@huggingface/tasks 0.13.1-test → 0.13.1-test2

package/src/tasks/chat-completion/spec/output.json

@@ -0,0 +1,206 @@
+{
+	"$id": "/inference/schemas/chat-completion/output.json",
+	"$schema": "http://json-schema.org/draft-06/schema#",
+	"description": "Chat Completion Output.\n\nAuto-generated from TGI specs.\nFor more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.",
+	"title": "ChatCompletionOutput",
+	"type": "object",
+	"required": ["id", "created", "model", "system_fingerprint", "choices", "usage"],
+	"properties": {
+		"choices": {
+			"type": "array",
+			"items": {
+				"$ref": "#/$defs/ChatCompletionOutputComplete"
+			}
+		},
+		"created": {
+			"type": "integer",
+			"format": "int64",
+			"example": "1706270835",
+			"minimum": 0
+		},
+		"id": {
+			"type": "string"
+		},
+		"model": {
+			"type": "string",
+			"example": "mistralai/Mistral-7B-Instruct-v0.2"
+		},
+		"system_fingerprint": {
+			"type": "string"
+		},
+		"usage": {
+			"$ref": "#/$defs/ChatCompletionOutputUsage"
+		}
+	},
+	"$defs": {
+		"ChatCompletionOutputComplete": {
+			"type": "object",
+			"required": ["index", "message", "finish_reason"],
+			"properties": {
+				"finish_reason": {
+					"type": "string"
+				},
+				"index": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"logprobs": {
+					"allOf": [
+						{
+							"$ref": "#/$defs/ChatCompletionOutputLogprobs"
+						}
+					],
+					"nullable": true
+				},
+				"message": {
+					"$ref": "#/$defs/ChatCompletionOutputMessage"
+				}
+			},
+			"title": "ChatCompletionOutputComplete"
+		},
+		"ChatCompletionOutputLogprobs": {
+			"type": "object",
+			"required": ["content"],
+			"properties": {
+				"content": {
+					"type": "array",
+					"items": {
+						"$ref": "#/$defs/ChatCompletionOutputLogprob"
+					}
+				}
+			},
+			"title": "ChatCompletionOutputLogprobs"
+		},
+		"ChatCompletionOutputLogprob": {
+			"type": "object",
+			"required": ["token", "logprob", "top_logprobs"],
+			"properties": {
+				"logprob": {
+					"type": "number",
+					"format": "float"
+				},
+				"token": {
+					"type": "string"
+				},
+				"top_logprobs": {
+					"type": "array",
+					"items": {
+						"$ref": "#/$defs/ChatCompletionOutputTopLogprob"
+					}
+				}
+			},
+			"title": "ChatCompletionOutputLogprob"
+		},
+		"ChatCompletionOutputTopLogprob": {
+			"type": "object",
+			"required": ["token", "logprob"],
+			"properties": {
+				"logprob": {
+					"type": "number",
+					"format": "float"
+				},
+				"token": {
+					"type": "string"
+				}
+			},
+			"title": "ChatCompletionOutputTopLogprob"
+		},
+		"ChatCompletionOutputMessage": {
+			"oneOf": [
+				{
+					"$ref": "#/$defs/ChatCompletionOutputTextMessage"
+				},
+				{
+					"$ref": "#/$defs/ChatCompletionOutputToolCallMessage"
+				}
+			],
+			"title": "ChatCompletionOutputMessage"
+		},
+		"ChatCompletionOutputTextMessage": {
+			"type": "object",
+			"required": ["role", "content"],
+			"properties": {
+				"content": {
+					"type": "string",
+					"example": "My name is David and I"
+				},
+				"role": {
+					"type": "string",
+					"example": "user"
+				}
+			},
+			"title": "ChatCompletionOutputTextMessage"
+		},
+		"ChatCompletionOutputToolCallMessage": {
+			"type": "object",
+			"required": ["role", "tool_calls"],
+			"properties": {
+				"role": {
+					"type": "string",
+					"example": "assistant"
+				},
+				"tool_calls": {
+					"type": "array",
+					"items": {
+						"$ref": "#/$defs/ChatCompletionOutputToolCall"
+					}
+				}
+			},
+			"title": "ChatCompletionOutputToolCallMessage"
+		},
+		"ChatCompletionOutputToolCall": {
+			"type": "object",
+			"required": ["id", "type", "function"],
+			"properties": {
+				"function": {
+					"$ref": "#/$defs/ChatCompletionOutputFunctionDefinition"
+				},
+				"id": {
+					"type": "string"
+				},
+				"type": {
+					"type": "string"
+				}
+			},
+			"title": "ChatCompletionOutputToolCall"
+		},
+		"ChatCompletionOutputFunctionDefinition": {
+			"type": "object",
+			"required": ["name", "arguments"],
+			"properties": {
+				"arguments": {},
+				"description": {
+					"type": "string",
+					"nullable": true
+				},
+				"name": {
+					"type": "string"
+				}
+			},
+			"title": "ChatCompletionOutputFunctionDefinition"
+		},
+		"ChatCompletionOutputUsage": {
+			"type": "object",
+			"required": ["prompt_tokens", "completion_tokens", "total_tokens"],
+			"properties": {
+				"completion_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"prompt_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"total_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				}
+			},
+			"title": "ChatCompletionOutputUsage"
+		}
+	}
+}

