Ratio Q To Kv Heads",
         "type": "integer"
      },
      "logit_scale_learnable": {
         "default": false,
         "title": "Logit Scale Learnable",
         "type": "boolean"
      },
      "attn_drop_prob": {
         "default": 0.0,
         "title": "Attn Drop Prob",
         "type": "number"
      },
      "proj_drop_prob": {
         "default": 0.0,
         "title": "Proj Drop Prob",
         "type": "number"
      },
      "max_attention_batch_size": {
         "default": -1,
         "description": "Runs attention by splitting the inputs into chunks of this size. 0 means no chunking. Useful for large inputs during inference.",
         "title": "Max Attention Batch Size",
         "type": "integer"
      },
      "norm_location": {
         "default": "post",
         "enum": [
            "pre",
            "post"
         ],
         "title": "Norm Location",
         "type": "string"
      },
      "layer_norm_eps": {
         "default": 1e-06,
         "title": "Layer Norm Eps",
         "type": "number"
      },
      "depth": {
         "title": "Depth",
         "type": "integer"
      },
      "window_size": {
         "maxItems": 3,
         "minItems": 3,
         "prefixItems": [
            {
               "type": "integer"
            },
            {
               "type": "integer"
            },
            {
               "type": "integer"
            }
         ],
         "title": "Window Size",
         "type": "array"
      },
      "use_relative_position_bias": {
         "default": true,
         "title": "Use Relative Position Bias",
         "type": "boolean"
      },
      "patch_merging": {
         "anyOf": [
            {
               "$ref": "#/$defs/SwinV23DPatchMergingConfig"
            },
            {
               "type": "null"
            }
         ],
         "default": null
      },
      "patch_splitting": {
         "anyOf": [
            {
               "$ref": "#/$defs/SwinV23DPatchSplittingConfig"
            },
            {
               "type": "null"
            }
         ],
         "default": null
      },
      "in_dim": {
         "anyOf": [
            {
               "type": "integer"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "title": "In Dim"
      },
      "out_dim": {
         "anyOf": [
            {
               "type": "integer"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "title": "Out Dim"
      }
   },
   "$defs": {
      "SwinV23DPatchMergingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "merge_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Merge Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "merge_window_size"
         ],
         "title": "SwinV23DPatchMergingConfig",
         "type": "object"
      },
      "SwinV23DPatchSplittingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "final_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Final Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "final_window_size"
         ],
         "title": "SwinV23DPatchSplittingConfig",
         "type": "object"
      }
   },
   "required": [
      "num_heads",
      "depth",
      "window_size"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• depth (int)

• dim (int)

• in_dim (int | None)

• logit_scale_learnable (bool)

• out_dim (int | None)

• patch_merging (vision_architectures.nets.swinv2_3d.SwinV23DPatchMergingConfig | None)

• patch_splitting (vision_architectures.nets.swinv2_3d.SwinV23DPatchSplittingConfig | None)

• use_relative_position_bias (bool)

• window_size (tuple[int, int, int])

Validators:

field depth: int [Required]

Validated by:

• validate

• validate_before

field window_size: tuple[int, int, int] [Required]

Validated by:

• validate

• validate_before

field use_relative_position_bias: bool = True

Validated by:

• validate

• validate_before

field patch_merging: SwinV23DPatchMergingConfig | None = None

Validated by:

• validate

• validate_before

field patch_splitting: SwinV23DPatchSplittingConfig | None = None

Validated by:

• validate

• validate_before

field in_dim: int | None = None

Validated by:

• validate

• validate_before

field dim: int = 0

dim at which attention is performed

Validated by:

• validate

• validate_before

field out_dim: int | None = None

Validated by:

• validate

• validate_before

field logit_scale_learnable: bool = False

Validated by:

• validate

• validate_before

property spatial_compression_ratio

get_out_patch_size(in_patch_size)

get_in_patch_size(out_patch_size)

pydantic model vision_architectures.nets.swinv2_3d.SwinV23DDecoderConfig

Bases: CustomBaseModel

Show JSON schema

{
   "title": "SwinV23DDecoderConfig",
   "type": "object",
   "properties": {
      "dim": {
         "title": "Dim",
         "type": "integer"
      },
      "stages": {
         "items": {
            "$ref": "#/$defs/SwinV23DStageConfig"
         },
         "title": "Stages",
         "type": "array"
      },
      "drop_prob": {
         "default": 0.0,
         "title": "Drop Prob",
         "type": "number"
      },
      "embed_spacing_info": {
         "default": false,
         "title": "Embed Spacing Info",
         "type": "boolean"
      }
   },
   "$defs": {
      "SwinV23DPatchMergingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "merge_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Merge Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "merge_window_size"
         ],
         "title": "SwinV23DPatchMergingConfig",
         "type": "object"
      },
      "SwinV23DPatchSplittingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "final_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Final Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "final_window_size"
         ],
         "title": "SwinV23DPatchSplittingConfig",
         "type": "object"
      },
      "SwinV23DStageConfig": {
         "properties": {
            "dim": {
               "default": 0,
               "description": "dim at which attention is performed",
               "title": "Dim",
               "type": "integer"
            },
            "mlp_ratio": {
               "default": 4,
               "title": "Mlp Ratio",
               "type": "integer"
            },
            "activation": {
               "default": "gelu",
               "title": "Activation",
               "type": "string"
            },
            "mlp_drop_prob": {
               "default": 0.0,
               "title": "Mlp Drop Prob",
               "type": "number"
            },
            "num_heads": {
               "description": "Number of query heads",
               "title": "Num Heads",
               "type": "integer"
            },
            "ratio_q_to_kv_heads": {
               "default": 1,
               "title": "Ratio Q To Kv Heads",
               "type": "integer"
            },
            "logit_scale_learnable": {
               "default": false,
               "title": "Logit Scale Learnable",
               "type": "boolean"
            },
            "attn_drop_prob": {
               "default": 0.0,
               "title": "Attn Drop Prob",
               "type": "number"
            },
            "proj_drop_prob": {
               "default": 0.0,
               "title": "Proj Drop Prob",
               "type": "number"
            },
            "max_attention_batch_size": {
               "default": -1,
               "description": "Runs attention by splitting the inputs into chunks of this size. 0 means no chunking. Useful for large inputs during inference.",
               "title": "Max Attention Batch Size",
               "type": "integer"
            },
            "norm_location": {
               "default": "post",
               "enum": [
                  "pre",
                  "post"
               ],
               "title": "Norm Location",
               "type": "string"
            },
            "layer_norm_eps": {
               "default": 1e-06,
               "title": "Layer Norm Eps",
               "type": "number"
            },
            "depth": {
               "title": "Depth",
               "type": "integer"
            },
            "window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Window Size",
               "type": "array"
            },
            "use_relative_position_bias": {
               "default": true,
               "title": "Use Relative Position Bias",
               "type": "boolean"
            },
            "patch_merging": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchMergingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "patch_splitting": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchSplittingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "in_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "In Dim"
            },
            "out_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "Out Dim"
            }
         },
         "required": [
            "num_heads",
            "depth",
            "window_size"
         ],
         "title": "SwinV23DStageConfig",
         "type": "object"
      }
   },
   "required": [
      "dim",
      "stages"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• dim (int)

• drop_prob (float)

• embed_spacing_info (bool)

• stages (list[vision_architectures.nets.swinv2_3d.SwinV23DStageConfig])

Validators:

• validate » all fields

field dim: int [Required]

Validated by:

• validate

• validate_before

field stages: list[SwinV23DStageConfig] [Required]

Validated by:

• validate

• validate_before

field drop_prob: float = 0.0

Validated by:

• validate

• validate_before

field embed_spacing_info: bool = False

Validated by:

• validate

• validate_before

populate()

validator validate  »  all fields

Base method for validating the model after creation.

