From cdd1ce0056c8ad98ec92325d97de3798d9b23719 Mon Sep 17 00:00:00 2001 From: Bobholamovic Date: Fri, 15 Nov 2024 13:55:20 +0800 Subject: [PATCH 1/5] Add PP-ShiTuV2 FastAPI app --- .../cv_pipelines/face_recognition.en.md | 7 +- .../cv_pipelines/face_recognition.md | 7 +- .../general_image_recognition.en.md | 659 ++++++++--------- .../cv_pipelines/general_image_recognition.md | 693 +++++++----------- .../image_anomaly_detection.en.md | 7 +- .../cv_pipelines/image_anomaly_detection.md | 7 +- .../cv_pipelines/image_classification.en.md | 7 +- .../cv_pipelines/image_classification.md | 7 +- .../image_multi_label_classification.en.md | 7 +- .../image_multi_label_classification.md | 7 +- .../cv_pipelines/instance_segmentation.en.md | 7 +- .../cv_pipelines/instance_segmentation.md | 7 +- .../cv_pipelines/object_detection.en.md | 7 +- .../cv_pipelines/object_detection.md | 7 +- .../pedestrian_attribute_recognition.en.md | 7 +- .../pedestrian_attribute_recognition.md | 7 +- .../cv_pipelines/semantic_segmentation.en.md | 7 +- .../cv_pipelines/semantic_segmentation.md | 7 +- .../cv_pipelines/small_object_detection.en.md | 7 +- .../cv_pipelines/small_object_detection.md | 7 +- .../vehicle_attribute_recognition.en.md | 7 +- .../vehicle_attribute_recognition.md | 7 +- ...ocument_scene_information_extraction.en.md | 7 +- .../document_scene_information_extraction.md | 7 +- .../tutorials/ocr_pipelines/OCR.en.md | 7 +- .../tutorials/ocr_pipelines/OCR.md | 7 +- .../ocr_pipelines/formula_recognition.en.md | 7 +- .../ocr_pipelines/formula_recognition.md | 7 +- .../ocr_pipelines/layout_parsing.en.md | 7 +- .../tutorials/ocr_pipelines/layout_parsing.md | 7 +- .../ocr_pipelines/seal_recognition.en.md | 7 +- .../ocr_pipelines/seal_recognition.md | 7 +- .../ocr_pipelines/table_recognition.en.md | 7 +- .../ocr_pipelines/table_recognition.md | 7 +- .../time_series_anomaly_detection.en.md | 7 +- .../time_series_anomaly_detection.md | 7 +- .../time_series_classification.en.md | 7 +- .../time_series_classification.md | 7 +- .../time_series_forecasting.en.md | 7 +- .../time_series_forecasting.md | 7 +- .../pipelines/ppchatocrv3/ppchatocrv3.py | 2 +- .../serving/_pipeline_apps/__init__.py | 8 + .../_pipeline_apps/anomaly_detection.py | 4 +- .../_pipeline_apps/formula_recognition.py | 4 +- .../_pipeline_apps/image_classification.py | 4 +- .../_pipeline_apps/instance_segmentation.py | 4 +- .../serving/_pipeline_apps/layout_parsing.py | 26 +- .../multi_label_image_classification.py | 4 +- .../_pipeline_apps/object_detection.py | 4 +- .../pipelines/serving/_pipeline_apps/ocr.py | 4 +- .../pedestrian_attribute_recognition.py | 4 +- .../serving/_pipeline_apps/pp_shitu_v2.py | 313 ++++++++ .../serving/_pipeline_apps/ppchatocrv3.py | 30 +- .../_pipeline_apps/seal_recognition.py | 4 +- .../_pipeline_apps/semantic_segmentation.py | 4 +- .../_pipeline_apps/small_object_detection.py | 4 +- .../_pipeline_apps/table_recognition.py | 8 +- .../pipelines/serving/_pipeline_apps/ts_ad.py | 4 +- .../pipelines/serving/_pipeline_apps/ts_fc.py | 4 +- .../vehicle_attribute_recognition.py | 4 +- paddlex/inference/pipelines/serving/app.py | 1 + .../pipelines/serving/file_storage.py | 80 -- .../inference/pipelines/serving/storage.py | 161 ++++ paddlex/inference/pipelines/serving/utils.py | 25 +- 64 files changed, 1295 insertions(+), 1033 deletions(-) create mode 100644 paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py delete mode 100644 paddlex/inference/pipelines/serving/file_storage.py create mode 100644 paddlex/inference/pipelines/serving/storage.py diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md index dbd35d7bf0..5a74a53ad5 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md @@ -374,9 +374,10 @@ Below are the API reference and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

@@ -425,7 +426,7 @@ Below are the API reference and multi-language service invocation examples:
-

The operations provided by the service are as follows:

+

The main operations provided by the service are as follows:

diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.md b/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.md index 149b2cb684..9042bd09c1 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.md @@ -391,9 +391,10 @@ data_root # 数据集根目录，目录名称可以改变
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -442,7 +443,7 @@ data_root # 数据集根目录，目录名称可以改变
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md index 9ab6bc00b1..3093c219a1 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md @@ -367,9 +367,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -377,32 +378,32 @@ Below are the API references and multi-language service invocation examples: - + - + - +
Name TypeDescriptionMeaning
errorCode integerError code. Fixed as 0.Error code. Fixed to 0.
errorMsg stringError message. Fixed as "Success".Error description. Fixed to "Success".
-

The response body may also have a result property of type object, which stores the operation result information.

+

The response body may also have a result property, which is an object type that stores operation result information.

    -
  • When the request is not processed successfully, the response body properties are as follows:
    • +
  • When the request is not processed successfully, the properties of the response body are as follows:
- + @@ -414,58 +415,235 @@ Below are the API references and multi-language service invocation examples: - +
Name TypeDescriptionMeaning
errorMsg stringError message.Error description.
-

Operations provided by the service are as follows:

+

The main operations provided by the service are as follows:

    -
  • infer
    • +
  • buildIndex
-

Classify images.

-

POST /image-classification

+

Build feature vector index.

+

POST /shitu-index-build

    -
  • The request body properties are as follows:
    • +
  • The properties of the request body are as follows:
- + + + + + + + +
Name TypeDescriptionMeaning Required
imageLabelPairsarrayImage-label pairs for building the index.Yes
+

Each element in imageLabelPairs is an object with the following properties:

+ + + + + + + + + + - + + + + + + + + +
NameTypeMeaning
image stringThe URL of an image file accessible by the service or the Base64 encoded result of the image file content.The URL of an image file accessible by the service, or the Base64 encoding result of the image file content.
labelstringLabel.
+
    +
  • When the request is processed successfully, the result of the response body has the following properties:
    • +
+ + + + + + + + + + + + + + + + + + + + +
NameTypeMeaning
indexKeystringThe key corresponding to the index, used to identify the established index. Can be used as input for other operations.
idMapobjectMapping from vector ID to label.
+
    +
  • addImagesToIndex
    • +
+

Add images (corresponding feature vectors) to the index.