package/src/tasks/chat-completion/spec/stream_output.json

@@ -0,0 +1,213 @@
+{
+	"$id": "/inference/schemas/chat-completion/stream_output.json",
+	"$schema": "http://json-schema.org/draft-06/schema#",
+	"description": "Chat Completion Stream Output.\n\nAuto-generated from TGI specs.\nFor more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.",
+	"title": "ChatCompletionStreamOutput",
+	"type": "object",
+	"required": ["id", "created", "model", "system_fingerprint", "choices"],
+	"properties": {
+		"choices": {
+			"type": "array",
+			"items": {
+				"$ref": "#/$defs/ChatCompletionStreamOutputChoice"
+			}
+		},
+		"created": {
+			"type": "integer",
+			"format": "int64",
+			"example": "1706270978",
+			"minimum": 0
+		},
+		"id": {
+			"type": "string"
+		},
+		"model": {
+			"type": "string",
+			"example": "mistralai/Mistral-7B-Instruct-v0.2"
+		},
+		"system_fingerprint": {
+			"type": "string"
+		},
+		"usage": {
+			"allOf": [
+				{
+					"$ref": "#/$defs/ChatCompletionStreamOutputUsage"
+				}
+			],
+			"nullable": true
+		}
+	},
+	"$defs": {
+		"ChatCompletionStreamOutputChoice": {
+			"type": "object",
+			"required": ["index", "delta"],
+			"properties": {
+				"delta": {
+					"$ref": "#/$defs/ChatCompletionStreamOutputDelta"
+				},
+				"finish_reason": {
+					"type": "string",
+					"nullable": true
+				},
+				"index": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"logprobs": {
+					"allOf": [
+						{
+							"$ref": "#/$defs/ChatCompletionStreamOutputLogprobs"
+						}
+					],
+					"nullable": true
+				}
+			},
+			"title": "ChatCompletionStreamOutputChoice"
+		},
+		"ChatCompletionStreamOutputDelta": {
+			"oneOf": [
+				{
+					"$ref": "#/$defs/ChatCompletionStreamOutputTextMessage"
+				},
+				{
+					"$ref": "#/$defs/ChatCompletionStreamOutputToolCallDelta"
+				}
+			],
+			"title": "ChatCompletionStreamOutputDelta"
+		},
+		"ChatCompletionStreamOutputTextMessage": {
+			"type": "object",
+			"required": ["role", "content"],
+			"properties": {
+				"content": {
+					"type": "string",
+					"example": "My name is David and I"
+				},
+				"role": {
+					"type": "string",
+					"example": "user"
+				}
+			},
+			"title": "ChatCompletionStreamOutputTextMessage"
+		},
+		"ChatCompletionStreamOutputToolCallDelta": {
+			"type": "object",
+			"required": ["role", "tool_calls"],
+			"properties": {
+				"role": {
+					"type": "string",
+					"example": "assistant"
+				},
+				"tool_calls": {
+					"$ref": "#/$defs/ChatCompletionStreamOutputDeltaToolCall"
+				}
+			},
+			"title": "ChatCompletionStreamOutputToolCallDelta"
+		},
+		"ChatCompletionStreamOutputDeltaToolCall": {
+			"type": "object",
+			"required": ["index", "id", "type", "function"],
+			"properties": {
+				"function": {
+					"$ref": "#/$defs/ChatCompletionStreamOutputFunction"
+				},
+				"id": {
+					"type": "string"
+				},
+				"index": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"type": {
+					"type": "string"
+				}
+			},
+			"title": "ChatCompletionStreamOutputDeltaToolCall"
+		},
+		"ChatCompletionStreamOutputFunction": {
+			"type": "object",
+			"required": ["arguments"],
+			"properties": {
+				"arguments": {
+					"type": "string"
+				},
+				"name": {
+					"type": "string",
+					"nullable": true
+				}
+			},
+			"title": "ChatCompletionStreamOutputFunction"
+		},
+		"ChatCompletionStreamOutputLogprobs": {
+			"type": "object",
+			"required": ["content"],
+			"properties": {
+				"content": {
+					"type": "array",
+					"items": {
+						"$ref": "#/$defs/ChatCompletionStreamOutputLogprob"
+					}
+				}
+			},
+			"title": "ChatCompletionStreamOutputLogprobs"
+		},
+		"ChatCompletionStreamOutputLogprob": {
+			"type": "object",
+			"required": ["token", "logprob", "top_logprobs"],
+			"properties": {
+				"logprob": {
+					"type": "number",
+					"format": "float"
+				},
+				"token": {
+					"type": "string"
+				},
+				"top_logprobs": {
+					"type": "array",
+					"items": {
+						"$ref": "#/$defs/ChatCompletionStreamOutputTopLogprob"
+					}
+				}
+			},
+			"title": "ChatCompletionStreamOutputLogprob"
+		},
+		"ChatCompletionStreamOutputTopLogprob": {
+			"type": "object",
+			"required": ["token", "logprob"],
+			"properties": {
+				"logprob": {
+					"type": "number",
+					"format": "float"
+				},
+				"token": {
+					"type": "string"
+				}
+			},
+			"title": "ChatCompletionStreamOutputTopLogprob"
+		},
+		"ChatCompletionStreamOutputUsage": {
+			"type": "object",
+			"required": ["prompt_tokens", "completion_tokens", "total_tokens"],
+			"properties": {
+				"completion_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"prompt_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				},
+				"total_tokens": {
+					"type": "integer",
+					"format": "int32",
+					"minimum": 0
+				}
+			},
+			"title": "ChatCompletionStreamOutputUsage"
+		}
+	}
+}