pydantic model vision_architectures.nets.swinv2_3d.SwinV23DConfig

Bases: SwinV23DDecoderConfig

Show JSON schema

{
   "title": "SwinV23DConfig",
   "type": "object",
   "properties": {
      "dim": {
         "title": "Dim",
         "type": "integer"
      },
      "stages": {
         "items": {
            "$ref": "#/$defs/SwinV23DStageConfig"
         },
         "title": "Stages",
         "type": "array"
      },
      "drop_prob": {
         "default": 0.0,
         "title": "Drop Prob",
         "type": "number"
      },
      "embed_spacing_info": {
         "default": false,
         "title": "Embed Spacing Info",
         "type": "boolean"
      },
      "in_channels": {
         "title": "In Channels",
         "type": "integer"
      },
      "patch_size": {
         "maxItems": 3,
         "minItems": 3,
         "prefixItems": [
            {
               "type": "integer"
            },
            {
               "type": "integer"
            },
            {
               "type": "integer"
            }
         ],
         "title": "Patch Size",
         "type": "array"
      },
      "image_size": {
         "anyOf": [
            {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "type": "array"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "required for learnable absolute position embeddings",
         "title": "Image Size"
      },
      "use_absolute_position_embeddings": {
         "default": true,
         "title": "Use Absolute Position Embeddings",
         "type": "boolean"
      },
      "learnable_absolute_position_embeddings": {
         "default": false,
         "title": "Learnable Absolute Position Embeddings",
         "type": "boolean"
      }
   },
   "$defs": {
      "SwinV23DPatchMergingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "merge_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Merge Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "merge_window_size"
         ],
         "title": "SwinV23DPatchMergingConfig",
         "type": "object"
      },
      "SwinV23DPatchSplittingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "final_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Final Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "final_window_size"
         ],
         "title": "SwinV23DPatchSplittingConfig",
         "type": "object"
      },
      "SwinV23DStageConfig": {
         "properties": {
            "dim": {
               "default": 0,
               "description": "dim at which attention is performed",
               "title": "Dim",
               "type": "integer"
            },
            "mlp_ratio": {
               "default": 4,
               "title": "Mlp Ratio",
               "type": "integer"
            },
            "activation": {
               "default": "gelu",
               "title": "Activation",
               "type": "string"
            },
            "mlp_drop_prob": {
               "default": 0.0,
               "title": "Mlp Drop Prob",
               "type": "number"
            },
            "num_heads": {
               "description": "Number of query heads",
               "title": "Num Heads",
               "type": "integer"
            },
            "ratio_q_to_kv_heads": {
               "default": 1,
               "title": "Ratio Q To Kv Heads",
               "type": "integer"
            },
            "logit_scale_learnable": {
               "default": false,
               "title": "Logit Scale Learnable",
               "type": "boolean"
            },
            "attn_drop_prob": {
               "default": 0.0,
               "title": "Attn Drop Prob",
               "type": "number"
            },
            "proj_drop_prob": {
               "default": 0.0,
               "title": "Proj Drop Prob",
               "type": "number"
            },
            "max_attention_batch_size": {
               "default": -1,
               "description": "Runs attention by splitting the inputs into chunks of this size. 0 means no chunking. Useful for large inputs during inference.",
               "title": "Max Attention Batch Size",
               "type": "integer"
            },
            "norm_location": {
               "default": "post",
               "enum": [
                  "pre",
                  "post"
               ],
               "title": "Norm Location",
               "type": "string"
            },
            "layer_norm_eps": {
               "default": 1e-06,
               "title": "Layer Norm Eps",
               "type": "number"
            },
            "depth": {
               "title": "Depth",
               "type": "integer"
            },
            "window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Window Size",
               "type": "array"
            },
            "use_relative_position_bias": {
               "default": true,
               "title": "Use Relative Position Bias",
               "type": "boolean"
            },
            "patch_merging": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchMergingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "patch_splitting": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchSplittingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "in_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "In Dim"
            },
            "out_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "Out Dim"
            }
         },
         "required": [
            "num_heads",
            "depth",
            "window_size"
         ],
         "title": "SwinV23DStageConfig",
         "type": "object"
      }
   },
   "required": [
      "dim",
      "stages",
      "in_channels",
      "patch_size"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• image_size (tuple[int, int, int] | None)

• in_channels (int)

• learnable_absolute_position_embeddings (bool)

• patch_size (tuple[int, int, int])

• use_absolute_position_embeddings (bool)

Validators:

• validate » all fields

field in_channels: int [Required]

Validated by:

• validate

• validate_before

field patch_size: tuple[int, int, int] [Required]

Validated by:

• validate

• validate_before

field image_size: tuple[int, int, int] | None = None

required for learnable absolute position embeddings

Validated by:

• validate

• validate_before

field use_absolute_position_embeddings: bool = True

Validated by:

• validate

• validate_before

field learnable_absolute_position_embeddings: bool = False

Validated by:

• validate

• validate_before

validator validate  »  all fields

Base method for validating the model after creation.

pydantic model vision_architectures.nets.swinv2_3d.Swin3DMIMConfig

Bases: SwinV23DConfig

Show JSON schema

{
   "title": "Swin3DMIMConfig",
   "type": "object",
   "properties": {
      "dim": {
         "title": "Dim",
         "type": "integer"
      },
      "stages": {
         "items": {
            "$ref": "#/$defs/SwinV23DStageConfig"
         },
         "title": "Stages",
         "type": "array"
      },
      "drop_prob": {
         "default": 0.0,
         "title": "Drop Prob",
         "type": "number"
      },
      "embed_spacing_info": {
         "default": false,
         "title": "Embed Spacing Info",
         "type": "boolean"
      },
      "in_channels": {
         "title": "In Channels",
         "type": "integer"
      },
      "patch_size": {
         "maxItems": 3,
         "minItems": 3,
         "prefixItems": [
            {
               "type": "integer"
            },
            {
               "type": "integer"
            },
            {
               "type": "integer"
            }
         ],
         "title": "Patch Size",
         "type": "array"
      },
      "image_size": {
         "anyOf": [
            {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "type": "array"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "required for learnable absolute position embeddings",
         "title": "Image Size"
      },
      "use_absolute_position_embeddings": {
         "default": true,
         "title": "Use Absolute Position Embeddings",
         "type": "boolean"
      },
      "learnable_absolute_position_embeddings": {
         "default": false,
         "title": "Learnable Absolute Position Embeddings",
         "type": "boolean"
      },
      "mim": {
         "additionalProperties": true,
         "title": "Mim",
         "type": "object"
      }
   },
   "$defs": {
      "SwinV23DPatchMergingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "merge_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Merge Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "merge_window_size"
         ],
         "title": "SwinV23DPatchMergingConfig",
         "type": "object"
      },
      "SwinV23DPatchSplittingConfig": {
         "properties": {
            "out_dim_ratio": {
               "title": "Out Dim Ratio",
               "type": "integer"
            },
            "final_window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Final Window Size",
               "type": "array"
            }
         },
         "required": [
            "out_dim_ratio",
            "final_window_size"
         ],
         "title": "SwinV23DPatchSplittingConfig",
         "type": "object"
      },
      "SwinV23DStageConfig": {
         "properties": {
            "dim": {
               "default": 0,
               "description": "dim at which attention is performed",
               "title": "Dim",
               "type": "integer"
            },
            "mlp_ratio": {
               "default": 4,
               "title": "Mlp Ratio",
               "type": "integer"
            },
            "activation": {
               "default": "gelu",
               "title": "Activation",
               "type": "string"
            },
            "mlp_drop_prob": {
               "default": 0.0,
               "title": "Mlp Drop Prob",
               "type": "number"
            },
            "num_heads": {
               "description": "Number of query heads",
               "title": "Num Heads",
               "type": "integer"
            },
            "ratio_q_to_kv_heads": {
               "default": 1,
               "title": "Ratio Q To Kv Heads",
               "type": "integer"
            },
            "logit_scale_learnable": {
               "default": false,
               "title": "Logit Scale Learnable",
               "type": "boolean"
            },
            "attn_drop_prob": {
               "default": 0.0,
               "title": "Attn Drop Prob",
               "type": "number"
            },
            "proj_drop_prob": {
               "default": 0.0,
               "title": "Proj Drop Prob",
               "type": "number"
            },
            "max_attention_batch_size": {
               "default": -1,
               "description": "Runs attention by splitting the inputs into chunks of this size. 0 means no chunking. Useful for large inputs during inference.",
               "title": "Max Attention Batch Size",
               "type": "integer"
            },
            "norm_location": {
               "default": "post",
               "enum": [
                  "pre",
                  "post"
               ],
               "title": "Norm Location",
               "type": "string"
            },
            "layer_norm_eps": {
               "default": 1e-06,
               "title": "Layer Norm Eps",
               "type": "number"
            },
            "depth": {
               "title": "Depth",
               "type": "integer"
            },
            "window_size": {
               "maxItems": 3,
               "minItems": 3,
               "prefixItems": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  },
                  {
                     "type": "integer"
                  }
               ],
               "title": "Window Size",
               "type": "array"
            },
            "use_relative_position_bias": {
               "default": true,
               "title": "Use Relative Position Bias",
               "type": "boolean"
            },
            "patch_merging": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchMergingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "patch_splitting": {
               "anyOf": [
                  {
                     "$ref": "#/$defs/SwinV23DPatchSplittingConfig"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null
            },
            "in_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "In Dim"
            },
            "out_dim": {
               "anyOf": [
                  {
                     "type": "integer"
                  },
                  {
                     "type": "null"
                  }
               ],
               "default": null,
               "title": "Out Dim"
            }
         },
         "required": [
            "num_heads",
            "depth",
            "window_size"
         ],
         "title": "SwinV23DStageConfig",
         "type": "object"
      }
   },
   "required": [
      "dim",
      "stages",
      "in_channels",
      "patch_size",
      "mim"
   ]
}