+

POST /shitu-index-add

+
    +
  • The properties of the request body are as follows:
    • +
+ + + + + + + + + + + + + + + + + + + + + +
NameTypeMeaningRequired
imageLabelPairsarrayImage-label pairs for building the index.Yes
indexKeystringThe key corresponding to the index. Provided by the buildIndex operation. Yes
+

Each element in imageLabelPairs is an object with the following properties:

+ + + + + + + + + + + + + + + + + + + + +
NameTypeMeaning
imagestringThe URL of an image file accessible by the service, or the Base64 encoding result of the image file content.
labelstringLabel.
+
    +
  • When the request is processed successfully, the result of the response body has the following properties:
    • +
+ + + + + + + + + - + - - + + + +
NameTypeMeaning
inferenceParamsidMap objectInference parameters.NoMapping from vector ID to label.
+
    +
  • removeImagesFromIndex
    • +
+

Remove images (corresponding feature vectors) from the index.

+

POST /shitu-index-remove

+
    +
  • The properties of the request body are as follows:
    • +
+ + + + + + + + + + + + + + + + + + + + +
NameTypeMeaningRequired
idsarrayIDs of the vectors to be removed from the index.Yes
indexKeystringThe key corresponding to the index. Provided by the buildIndex operation.Yes
-

The properties of inferenceParams are as follows:

+
    +
  • When the request is processed successfully, the result of the response body has the following properties:
    • +
- + + + + + + + + + + +
Name TypeDescriptionMeaning
idMapobjectMapping from vector ID to label.
+
    +
  • infer
    • +
+

Perform image recognition.

+

POST /shitu-infer

+
    +
  • The properties of the request body are as follows:
    • +
+ + + + + + - - - + + + + + + + + + @@ -478,61 +656,72 @@ Below are the API references and multi-language service invocation examples: - + - + - + - +
NameTypeMeaning Required
topKintegerOnly the top topK categories with the highest scores will be retained in the results.imagestringThe URL of an image file accessible by the service, or the Base64 encoding result of the image file content.Yes
indexKeystringThe key corresponding to the index. Provided by the buildIndex operation. No
Name TypeDescriptionMeaning
categoriesdetectedObjects arrayImage category information.Information of the detected targets.
image stringThe image classification result image. The image is in JPEG format and encoded using Base64.Recognition result image. The image is in JPEG format, encoded with Base64.
-

Each element in categories is an object with the following properties:

+

Each element in detectedObjects is an object with the following properties:

- + - - - + + + + + + + + + + + + + + + +
Name TypeDescriptionMeaning
idintegerCategory ID.bboxarrayTarget location. The elements in the array are the x-coordinate of the upper-left corner, the y-coordinate of the upper-left corner, the x-coordinate of the lower-right corner, and the y-coordinate of the lower-right corner, respectively.
recResultsarrayRecognition results.
scorenumberDetection score.
+

Each element in recResults is an object with the following properties:

+ + + + + + + + - + - + - +
NameTypeMeaning
namelabel stringCategory name.Label.
score numberCategory score.Recognition score.
-

An example of result is as follows:

-
{
-"categories": [
-{
-"id": 5,
-"name": "Rabbit",
-"score": 0.93
-}
-],
-"image": "xxxxxx"
-}
-
+
Multi-Language Service Invocation Examples @@ -541,336 +730,72 @@ Below are the API references and multi-language service invocation examples: 
import base64
+import pprint
+import sys
+
 import requests
 
-API_URL = "http://localhost:8080/image-classification"
-image_path = "./demo.jpg"
+API_BASE_URL = "http://0.0.0.0:8080"
+
+base_image_label_pairs = [
+    {"image": "./demo0.jpg", "label": "rabbit"},
+    {"image": "./demo1.jpg", "label": "rabbit"},
+    {"image": "./demo2.jpg", "label": "puppy"},
+]
+image_label_pairs_to_add = [
+    {"image": "./demo3.jpg", "label": "puppy"},
+]
+infer_image_path = "./demo4.jpg"
 output_image_path = "./out.jpg"
 
-with open(image_path, "rb") as file:
+for pair in base_image_label_pairs:
+    with open(pair["image"], "rb") as file:
+        image_bytes = file.read()
+        image_data = base64.b64encode(image_bytes).decode("ascii")
+    pair["image"] = image_data
+
+payload = {"imageLabelPairs": base_image_label_pairs}
+resp_index_build = requests.post(f"{API_BASE_URL}/shitu-index-build", json=payload)
+if resp_index_build.status_code != 200:
+    print(f"Request to shitu-index-build failed with status code {resp_index_build}.")
+    pprint.pp(resp_index_build.json())
+    sys.exit(1)
+result_index_build = resp_index_build.json()["result"]
+print(f"Number of images indexed: {len(result_index_build['idMap'])}")
+
+for pair in image_label_pairs_to_add:
+    with open(pair["image"], "rb") as file:
+        image_bytes = file.read()
+        image_data = base64.b64encode(image_bytes).decode("ascii")
+    pair["image"] = image_data
+
+payload = {"imageLabelPairs": image_label_pairs_to_add, "indexKey": result_index_build["indexKey"]}
+resp_index_add = requests.post(f"{API_BASE_URL}/shitu-index-add", json=payload)
+if resp_index_add.status_code != 200:
+    print(f"Request to shitu-index-add failed with status code {resp_index_add}.")
+    pprint.pp(resp_index_add.json())
+    sys.exit(1)
+result_index_add = resp_index_add.json()["result"]
+print(f"Number of images indexed: {len(result_index_add['idMap'])}")
+
+with open(infer_image_path, "rb") as file:
     image_bytes = file.read()
     image_data = base64.b64encode(image_bytes).decode("ascii")
 
-payload = {"image": image_data}
-
-response = requests.post(API_URL, json=payload)
+payload = {"image": image_data, "indexKey": result_index_build["indexKey"]}
+resp_infer = requests.post(f"{API_BASE_URL}/shitu-infer", json=payload)
+if resp_infer.status_code != 200:
+    print(f"Request to shitu-infer failed with status code {resp_infer}.")
+    pprint.pp(resp_infer.json())
+    sys.exit(1)
+result_infer = resp_infer.json()["result"]
 