package/src/tasks/common-definitions.json

@@ -0,0 +1,100 @@
+{
+	"$id": "/inference/schemas/common-definitions.json",
+	"$schema": "http://json-schema.org/draft-06/schema#",
+	"description": "(Incomplete!) Common type definitions shared by several tasks",
+	"definitions": {
+		"ClassificationOutputTransform": {
+			"title": "ClassificationOutputTransform",
+			"type": "string",
+			"description": "The function to apply to the model outputs in order to retrieve the scores.",
+			"enum": ["sigmoid", "softmax", "none"]
+		},
+		"ClassificationOutput": {
+			"title": "ClassificationOutput",
+			"type": "object",
+			"properties": {
+				"label": {
+					"type": "string",
+					"description": "The predicted class label."
+				},
+				"score": {
+					"type": "number",
+					"description": "The corresponding probability."
+				}
+			},
+			"required": ["label", "score"]
+		},
+		"GenerationParameters": {
+			"title": "GenerationParameters",
+			"description": "Ad-hoc parametrization of the text generation process",
+			"type": "object",
+			"properties": {
+				"temperature": {
+					"type": "number",
+					"description": "The value used to modulate the next token probabilities."
+				},
+				"top_k": {
+					"type": "integer",
+					"description": "The number of highest probability vocabulary tokens to keep for top-k-filtering."
+				},
+				"top_p": {
+					"type": "number",
+					"description": "If set to float < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation."
+				},
+				"typical_p": {
+					"type": "number",
+					"description": " Local typicality measures how similar the conditional probability of predicting a target token next is to the expected conditional probability of predicting a random token next, given the partial text already generated. If set to float < 1, the smallest set of the most locally typical tokens with probabilities that add up to typical_p or higher are kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details."
+				},
+				"epsilon_cutoff": {
+					"type": "number",
+					"description": "If set to float strictly between 0 and 1, only tokens with a conditional probability greater than epsilon_cutoff will be sampled. In the paper, suggested values range from 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191) for more details."
+				},
+				"eta_cutoff": {
+					"type": "number",
+					"description": "Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to float strictly between 0 and 1, a token is only considered if it is greater than either eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model. See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191) for more details."
+				},
+				"max_length": {
+					"type": "integer",
+					"description": "The maximum length (in tokens) of the generated text, including the input."
+				},
+				"max_new_tokens": {
+					"type": "integer",
+					"description": "The maximum number of tokens to generate. Takes precedence over max_length."
+				},
+				"min_length": {
+					"type": "integer",
+					"description": "The minimum length (in tokens) of the generated text, including the input."
+				},
+				"min_new_tokens": {
+					"type": "integer",
+					"description": "The minimum number of tokens to generate. Takes precedence over min_length."
+				},
+				"do_sample": {
+					"type": "boolean",
+					"description": "Whether to use sampling instead of greedy decoding when generating new tokens."
+				},
