mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-09-09 11:24:53 +00:00
fmt
This commit is contained in:
OlivierDehaene
parent
8d7a0c1992
commit
a448acbfbe
@ -210,13 +210,62 @@
|
||||
},
|
||||
"components": {
|
||||
"schemas": {
|
||||
"BestOfSequence": {
|
||||
"type": "object",
|
||||
"required": [
|
||||
"generated_text",
|
||||
"finish_reason",
|
||||
"generated_tokens",
|
||||
"prefill",
|
||||
"tokens"
|
||||
],
|
||||
"properties": {
|
||||
"finish_reason": {
|
||||
"$ref": "#/components/schemas/FinishReason"
|
||||
},
|
||||
"generated_text": {
|
||||
"type": "string",
|
||||
"example": "test"
|
||||
},
|
||||
"generated_tokens": {
|
||||
"type": "integer",
|
||||
"format": "int32",
|
||||
"example": 1
|
||||
},
|
||||
"prefill": {
|
||||
"type": "array",
|
||||
"items": {
|
||||
"$ref": "#/components/schemas/PrefillToken"
|
||||
}
|
||||
},
|
||||
"seed": {
|
||||
"type": "integer",
|
||||
"format": "int64",
|
||||
"example": 42
|
||||
},
|
||||
"tokens": {
|
||||
"type": "array",
|
||||
"items": {
|
||||
"$ref": "#/components/schemas/Token"
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"Details": {
|
||||
"type": "object",
|
||||
"required": [
|
||||
"finish_reason",
|
||||
"generated_tokens"
|
||||
"generated_tokens",
|
||||
"prefill",
|
||||
"tokens"
|
||||
],
|
||||
"properties": {
|
||||
"best_of_sequences": {
|
||||
"type": "array",
|
||||
"items": {
|
||||
"$ref": "#/components/schemas/BestOfSequence"
|
||||
}
|
||||
},
|
||||
"finish_reason": {
|
||||
"$ref": "#/components/schemas/FinishReason"
|
||||
},
|
||||
@ -247,11 +296,15 @@
|
||||
"ErrorResponse": {
|
||||
"type": "object",
|
||||
"required": [
|
||||
"error"
|
||||
"error",
|
||||
"error_type"
|
||||
],
|
||||
"properties": {
|
||||
"error": {
|
||||
"type": "string"
|
||||
},
|
||||
"error_type": {
|
||||
"type": "string"
|
||||
}
|
||||
}
|
||||
},
|
||||
@ -266,6 +319,13 @@
|
||||
"GenerateParameters": {
|
||||
"type": "object",
|
||||
"properties": {
|
||||
"best_of": {
|
||||
"type": "integer",
|
||||
"default": "null",
|
||||
"example": 1,
|
||||
"nullable": true,
|
||||
"exclusiveMinimum": 0.0
|
||||
},
|
||||
"details": {
|
||||
"type": "boolean",
|
||||
"default": "true"
|
||||
@ -297,7 +357,11 @@
|
||||
},
|
||||
"seed": {
|
||||
"type": "integer",
|
||||
"format": "int64"
|
||||
"format": "int64",
|
||||
"default": "null",
|
||||
"example": "null",
|
||||
"nullable": true,
|
||||
"exclusiveMinimum": 0.0
|
||||
},
|
||||
"stop": {
|
||||
"type": "array",
|
||||
@ -334,6 +398,15 @@
|
||||
"maximum": 1.0,
|
||||
"exclusiveMinimum": 0.0
|
||||
},
|
||||
"typical_p": {
|
||||
"type": "number",
|
||||
"format": "float",
|
||||
"default": "null",
|
||||
"example": 0.95,
|
||||
"nullable": true,
|
||||
"maximum": 1.0,
|
||||
"exclusiveMinimum": 0.0
|
||||
},
|
||||
"watermark": {
|
||||
"type": "boolean",
|
||||
"default": "false",
|
||||
|
||||
@ -87,21 +87,21 @@ async fn health(infer: Extension<Infer>) -> Result<(), (StatusCode, Json<ErrorRe
|
||||
|
||||
/// Generate tokens
|
||||
#[utoipa::path(
|
||||
post,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/generate",
|
||||
request_body = GenerateRequest,
|
||||
responses(
|
||||
(status = 200, description = "Generated Text", body = GenerateResponse),
|
||||
(status = 424, description = "Generation Error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Request failed during generation"})),
|
||||
(status = 429, description = "Model is overloaded", body = ErrorResponse,
|
||||
example = json ! ({"error": "Model is overloaded"})),
|
||||
(status = 422, description = "Input validation error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Input validation error"})),
|
||||
(status = 500, description = "Incomplete generation", body = ErrorResponse,
|
||||
example = json ! ({"error": "Incomplete generation"})),
|
||||
)
|
||||
post,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/generate",
|
||||
request_body = GenerateRequest,
|
||||
responses(
|
||||
(status = 200, description = "Generated Text", body = GenerateResponse),
|
||||
(status = 424, description = "Generation Error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Request failed during generation"})),
|
||||
(status = 429, description = "Model is overloaded", body = ErrorResponse,
|
||||
example = json ! ({"error": "Model is overloaded"})),
|
||||
(status = 422, description = "Input validation error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Input validation error"})),
|
||||