Config:

• arbitrary_types_allowed: bool = True

• extra: str = ignore

• validate_default: bool = True

• validate_assignment: bool = True

• validate_return: bool = True

Fields:

• mim (dict)

Validators:

field mim: dict [Required]

Validated by:

• validate

• validate_before

class vision_architectures.nets.swinv2_3d.SwinV23DLayerLogitScale(num_heads)

Bases: Module

__init__(num_heads)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward()

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DLayer(config={}, checkpointing_level=0, **kwargs)

Bases: Module

__init__(config={}, checkpointing_level=0, **kwargs)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DBlock(config={}, checkpointing_level=0, **kwargs)

Bases: Module

__init__(config={}, checkpointing_level=0, **kwargs)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(hidden_states, channels_first=True, return_intermediates=False)

Please note that the layer_outputs returned if return_intermediates=True will always be in channels_last format

class vision_architectures.nets.swinv2_3d.SwinV23DPatchMerging(merge_window_size, in_dim, out_dim, checkpointing_level=0)

Bases: Module

__init__(merge_window_size, in_dim, out_dim, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DPatchSplitting(final_window_size, in_dim, out_dim, checkpointing_level=0)

Bases: Module

__init__(final_window_size, in_dim, out_dim, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DStage(stage_config, checkpointing_level=0)

Bases: Module

__init__(stage_config, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DEncoder(config, checkpointing_level=0)

Bases: Module, PyTorchModelHubMixin

__init__(config, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DDecoder(config, checkpointing_level=0)

Bases: Module, PyTorchModelHubMixin

__init__(config, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(*args, **kwargs)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DModel(config, checkpointing_level=0)

Bases: Module, PyTorchModelHubMixin

__init__(config, checkpointing_level=0)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(pixel_values, spacings=None, crop_offsets=None, channels_first=True, return_intermediates=False)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DReconstructionDecoder(config)

Bases: Module

__init__(config)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(encodings)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DMIM(swin_config, decoder_config, mim_config)

Bases: Module

__init__(swin_config, decoder_config, mim_config)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

mask_image(pixel_values)

class vision_architectures.nets.swinv2_3d.SwinV23DSimMIM(*args, **kwargs)

Bases: SwinV23DMIM, PyTorchModelHubMixin

__init__(swin_config, decoder_config, mim_config)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

static loss_fn(pred, target, reduction='mean')

forward(pixel_values, spacings=None)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class vision_architectures.nets.swinv2_3d.SwinV23DVAEMIM(*args, **kwargs)

Bases: SwinV23DMIM, PyTorchModelHubMixin

__init__(swin_config, decoder_config, mim_config)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

get_beta()

reparameterize(mu, logvar)

static reconstruction_loss_fn(pred, target, loss_type='l2', reduction='mean')

static kl_divergence_loss_fn(mu, logvar)

forward(pixel_values, spacings=None, reconstruction_loss_type='l2')

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.