+				"early_stopping": {
+					"type": ["boolean", "string"],
+					"description": "Controls the stopping condition for beam-based methods.",
+					"enum": ["never", true, false]
+				},
+				"num_beams": {
+					"type": "integer",
+					"description": "Number of beams to use for beam search."
+				},
+				"num_beam_groups": {
+					"type": "integer",
+					"description": "Number of groups to divide num_beams into in order to ensure diversity among different groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details."
+				},
+				"penalty_alpha": {
+					"type": "number",
+					"description": "The value balances the model confidence and the degeneration penalty in contrastive search decoding."
+				},
+				"use_cache": {
+					"type": "boolean",
+					"description": "Whether the model should use the past last key/values attentions to speed up decoding"
+				}
+			}
+		}
+	}
+}

package/src/tasks/depth-estimation/about.md

@@ -0,0 +1,45 @@
+## Use Cases
+
+Depth estimation models can be used to estimate the depth of different objects present in an image.
+
+### Estimation of Volumetric Information
+Depth estimation models are widely used to study volumetric formation of objects present inside an image. This is an important use case in the domain of computer graphics.
+
+### 3D Representation
+
+Depth estimation models can also be used to develop a 3D representation from a 2D image.
+
+## Depth Estimation Subtasks
+
+There are two depth estimation subtasks.
+
+- **Absolute depth estimation**: Absolute (or metric) depth estimation aims to provide exact depth measurements from the camera. Absolute depth estimation models output depth maps with real-world distances in meter or feet.
+
+- **Relative depth estimation**: Relative depth estimation aims to predict the depth order of objects or points in a scene without providing the precise measurements.
+
+## Inference
+
+With the `transformers` library, you can use the `depth-estimation` pipeline to infer with image classification models. You can initialize the pipeline with a model id from the Hub. If you do not provide a model id it will initialize with [Intel/dpt-large](https://huggingface.co/Intel/dpt-large) by default. When calling the pipeline you just need to specify a path, http link or an image loaded in PIL. Additionally, you can find a comprehensive list of various depth estimation models at [this link](https://huggingface.co/models?pipeline_tag=depth-estimation).
+
+```python
+from transformers import pipeline
+
+estimator = pipeline(task="depth-estimation", model="Intel/dpt-large")
+result = estimator(images="http://images.cocodataset.org/val2017/000000039769.jpg")
+result
+
+# {'predicted_depth': tensor([[[ 6.3199,  6.3629,  6.4148,  ..., 10.4104, 10.5109, 10.3847],
+#           [ 6.3850,  6.3615,  6.4166,  ..., 10.4540, 10.4384, 10.4554],
+#           [ 6.3519,  6.3176,  6.3575,  ..., 10.4247, 10.4618, 10.4257],
+#           ...,
+#           [22.3772, 22.4624, 22.4227,  ..., 22.5207, 22.5593, 22.5293],
+#           [22.5073, 22.5148, 22.5114,  ..., 22.6604, 22.6344, 22.5871],
+#           [22.5176, 22.5275, 22.5218,  ..., 22.6282, 22.6216, 22.6108]]]),
+#  'depth': <PIL.Image.Image image mode=L size=640x480 at 0x7F1A8BFE5D90>}
+
+# You can visualize the result just by calling `result["depth"]`.
+```
+
+## Useful Resources
+
+- [Monocular depth estimation task guide](https://huggingface.co/docs/transformers/tasks/monocular_depth_estimation)

