ALF&SAM-API文档

/api/caninterp

Interprets CAN messages based on ISO 14229 standard.
Provides detailed diagnostic session analysis and negative response code explanations.

Request Body

JSON

{
  "user": "User_ID",
  "query": "Tx:0x10 0x02;Rx:0x7F 0x10 0x7E"
}

Expected Response

JSON

{
  "result": "当使用 DiagnosticSessionControl (0x10) service 下的ProgrammingSession (0x02)的时候取得了负相应 NRC 0x7E, 表示: the server does not support the requested sub-function in the session currently active"
}

/api/getstudydetails

Retrieves all published research papers from ALF & SAM.
Supports natural language queries for paper discovery.

Request Body

JSON

{
  "user": "User_ID",
  "query": "Tell me how to design a deeplearning model for bluetooth keyless entry."
}

Response Format

Markdown

### 标题
...

### 作者
...

### 摘要
...

### 链接
https://www.researchgate.net...

/api/det

这个服务主要是响应客户端发送过来的POST请求，并且从中抽取出DICOM对象以及检测算法的必要参数
最终服务会根据这些参数返回检测结果(OK or NG)，以及标注过的图片。
关于这个服务的更多信息可以参考这个Notebook。

Request Body

JSON

{
  "dicom_stream": "DICOM对象，格式为base64编码",
  "grey_lvl_upper": "检测使用的图片HU上限，DICOM会根据这个值截断并且归一化到1.0（default为10000）"
  "grey_lvl_lower": "检测使用的图片HU下限，DICOM会根据这个值截断并且归一化到0.0（default为0）"
  "chunk_size": "原始DICOM转换成的灰度图会按照这个值切成方块，根据环境变量里的检测器输入大小做必要的resize（default为640）"
}

Expected Response

JSON

{
  "image_stream": "标注过的图片"
  "is_ok": "是否是良品（OK or NG）"
}

/api/ontorerank

Rerank document list from a vanilla RAG system
You need to provide query graph, list of graph each one from a retrieved document
This tool is designed for restrained search defined by an ontology.
More details of this tool can be found in this link and this notebook.

Request Body

JSON

{
  "user": "User_ID",
  "query_kg": "graph from your question"
  "doc_kg": "A List of graph from your doc"
}

Response Format

JSON

{
  "score": "Rerank score one for each document",
}

Ontology Example

JSON

{
	"entities":[
		"NamedPizza - a pizza with specific name",
		...
	],
	"object_property": [
		"is a kind of",
		"hasTopping",
		...
	]
	"graph": [
		["NamedPizza", "hasTopping", "PizzaTopping"],
		["PizzaTopping", "is a kind of", "Topping"],
		...
	]
}

Usage Example

JSON

Request:
{
	"user": "XXX",
	"query_kg":[["NamedPizza", "hasSpiciness", "Medium"], ["NamedPizza", "hasTopping", "vegetable"]],
	"doc_kg":[[["Veneziana", "hasTopping", "OnionTopping"], ["OnionTopping", "is a kind of", "VegetableTopping"], ["OnionTopping", "hasSpiciness", "Medium"]], [["Siciliana", "hasTopping", "GarlicTopping"], ["GarlicTopping", "is a kind of", "VegetableTopping"], ["GarlicTopping", "hasSpiciness", "Medium"]], [["Fiorentina", "hasTopping", "GarlicTopping"], ["GarlicTopping", "is a kind of", "VegetableTopping"], ["GarlicTopping", "hasSpiciness", "Medium"]], [["FruttiDiMare", "hasTopping", "GarlicTopping"], ["GarlicTopping", "is a kind of", "VegetableTopping"], ["GarlicTopping", "hasSpiciness", "Medium"]], [["PolloAdAstra", "hasTopping", "GarlicTopping"], ["GarlicTopping", "is a kind of", "VegetableTopping"], ["GarlicTopping", "hasSpiciness", "Medium"]], [["FourSeasons", "hasTopping", "PeperoniSausageTopping"], ["PeperoniSausageTopping", "is a kind of", "MeatTopping"], ["PeperoniSausageTopping", "hasSpiciness", "Medium"]], [["AmericanHot", "hasTopping", "PeperoniSausageTopping"], ["PeperoniSausageTopping", "is a kind of", "MeatTopping"], ["PeperoniSausageTopping", "hasSpiciness", "Medium"]], [["Parmense", "hasTopping", "AsparagusTopping"], ["AsparagusTopping", "is a kind of", "VegetableTopping"], ["AsparagusTopping", "hasSpiciness", "Mild"]], [["Veneziana", "hasTopping", "SultanaTopping"], ["SultanaTopping", "is a kind of", "FruitTopping"], ["SultanaTopping", "hasSpiciness", "Medium"]], [["SloppyGiuseppe", "hasTopping", "OnionTopping"], ["OnionTopping", "is a kind of", "VegetableTopping"], ["OnionTopping", "hasSpiciness", "Medium"]]]
}

Response:
{
	{"result": "0.92, 0.92, 0.92, 0.92, 0.92, 0.86, 0.86, 0.82, 0.87, 0.92", "error": ""}
}