-assert response.status_code == 200
-result = response.json()["result"]
 with open(output_image_path, "wb") as file:
-    file.write(base64.b64decode(result["image"]))
+    file.write(base64.b64decode(result_infer["image"]))
 print(f"Output image saved at {output_image_path}")
-print("\nCategories:")
-print(result["categories"])
+print("\nDetected objects:")
+pprint.pp(result_infer["detectedObjects"])
-
C++ - -
#include <iostream>
-#include "cpp-httplib/httplib.h" // https://github.com/Huiyicc/cpp-httplib
-#include "nlohmann/json.hpp" // https://github.com/nlohmann/json
-#include "base64.hpp" // https://github.com/tobiaslocker/base64
-
-int main() {
-    httplib::Client client("localhost:8080");
-    const std::string imagePath = "./demo.jpg";
-    const std::string outputImagePath = "./out.jpg";
-
-    httplib::Headers headers = {
-        {"Content-Type", "application/json"}
-    };
-
-    std::ifstream file(imagePath, std::ios::binary | std::ios::ate);
-    std::streamsize size = file.tellg();
-    file.seekg(0, std::ios::beg);
-
-    std::vector<char> buffer(size);
-    if (!file.read(buffer.data(), size)) {
-        std::cerr << "Error reading file." << std::endl;
-        return 1;
-    }
-    std::string bufferStr(reinterpret_cast<const char*>(buffer.data()), buffer.size());
-    std::string encodedImage = base64::to_base64(bufferStr);
-
-    nlohmann::json jsonObj;
-    jsonObj["image"] = encodedImage;
-    std::string body = jsonObj.dump();
-
-    auto response = client.Post("/image-classification", headers, body, "application/json");
-    if (response && response->status == 200) {
-        nlohmann::json jsonResponse = nlohmann::json::parse(response->body);
-        auto result = jsonResponse["result"];
-
-        encodedImage = result["image"];
-        std::string decodedString = base64::from_base64(encodedImage);
-        std::vector<unsigned char> decodedImage(decodedString.begin(), decodedString.end());
-        std::ofstream outputImage(outPutImagePath, std::ios::binary | std::ios::out);
-        if (outputImage.is_open()) {
-            outputImage.write(reinterpret_cast<char*>(decodedImage.data()), decodedImage.size());
-            outputImage.close();
-            std::cout << "Output image saved at " << outPutImagePath << std::endl;
-        } else {
-            std::cerr << "Unable to open file for writing: " << outPutImagePath << std::endl;
-        }
-
-        auto categories = result["categories"];
-        std::cout << "\nCategories:" << std::endl;
-        for (const auto& category : categories) {
-            std::cout << category << std::endl;
-        }
-    } else {
-        std::cout << "Failed to send HTTP request." << std::endl;
-        return 1;
-    }
-
-    return 0;
-}
-
- -
Java - -
import okhttp3.*;
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.fasterxml.jackson.databind.JsonNode;
-import com.fasterxml.jackson.databind.node.ObjectNode;
-
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.util.Base64;
-
-public class Main {
-    public static void main(String[] args) throws IOException {
-        String API_URL = "http://localhost:8080/image-classification";
-        String imagePath = "./demo.jpg";
-        String outputImagePath = "./out.jpg";
-
-        File file = new File(imagePath);
-        byte[] fileContent = java.nio.file.Files.readAllBytes(file.toPath());
-        String imageData = Base64.getEncoder().encodeToString(fileContent);
-
-        ObjectMapper objectMapper = new ObjectMapper();
-        ObjectNode params = objectMapper.createObjectNode();
-        params.put("image", imageData);
-
-        OkHttpClient client = new OkHttpClient();
-        MediaType JSON = MediaType.Companion.get("application/json; charset=utf-8");
-        RequestBody body = RequestBody.Companion.create(params.toString(), JSON);
-        Request request = new Request.Builder()
-                .url(API_URL)
-                .post(body)
-                .build();
-
-        try (Response response = client.newCall(request).execute()) {
-            if (response.isSuccessful()) {
-                String responseBody = response.body().string();
-                JsonNode resultNode = objectMapper.readTree(responseBody);
-                JsonNode result = resultNode.get("result");
-                String base64Image = result.get("image").asText();
-                JsonNode categories = result.get("categories");
-
-                byte[] imageBytes = Base64.getDecoder().decode(base64Image);
-                try (FileOutputStream fos = new FileOutputStream(outputImagePath)) {
-                    fos.write(imageBytes);
-                }
-                System.out.println("Output image saved at " + outputImagePath);
-                System.out.println("\nCategories: " + categories.toString());
-            } else {
-                System.err.println("Request failed with code: " + response.code());
-            }
-        }
-    }
-}
-
- -
Go - -
package main
-
-import (
-    "bytes"
-    "encoding/base64"
-    "encoding/json"
-    "fmt"
-    "io/ioutil"
-    "net/http"
-)
-
-func main() {
-    API_URL := "http://localhost:8080/image-classification"
-    imagePath := "./demo.jpg"
-    outputImagePath := "./out.jpg"
-
-    imageBytes, err := ioutil.ReadFile(imagePath)
-    if err != nil {
-        fmt.Println("Error reading image file:", err)
-        return
-    }
-    imageData := base64.StdEncoding.EncodeToString(imageBytes)
-
-    payload := map[string]string{"image": imageData}
-    payloadBytes, err := json.Marshal(payload)
-    if err != nil {
-        fmt.Println("Error marshaling payload:", err)
-        return
-    }
-
-    client := &http.Client{}
-    req, err := http.NewRequest("POST", API_URL, bytes.NewBuffer(payloadBytes))
-    if err != nil {
-        fmt.Println("Error creating request:", err)
-        return
-    }
-
-    res, err := client.Do(req)
-    if err != nil {
-        fmt.Println("Error sending request:", err)
-        return
-    }
-    defer res.Body.Close()
-
-    body, err := ioutil.ReadAll(res.Body)
-    if err != nil {
-        fmt.Println("Error reading response body:", err)
-        return
-    }
-    type Response struct {
-        Result struct {
-            Image      string   `json:"image"`
-            Categories []map[string]interface{} `json:"categories"`
-        } `json:"result"`
-    }
-    var respData Response
-    err = json.Unmarshal([]byte(string(body)), &respData)
-    if err != nil {
-        fmt.Println("Error unmarshaling response body:", err)
-        return
-    }
-
-    outputImageData, err := base64.StdEncoding.DecodeString(respData.Result.Image)
-    if err != nil {
-        fmt.Println("Error decoding base64 image data:", err)
-        return
-    }
-    err = ioutil.WriteFile(outputImagePath, outputImageData, 0644)
-    if err != nil {
-        fmt.Println("Error writing image to file:", err)
-        return
-    }
-    fmt.Printf("Image saved at %s.jpg\n", outputImagePath)
-    fmt.Println("\nCategories:")
-    for _, category := range respData.Result.Categories {
-        fmt.Println(category)
-    }
-}
-
- -
C# - -
using System;
-using System.IO;
-using System.Net.Http;
-using System.Net.Http.Headers;
-using System.Text;
-using System.Threading.Tasks;
-using Newtonsoft.Json.Linq;
-
-class Program
-{
-    static readonly string API_URL = "http://localhost:8080/image-classification";
-    static readonly string imagePath = "./demo.jpg";
-    static readonly string outputImagePath = "./out.jpg";
-
-    static async Task Main(string[] args)
-    {
-        var httpClient = new HttpClient();
-
-        byte[] imageBytes = File.ReadAllBytes(imagePath);
-        string image_data = Convert.ToBase64String(imageBytes);
-
-        var payload = new JObject{ { "image", image_data } };
-        var content = new StringContent(payload.ToString(), Encoding.UTF8, "application/json");
-
-        HttpResponseMessage response = await httpClient.PostAsync(API_URL, content);
-        response.EnsureSuccessStatusCode();
-
-        string responseBody = await response.Content.ReadAsStringAsync();
-        JObject jsonResponse = JObject.Parse(responseBody);
-
-        string base64Image = jsonResponse["result"]["image"].ToString();
-        byte[] outputImageBytes = Convert.FromBase64String(base64Image);
-
-        File.WriteAllBytes(outputImagePath, outputImageBytes);
-        Console.WriteLine($"Output image saved at {outputImagePath}");
-        Console.WriteLine("\nCategories:");
-        Console.WriteLine(jsonResponse["result"]["categories"].ToString());
-    }
-}
-
- -
Node.js - -
const axios = require('axios');
-const fs = require('fs');
-
-const API_URL = 'http://localhost:8080/image-classification'
-const imagePath = './demo.jpg'
-const outputImagePath = "./out.jpg";
-
-let config = {
-   method: 'POST',
-   maxBodyLength: Infinity,
-   url: API_URL,
-   data: JSON.stringify({
-    'image': encodeImageToBase64(imagePath)
-  })
-};
-
-function encodeImageToBase64(filePath) {
-  const bitmap = fs.readFileSync(filePath);
-  return Buffer.from(bitmap).toString('base64');
-}
-
-axios.request(config)
-.then((response) => {
-    const result = response.data["result"];
-    const imageBuffer = Buffer.from(result["image"], 'base64');
-    fs.writeFile(outputImagePath, imageBuffer, (err) => {
-      if (err) throw err;
-      console.log(`Output image saved at ${outputImagePath}`);
-    });
-    console.log("\nCategories:");
-    console.log(result["categories"]);
-})
-.catch((error) => {
-  console.log(error);
-});
-
-
PHP - -
<?php
-
-$API_URL = "http://localhost:8080/image-classification";
-$image_path = "./demo.jpg";
-$output_image_path = "./out.jpg";
-
-$image_data = base64_encode(file_get_contents($image_path));
-$payload = array("image" => $image_data);
-
-$ch = curl_init($API_URL);
-curl_setopt($ch, CURLOPT_POST, true);
-curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
-curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
-$response = curl_exec($ch);
-curl_close($ch);
-
-$result = json_decode($response, true)["result"];
-file_put_contents($output_image_path, base64_decode($result["image"]));
-echo "Output image saved at " . $output_image_path . "\n";
-echo "\nCategories:\n";
-print_r($result["categories"]);
-?>
-
-

diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md index 99c8d70328..961a83873d 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md @@ -397,9 +397,10 @@ data_root # 数据集根目录，目录名称可以改变
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

@@ -448,12 +449,12 @@ data_root # 数据集根目录，目录名称可以改变
-

服务提供的操作如下：

+

服务提供的主要操作如下：

-

获取图像OCR结果。

-

POST /ocr

+

构建特征向量索引。

+

POST /shitu-index-build

@@ -468,20 +469,191 @@ data_root # 数据集根目录，目录名称可以改变 +imageLabelPairs +array +用于构建索引的图像-标签对。 +是 + + + +

imageLabelPairs中的每个元素为一个object，具有如下属性：

+ + + + + + + + + + + + + + + + + +
名称类型含义
image string 服务可访问的图像文件的URL或图像文件内容的Base64编码结果。
labelstring标签。
+ + + + + + + + + + + + + + + + + + + + + +
名称类型含义
indexKeystring索引对应的键，用于标识建立的索引。可用作其他操作的输入。
idMapobject向量ID到标签的映射。
+ +

将图像（对应的特征向量）加入索引。

+

POST /shitu-index-add

+ + + + + + + + + + + + + + + - + + + + + + +
名称类型含义是否必填
imageLabelPairsarray用于构建索引的图像-标签对。 是
inferenceParamsindexKeystring索引对应的键。由buildIndex操作提供。是
+

imageLabelPairs中的每个元素为一个object，具有如下属性：

+ + + + + + + + + + + + + + + + + + + + +
名称类型含义
imagestring服务可访问的图像文件的URL或图像文件内容的Base64编码结果。
labelstring标签。
+ + + + + + + + + + + + - - + + + +
名称类型含义
idMap object推理参数。否向量ID到标签的映射。
+ +

从索引中移除图像（对应的特征向量）。

+

POST /shitu-index-remove

+ + + + + + + + + + + + + + + + + + + + + + + + +
名称类型含义是否必填
idsarray需要从索引中移除的向量的ID。是
indexKeystring索引对应的键。由buildIndex操作提供。是
+ + + + + + + + + + + + + +
名称类型含义
idMapobject向量ID到标签的映射。
-

inferenceParams的属性如下：

+ +

进行图像识别。

+

POST /shitu-infer

+ @@ -493,9 +665,15 @@ data_root # 数据集根目录，目录名称可以改变 - - - + + + + + + + + + @@ -513,18 +691,18 @@ data_root # 数据集根目录，目录名称可以改变 - + - + - +
maxLongSideinteger推理时，若文本检测模型的输入图像较长边的长度大于maxLongSide，则将对图像进行缩放，使其较长边的长度等于maxLongSide。imagestring服务可访问的图像文件的URL或图像文件内容的Base64编码结果。是
indexKeystring索引对应的键。由buildIndex操作提供。 否
textsdetectedObjects array文本位置、内容和得分。检测到的目标的信息。
image stringOCR结果图，其中标注检测到的文本位置。图像为JPEG格式，使用Base64编码。识别结果图。图像为JPEG格式，使用Base64编码。
-