package/src/tasks/depth-estimation/data.ts

@@ -0,0 +1,70 @@
+import type { TaskDataCustom } from "../index.js";
+
+const taskData: TaskDataCustom = {
+	datasets: [
+		{
+			description: "NYU Depth V2 Dataset: Video dataset containing both RGB and depth sensor data.",
+			id: "sayakpaul/nyu_depth_v2",
+		},
+		{
+			description: "Monocular depth estimation benchmark based without noise and errors.",
+			id: "depth-anything/DA-2K",
+		},
+	],
+	demo: {
+		inputs: [
+			{
+				filename: "depth-estimation-input.jpg",
+				type: "img",
+			},
+		],
+		outputs: [
+			{
+				filename: "depth-estimation-output.png",
+				type: "img",
+			},
+		],
+	},
+	metrics: [],
+	models: [
+		{
+			description: "Cutting-edge depth estimation model.",
+			id: "depth-anything/Depth-Anything-V2-Large",
+		},
+		{
+			description: "A strong monocular depth estimation model.",
+			id: "jingheya/lotus-depth-g-v1-0",
+		},
+		{
+			description: "A depth estimation model that predicts depth in videos.",
+			id: "tencent/DepthCrafter",
+		},
+		{
+			description: "A robust depth estimation model.",
+			id: "apple/DepthPro",
+		},
+	],
+	spaces: [
+		{
+			description: "An application that predicts the depth of an image and then reconstruct the 3D model as voxels.",
+			id: "radames/dpt-depth-estimation-3d-voxels",
+		},
+		{
+			description: "An application for bleeding-edge depth estimation.",
+			id: "akhaliq/depth-pro",
+		},
+		{
+			description: "An application on cutting-edge depth estimation in videos.",
+			id: "tencent/DepthCrafter",
+		},
+		{
+			description: "A human-centric depth estimation application.",
+			id: "facebook/sapiens-depth",
+		},
+	],
+	summary: "Depth estimation is the task of predicting depth of the objects present in an image.",
+	widgetModels: [""],
+	youtubeId: "",
+};
+
+export default taskData;

package/src/tasks/depth-estimation/inference.ts

@@ -0,0 +1,35 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+
+/**
+ * Inputs for Depth Estimation inference
+ */
+export interface DepthEstimationInput {
+	/**
+	 * The input image data
+	 */
+	inputs: unknown;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: { [key: string]: unknown };
+	[property: string]: unknown;
+}
+
+/**
+ * Outputs of inference for the Depth Estimation task
+ */
+export interface DepthEstimationOutput {
+	/**
+	 * The predicted depth as an image
+	 */
+	depth?: unknown;
+	/**
+	 * The predicted depth as a tensor
+	 */
+	predicted_depth?: unknown;
+	[property: string]: unknown;
+}

package/src/tasks/depth-estimation/spec/input.json

@@ -0,0 +1,25 @@
+{
+	"$id": "/inference/schemas/depth-estimation/input.json",
+	"$schema": "http://json-schema.org/draft-06/schema#",
+	"description": "Inputs for Depth Estimation inference",
+	"title": "DepthEstimationInput",
+	"type": "object",
+	"properties": {
+		"inputs": {
+			"description": "The input image data"
+		},
+		"parameters": {
+			"description": "Additional inference parameters",
+			"$ref": "#/$defs/DepthEstimationParameters"
+		}
+	},
+	"$defs": {
+		"DepthEstimationParameters": {
+			"title": "DepthEstimationParameters",
+			"description": "Additional inference parameters for Depth Estimation",
+			"type": "object",
+			"properties": {}
+		}
+	},
+	"required": ["inputs"]
+}