(status = 500, description = "Incomplete generation", body = ErrorResponse,
|
||||
example = json ! ({"error": "Incomplete generation"})),
|
||||
)
|
||||
)]
|
||||
#[instrument(
|
||||
skip(infer),
|
||||
@ -253,26 +253,26 @@ async fn generate(
|
||||
|
||||
/// Generate a stream of token using Server-Sent Events
|
||||
#[utoipa::path(
|
||||
post,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/generate_stream",
|
||||
request_body = GenerateRequest,
|
||||
responses(
|
||||
(status = 200, description = "Generated Text", body = StreamResponse,
|
||||
content_type = "text/event-stream"),
|
||||
(status = 424, description = "Generation Error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Request failed during generation"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 429, description = "Model is overloaded", body = ErrorResponse,
|
||||
example = json ! ({"error": "Model is overloaded"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 422, description = "Input validation error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Input validation error"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 500, description = "Incomplete generation", body = ErrorResponse,
|
||||
example = json ! ({"error": "Incomplete generation"}),
|
||||
content_type = "text/event-stream"),
|
||||
)
|
||||
post,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/generate_stream",
|
||||
request_body = GenerateRequest,
|
||||
responses(
|
||||
(status = 200, description = "Generated Text", body = StreamResponse,
|
||||
content_type = "text/event-stream"),
|
||||
(status = 424, description = "Generation Error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Request failed during generation"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 429, description = "Model is overloaded", body = ErrorResponse,
|
||||
example = json ! ({"error": "Model is overloaded"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 422, description = "Input validation error", body = ErrorResponse,
|
||||
example = json ! ({"error": "Input validation error"}),
|
||||
content_type = "text/event-stream"),
|
||||
(status = 500, description = "Incomplete generation", body = ErrorResponse,
|
||||
example = json ! ({"error": "Incomplete generation"}),
|
||||
content_type = "text/event-stream"),
|
||||
)
|
||||
)]
|
||||
#[instrument(
|
||||
skip(infer),
|
||||
@ -434,10 +434,10 @@ async fn generate_stream(
|
||||
|
||||
/// Prometheus metrics scrape endpoint
|
||||
#[utoipa::path(
|
||||
get,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/metrics",
|
||||
responses((status = 200, description = "Prometheus Metrics", body = String))
|
||||
get,
|
||||
tag = "Text Generation Inference",
|
||||
path = "/metrics",
|
||||
responses((status = 200, description = "Prometheus Metrics", body = String))
|
||||
)]
|
||||
async fn metrics(prom_handle: Extension<PrometheusHandle>) -> String {
|
||||
prom_handle.render()
|
||||
@ -463,36 +463,36 @@ pub async fn run(
|
||||
// OpenAPI documentation
|
||||
#[derive(OpenApi)]
|
||||
#[openapi(
|
||||
paths(
|
||||
generate,
|
||||
generate_stream,
|
||||
metrics,
|
||||
),
|
||||
components(
|
||||
schemas(
|
||||
GenerateRequest,
|
||||
GenerateParameters,
|
||||
PrefillToken,
|
||||
Token,
|
||||
GenerateResponse,
|
||||
BestOfSequence,
|
||||
Details,
|
||||
FinishReason,
|
||||
StreamResponse,
|
||||
StreamDetails,
|
||||
ErrorResponse,
|
||||
)
|
||||
),
|
||||
tags(
|
||||
(name = "Text Generation Inference", description = "Hugging Face Text Generation Inference API")
|
||||
),
|
||||
info(
|
||||
title = "Text Generation Inference",
|
||||
license(
|
||||
name = "Apache 2.0",
|
||||
url = "https://www.apache.org/licenses/LICENSE-2.0"
|
||||
)
|
||||
)
|
||||
paths(
|
||||
generate,
|
||||
generate_stream,
|
||||
metrics,
|
||||
),
|
||||
components(
|
||||
schemas(
|
||||
GenerateRequest,
|
||||
GenerateParameters,
|
||||
PrefillToken,
|
||||
Token,
|
||||
GenerateResponse,
|
||||
BestOfSequence,
|
||||
Details,
|
||||
FinishReason,
|
||||
StreamResponse,
|
||||
StreamDetails,
|
||||
ErrorResponse,
|
||||
)
|
||||
),
|
||||
tags(
|
||||
(name = "Text Generation Inference", description = "Hugging Face Text Generation Inference API")
|
||||
),
|
||||
info(
|
||||
title = "Text Generation Inference",
|
||||
license(
|
||||
name = "Apache 2.0",
|
||||
url = "https://www.apache.org/licenses/LICENSE-2.0"
|
||||
)
|
||||
)
|
||||
)]
|
||||
struct ApiDoc;
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user