texts中的每个元素为一个object，具有如下属性：

+

detectedObjects中的每个元素为一个object，具有如下属性：

@@ -535,73 +713,45 @@ data_root # 数据集根目录，目录名称可以改变 - + + + + + + - + - + + + + + +
polybboxarray目标位置。数组中元素依次为边界框左上角x坐标、左上角y坐标、右下角x坐标以及右下角y坐标。
recResults array文本位置。数组中元素依次为包围文本的多边形的顶点坐标。识别结果。
textscorenumber检测得分。
+

recResults中的每个元素为一个object，具有如下属性：

+ + + + + + + + + + + - + - +
名称类型含义
label string文本内容。标签。
score number文本识别得分。识别得分。
-

result示例如下：

-
{
-"texts": [
-{
-"poly": [
-[
-444,
-244
-],
-[
-705,
-244
-],
-[
-705,
-311
-],
-[
-444,
-311
-]
-],
-"text": "北京南站",
-"score": 0.9
-},
-{
-"poly": [
-[
-992,
-248
-],
-[
-1263,
-251
-],
-[
-1263,
-318
-],
-[
-992,
-315
-]
-],
-"text": "天津站",
-"score": 0.5
-}
-],
-"image": "xxxxxx"
-}
-
+
多语言调用服务示例 @@ -610,358 +760,71 @@ data_root # 数据集根目录，目录名称可以改变 
import base64
+import pprint
+import sys
+
 import requests
 
-API_URL = "http://localhost:8080/ocr" # 服务URL
-image_path = "./demo.jpg"
+API_BASE_URL = "http://0.0.0.0:8080"
+
+base_image_label_pairs = [
+    {"image": "./demo0.jpg", "label": "兔子"},
+    {"image": "./demo1.jpg", "label": "兔子"},
+    {"image": "./demo2.jpg", "label": "小狗"},
+]
+image_label_pairs_to_add = [
+    {"image": "./demo3.jpg", "label": "小狗"},
+]
+infer_image_path = "./demo4.jpg"
 output_image_path = "./out.jpg"
 
-# 对本地图像进行Base64编码
-with open(image_path, "rb") as file:
+for pair in base_image_label_pairs:
+    with open(pair["image"], "rb") as file:
+        image_bytes = file.read()
+        image_data = base64.b64encode(image_bytes).decode("ascii")
+    pair["image"] = image_data
+
+payload = {"imageLabelPairs": base_image_label_pairs}
+resp_index_build = requests.post(f"{API_BASE_URL}/shitu-index-build", json=payload)
+if resp_index_build.status_code != 200:
+    print(f"Request to shitu-index-build failed with status code {resp_index_build}.")
+    pprint.pp(resp_index_build.json())
+    sys.exit(1)
+result_index_build = resp_index_build.json()["result"]
+print(f"Number of images indexed: {len(result_index_build['idMap'])}")
+
+for pair in image_label_pairs_to_add:
+    with open(pair["image"], "rb") as file:
+        image_bytes = file.read()
+        image_data = base64.b64encode(image_bytes).decode("ascii")
+    pair["image"] = image_data
+
+payload = {"imageLabelPairs": image_label_pairs_to_add, "indexKey": result_index_build["indexKey"]}
+resp_index_add = requests.post(f"{API_BASE_URL}/shitu-index-add", json=payload)
+if resp_index_add.status_code != 200:
+    print(f"Request to shitu-index-add failed with status code {resp_index_add}.")
+    pprint.pp(resp_index_add.json())
+    sys.exit(1)
+result_index_add = resp_index_add.json()["result"]
+print(f"Number of images indexed: {len(result_index_add['idMap'])}")
+
+with open(infer_image_path, "rb") as file:
     image_bytes = file.read()
     image_data = base64.b64encode(image_bytes).decode("ascii")
 
-payload = {"image": image_data}  # Base64编码的文件内容或者图像URL
-
-# 调用API
-response = requests.post(API_URL, json=payload)
+payload = {"image": image_data, "indexKey": result_index_build["indexKey"]}
+resp_infer = requests.post(f"{API_BASE_URL}/shitu-infer", json=payload)
+if resp_infer.status_code != 200:
+    print(f"Request to shitu-infer failed with status code {resp_infer}.")
+    pprint.pp(resp_infer.json())
+    sys.exit(1)
+result_infer = resp_infer.json()["result"]
 
-# 处理接口返回数据
-assert response.status_code == 200
-result = response.json()["result"]
 with open(output_image_path, "wb") as file:
-    file.write(base64.b64decode(result["image"]))
+    file.write(base64.b64decode(result_infer["image"]))
 print(f"Output image saved at {output_image_path}")
-print("\nDetected texts:")
-print(result["texts"])
-
- -
C++ - -
#include <iostream>
-#include "cpp-httplib/httplib.h" // https://github.com/Huiyicc/cpp-httplib
-#include "nlohmann/json.hpp" // https://github.com/nlohmann/json
-#include "base64.hpp" // https://github.com/tobiaslocker/base64
-
-int main() {
-    httplib::Client client("localhost:8080");
-    const std::string imagePath = "./demo.jpg";
-    const std::string outputImagePath = "./out.jpg";
-
-    httplib::Headers headers = {
-        {"Content-Type", "application/json"}
-    };
-
-    // 对本地图像进行Base64编码
-    std::ifstream file(imagePath, std::ios::binary | std::ios::ate);
-    std::streamsize size = file.tellg();
-    file.seekg(0, std::ios::beg);
-
-    std::vector<char> buffer(size);
-    if (!file.read(buffer.data(), size)) {
-        std::cerr << "Error reading file." << std::endl;
-        return 1;
-    }
-    std::string bufferStr(reinterpret_cast<const char*>(buffer.data()), buffer.size());
-    std::string encodedImage = base64::to_base64(bufferStr);
-
-    nlohmann::json jsonObj;
-    jsonObj["image"] = encodedImage;
-    std::string body = jsonObj.dump();
-
-    // 调用API
-    auto response = client.Post("/ocr", headers, body, "application/json");
-    // 处理接口返回数据
-    if (response && response->status == 200) {
-        nlohmann::json jsonResponse = nlohmann::json::parse(response->body);
-        auto result = jsonResponse["result"];
-
-        encodedImage = result["image"];
-        std::string decodedString = base64::from_base64(encodedImage);
-        std::vector<unsigned char> decodedImage(decodedString.begin(), decodedString.end());
-        std::ofstream outputImage(outPutImagePath, std::ios::binary | std::ios::out);
-        if (outputImage.is_open()) {
-            outputImage.write(reinterpret_cast<char*>(decodedImage.data()), decodedImage.size());
-            outputImage.close();
-            std::cout << "Output image saved at " << outPutImagePath << std::endl;
-        } else {
-            std::cerr << "Unable to open file for writing: " << outPutImagePath << std::endl;
-        }
-
-        auto texts = result["texts"];
-        std::cout << "\nDetected texts:" << std::endl;
-        for (const auto& text : texts) {
-            std::cout << text << std::endl;
-        }
-    } else {
-        std::cout << "Failed to send HTTP request." << std::endl;
-        return 1;
-    }
-
-    return 0;
-}
-
- -
Java - -
import okhttp3.*;
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.fasterxml.jackson.databind.JsonNode;
-import com.fasterxml.jackson.databind.node.ObjectNode;
-
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.util.Base64;
-
-public class Main {
-    public static void main(String[] args) throws IOException {
-        String API_URL = "http://localhost:8080/ocr"; // 服务URL
-        String imagePath = "./demo.jpg"; // 本地图像
-        String outputImagePath = "./out.jpg"; // 输出图像
-
-        // 对本地图像进行Base64编码
-        File file = new File(imagePath);
-        byte[] fileContent = java.nio.file.Files.readAllBytes(file.toPath());
-        String imageData = Base64.getEncoder().encodeToString(fileContent);
-
-        ObjectMapper objectMapper = new ObjectMapper();
-        ObjectNode params = objectMapper.createObjectNode();
-        params.put("image", imageData); // Base64编码的文件内容或者图像URL
-
-        // 创建 OkHttpClient 实例
-        OkHttpClient client = new OkHttpClient();
-        MediaType JSON = MediaType.Companion.get("application/json; charset=utf-8");
-        RequestBody body = RequestBody.Companion.create(params.toString(), JSON);
-        Request request = new Request.Builder()
-                .url(API_URL)
-                .post(body)
-                .build();
-
-        // 调用API并处理接口返回数据
-        try (Response response = client.newCall(request).execute()) {
-            if (response.isSuccessful()) {
-                String responseBody = response.body().string();
-                JsonNode resultNode = objectMapper.readTree(responseBody);
-                JsonNode result = resultNode.get("result");
-                String base64Image = result.get("image").asText();
-                JsonNode texts = result.get("texts");
-
-                byte[] imageBytes = Base64.getDecoder().decode(base64Image);
-                try (FileOutputStream fos = new FileOutputStream(outputImagePath)) {
-                    fos.write(imageBytes);
-                }
-                System.out.println("Output image saved at " + outputImagePath);
-                System.out.println("\nDetected texts: " + texts.toString());
-            } else {
-                System.err.println("Request failed with code: " + response.code());
-            }
-        }
-    }
-}
-
- -
Go - -
package main
-
-import (
-    "bytes"
-    "encoding/base64"
-    "encoding/json"
-    "fmt"
-    "io/ioutil"
-    "net/http"
-)
-
-func main() {
-    API_URL := "http://localhost:8080/ocr"
-    imagePath := "./demo.jpg"
-    outputImagePath := "./out.jpg"
-
-    // 对本地图像进行Base64编码
-    imageBytes, err := ioutil.ReadFile(imagePath)
-    if err != nil {
-        fmt.Println("Error reading image file:", err)
-        return
-    }
-    imageData := base64.StdEncoding.EncodeToString(imageBytes)
-
-    payload := map[string]string{"image": imageData} // Base64编码的文件内容或者图像URL
-    payloadBytes, err := json.Marshal(payload)
-    if err != nil {
-        fmt.Println("Error marshaling payload:", err)
-        return
-    }
-
-    // 调用API
-    client := &http.Client{}
-    req, err := http.NewRequest("POST", API_URL, bytes.NewBuffer(payloadBytes))
-    if err != nil {
-        fmt.Println("Error creating request:", err)
-        return
-    }
-
-    res, err := client.Do(req)
-    if err != nil {
-        fmt.Println("Error sending request:", err)
-        return
-    }
-    defer res.Body.Close()
-
-    // 处理接口返回数据
-    body, err := ioutil.ReadAll(res.Body)
-    if err != nil {
-        fmt.Println("Error reading response body:", err)
-        return
-    }
-    type Response struct {
-        Result struct {
-            Image      string   `json:"image"`
-            Texts []map[string]interface{} `json:"texts"`
-        } `json:"result"`
-    }
-    var respData Response
-    err = json.Unmarshal([]byte(string(body)), &respData)
-    if err != nil {
-        fmt.Println("Error unmarshaling response body:", err)
-        return
-    }
-
-    outputImageData, err := base64.StdEncoding.DecodeString(respData.Result.Image)
-    if err != nil {
-        fmt.Println("Error decoding base64 image data:", err)
-        return
-    }
-    err = ioutil.WriteFile(outputImagePath, outputImageData, 0644)
-    if err != nil {
-        fmt.Println("Error writing image to file:", err)
-        return
-    }
-    fmt.Printf("Image saved at %s.jpg\n", outputImagePath)
-    fmt.Println("\nDetected texts:")
-    for _, text := range respData.Result.Texts {
-        fmt.Println(text)
-    }
-}
-
- -
C# - -
using System;
-using System.IO;
-using System.Net.Http;
-using System.Net.Http.Headers;
-using System.Text;
-using System.Threading.Tasks;
-using Newtonsoft.Json.Linq;
-
-class Program
-{
-    static readonly string API_URL = "http://localhost:8080/ocr";
-    static readonly string imagePath = "./demo.jpg";
-    static readonly string outputImagePath = "./out.jpg";
-
-    static async Task Main(string[] args)
-    {
-        var httpClient = new HttpClient();
-
-        // 对本地图像进行Base64编码
-        byte[] imageBytes = File.ReadAllBytes(imagePath);
-        string image_data = Convert.ToBase64String(imageBytes);
-
-        var payload = new JObject{ { "image", image_data } }; // Base64编码的文件内容或者图像URL
-        var content = new StringContent(payload.ToString(), Encoding.UTF8, "application/json");
-
-        // 调用API
-        HttpResponseMessage response = await httpClient.PostAsync(API_URL, content);
-        response.EnsureSuccessStatusCode();
-
-        // 处理接口返回数据
-        string responseBody = await response.Content.ReadAsStringAsync();
-        JObject jsonResponse = JObject.Parse(responseBody);
-
-        string base64Image = jsonResponse["result"]["image"].ToString();
-        byte[] outputImageBytes = Convert.FromBase64String(base64Image);
-
-        File.WriteAllBytes(outputImagePath, outputImageBytes);
-        Console.WriteLine($"Output image saved at {outputImagePath}");
-        Console.WriteLine("\nDetected texts:");
-        Console.WriteLine(jsonResponse["result"]["texts"].ToString());
-    }
-}
-
- -
Node.js - -
const axios = require('axios');
-const fs = require('fs');
-
-const API_URL = 'http://localhost:8080/ocr'
-const imagePath = './demo.jpg'
-const outputImagePath = "./out.jpg";
-
-let config = {
-   method: 'POST',
-   maxBodyLength: Infinity,
-   url: API_URL,
-   data: JSON.stringify({
-    'image': encodeImageToBase64(imagePath)  // Base64编码的文件内容或者图像URL
-  })
-};
-
-// 对本地图像进行Base64编码
-function encodeImageToBase64(filePath) {
-  const bitmap = fs.readFileSync(filePath);
-  return Buffer.from(bitmap).toString('base64');
-}
-
-// 调用API
-axios.request(config)
-.then((response) => {
-    // 处理接口返回数据
-    const result = response.data["result"];
-    const imageBuffer = Buffer.from(result["image"], 'base64');
-    fs.writeFile(outputImagePath, imageBuffer, (err) => {
-      if (err) throw err;
-      console.log(`Output image saved at ${outputImagePath}`);
-    });
-    console.log("\nDetected texts:");
-    console.log(result["texts"]);
-})
-.catch((error) => {
-  console.log(error);
-});
-
- -
PHP - -
<?php
-
-$API_URL = "http://localhost:8080/ocr"; // 服务URL
-$image_path = "./demo.jpg";
-$output_image_path = "./out.jpg";
-
-// 对本地图像进行Base64编码
-$image_data = base64_encode(file_get_contents($image_path));
-$payload = array("image" => $image_data); // Base64编码的文件内容或者图像URL
-
-// 调用API
-$ch = curl_init($API_URL);
-curl_setopt($ch, CURLOPT_POST, true);
-curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
-curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
-$response = curl_exec($ch);
-curl_close($ch);
-
-// 处理接口返回数据
-$result = json_decode($response, true)["result"];
-file_put_contents($output_image_path, base64_decode($result["image"]));
-echo "Output image saved at " . $output_image_path . "\n";
-echo "\nDetected texts:\n";
-print_r($result["texts"]);
-
-?>
+print("\nDetected objects:")
+pprint.pp(result_infer["detectedObjects"])

diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.en.md index 7513831482..79bfda81d4 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.en.md @@ -217,9 +217,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

@@ -268,7 +269,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.md index 234a435c18..d4d248d3db 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.md @@ -216,9 +216,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -267,7 +268,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.en.md index 29a67640f0..9349787118 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.en.md @@ -803,9 +803,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -854,7 +855,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.md index d87400ae3a..80d354e9b5 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_classification.md @@ -802,9 +802,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -853,7 +854,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.en.md index f975af8e44..6560cf2c7c 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.en.md @@ -243,9 +243,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -294,7 +295,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.md b/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.md index d77eeea049..96a9c8cd78 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/image_multi_label_classification.md @@ -246,9 +246,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -297,7 +298,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.en.md index 9b64abeae2..6ad46972d4 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.en.md @@ -332,9 +332,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -383,7 +384,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.md b/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.md index 7af3b1c707..96518fa990 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/instance_segmentation.md @@ -327,9 +327,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -378,7 +379,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.en.md index ec4fb5faa2..e8e52dd55d 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.en.md @@ -492,9 +492,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -543,7 +544,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.md b/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.md index b6c64a8f24..763b3b81e6 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/object_detection.md @@ -493,9 +493,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -544,7 +545,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.en.md index 6fb22cfc3a..3bae963e2c 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.en.md @@ -250,9 +250,10 @@ Below are the API reference and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -301,7 +302,7 @@ Below are the API reference and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.md b/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.md index 3e46317514..918c3cf682 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/pedestrian_attribute_recognition.md @@ -250,9 +250,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -301,7 +302,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md index da1a4bfd71..9b7a4e0804 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md @@ -364,9 +364,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -415,7 +416,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.md b/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.md index 7fc0cd72e8..5305487667 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.md @@ -361,9 +361,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -412,7 +413,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md index ad8d4cbb28..ba5db67ce7 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md @@ -240,9 +240,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -291,7 +292,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md index 0f45a68ab0..05ee161f76 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md @@ -237,9 +237,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -288,7 +289,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.en.md index a71dd36107..87fa9a1826 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.en.md @@ -248,9 +248,10 @@ Below are the API reference and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -299,7 +300,7 @@ Below are the API reference and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.md b/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.md index 6b08cfa798..60250684c7 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/vehicle_attribute_recognition.md @@ -247,9 +247,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -298,7 +299,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.en.md b/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.en.md index c97a080672..bec85e481c 100644 --- a/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.en.md +++ b/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.en.md @@ -617,9 +617,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -668,7 +669,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • analyzeImages
diff --git a/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.md b/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.md index d3d2b8e94e..24384c4921 100644 --- a/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.md +++ b/docs/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction.md @@ -586,9 +586,10 @@ chat_result.print()
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -637,7 +638,7 @@ chat_result.print()
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • analyzeImages
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md b/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md index 0d431adaa0..8dab252a2d 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md @@ -309,9 +309,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -360,7 +361,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md b/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md index 5379ef7579..f8942179df 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md @@ -311,9 +311,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -362,7 +363,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.en.md b/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.en.md index 7119685673..9f7e031b1d 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.en.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.en.md @@ -290,9 +290,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -341,7 +342,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md b/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md index 512cdf3d04..73f30d4a9d 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/formula_recognition.md @@ -291,9 +291,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -342,7 +343,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.en.md b/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.en.md index 2588f78009..84bb8d1904 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.en.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.en.md @@ -529,9 +529,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body attributes are as follows:
@@ -580,7 +581,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.md b/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.md index cd08565101..2e8c8c3198 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.md @@ -532,9 +532,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -583,7 +584,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md b/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md index c6196c068a..5e6a54835e 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md @@ -439,9 +439,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -490,7 +491,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md b/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md index 4dcda8fc09..9b12cd12f3 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md @@ -449,9 +449,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -500,7 +501,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.en.md b/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.en.md index 977316d1f2..a0b881a2ed 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.en.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.en.md @@ -347,9 +347,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -398,7 +399,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.md b/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.md index 56b47ce845..8715e7a351 100644 --- a/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.md +++ b/docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.md @@ -431,9 +431,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -482,7 +483,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.en.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.en.md index 54175e0eaa..fe36eb33f7 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.en.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.en.md @@ -264,9 +264,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the properties of the response body are as follows:
@@ -315,7 +316,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.md index 33a0563751..b7c91bad02 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_anomaly_detection.md @@ -268,9 +268,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -319,7 +320,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.en.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.en.md index 409a8bffe4..40ff892ebb 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.en.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.en.md @@ -229,9 +229,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -280,7 +281,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service:

+

Main operations provided by the service:

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.md index 6ee65c8654..a1eecdf28d 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_classification.md @@ -226,9 +226,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -277,7 +278,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.en.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.en.md index aac2accc45..bb850d7b24 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.en.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.en.md @@ -274,9 +274,10 @@ Below are the API references and multi-language service invocation examples:
API Reference -

For all operations provided by the service:

+

For main operations provided by the service:

    -
  • Both the response body and the request body for POST requests are JSON data (JSON objects).
    • +
  • The HTTP request method is POST.
    • +
  • The request body and the response body are both JSON data (JSON objects).
  • When the request is processed successfully, the response status code is 200, and the response body properties are as follows:
@@ -325,7 +326,7 @@ Below are the API references and multi-language service invocation examples:
-

Operations provided by the service are as follows:

+

Main operations provided by the service are as follows:

  • infer
diff --git a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.md b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.md index 3e14c09f7a..1e1d8af696 100644 --- a/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.md +++ b/docs/pipeline_usage/tutorials/time_series_pipelines/time_series_forecasting.md @@ -274,9 +274,10 @@ for res in output:
API参考 -

对于服务提供的所有操作：

+

对于服务提供的主要操作：

    -
  • 响应体以及POST请求的请求体均为JSON数据（JSON对象）。
    • +
  • HTTP请求方法为POST。
    • +
  • 请求体和响应体均为JSON数据（JSON对象）。
  • 当请求处理成功时，响应状态码为200，响应体的属性如下：
@@ -325,7 +326,7 @@ for res in output:
-

服务提供的操作如下：

+

服务提供的主要操作如下：

  • infer
diff --git a/paddlex/inference/pipelines/ppchatocrv3/ppchatocrv3.py b/paddlex/inference/pipelines/ppchatocrv3/ppchatocrv3.py index c207eb8ff2..4b9d558f98 100644 --- a/paddlex/inference/pipelines/ppchatocrv3/ppchatocrv3.py +++ b/paddlex/inference/pipelines/ppchatocrv3/ppchatocrv3.py @@ -240,7 +240,7 @@ def get_visual_result( img_info_list = list(self.img_reader(inputs))[0] elif isinstance(inputs, list): assert not any( - s.endswith(".pdf") for s in inputs + isinstance(s, str) and s.endswith(".pdf") for s in inputs ), "List containing pdf is not supported; only a list of images or a single PDF is supported." img_info_list = [x[0] for x in list(self.img_reader(inputs))] diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py b/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py index 998748c55f..9083e946b3 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py @@ -24,6 +24,7 @@ from ...formula_recognition import FormulaRecognitionPipeline from ...layout_parsing import LayoutParsingPipeline from ...ocr import OCRPipeline +from ...pp_shitu_v2 import ShiTuV2Pipeline from ...ppchatocrv3 import PPChatOCRPipeline from ...seal_recognition import SealOCRPipeline from ...single_model_pipeline import ( @@ -58,6 +59,7 @@ from .vehicle_attribute_recognition import ( create_pipeline_app as create_vehicle_attribute_recognition_app, ) +from .pp_shitu_v2 import create_pipeline_app as create_pp_shitu_v2_app from .ppchatocrv3 import create_pipeline_app as create_ppchatocrv3_app from .seal_recognition import create_pipeline_app as create_seal_recognition_app from .semantic_segmentation import ( @@ -180,6 +182,12 @@ def create_pipeline_app( "Expected `pipeline` to be an instance of `VehicleAttributeRecPipeline`." ) return create_vehicle_attribute_recognition_app(pipeline, app_config) + elif pipeline_name == "PP-ShiTuV2": + if not isinstance(pipeline, ShiTuV2Pipeline): + raise TypeError( + "Expected `pipeline` to be an instance of `ShiTuV2Pipeline`." + ) + return create_pp_shitu_v2_app(pipeline, app_config) else: if BasePipeline.get(pipeline_name): raise ValueError( diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/anomaly_detection.py b/paddlex/inference/pipelines/serving/_pipeline_apps/anomaly_detection.py index c9751ef2fb..2dd568b462 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/anomaly_detection.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/anomaly_detection.py @@ -60,8 +60,8 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: pred = result["pred"][0].tolist() size = [len(pred), len(pred[0])] label_map = [item for sublist in pred for item in sublist] - output_image_base64 = serving_utils.image_to_base64( - result.img.convert("RGB") + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img.convert("RGB")) ) return ResultResponse( diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/formula_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/formula_recognition.py index acaeb7bee3..ae139025f7 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/formula_recognition.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/formula_recognition.py @@ -88,7 +88,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: latex=latex, ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/image_classification.py b/paddlex/inference/pipelines/serving/_pipeline_apps/image_classification.py index fe01e7df3e..dd1a89fca6 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/image_classification.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/image_classification.py @@ -83,7 +83,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: zip(result["class_ids"], cat_names, result["scores"]), None, top_k ): categories.append(Category(id=id_, name=name, score=score)) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/instance_segmentation.py b/paddlex/inference/pipelines/serving/_pipeline_apps/instance_segmentation.py index c90ee714e4..d434a94129 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/instance_segmentation.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/instance_segmentation.py @@ -94,7 +94,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: mask=mask, ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/layout_parsing.py b/paddlex/inference/pipelines/serving/_pipeline_apps/layout_parsing.py index aa1744cc49..841663a0e2 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/layout_parsing.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/layout_parsing.py @@ -15,7 +15,6 @@ import os from typing import Final, List, Literal, Optional, Tuple -import cv2 import numpy as np from fastapi import FastAPI, HTTPException from numpy.typing import ArrayLike @@ -24,7 +23,7 @@ from .....utils import logging from ...layout_parsing import LayoutParsingPipeline -from .. import file_storage +from ..storage import SupportsGetURL, Storage, create_storage from .. import utils as serving_utils from ..app import AppConfig, create_app from ..models import Response, ResultResponse @@ -72,16 +71,17 @@ def _postprocess_image( img: ArrayLike, request_id: str, filename: str, - file_storage_config: file_storage.FileStorageConfig, + file_storage: Optional[Storage], ) -> str: key = f"{request_id}/{filename}" ext = os.path.splitext(filename)[1] img = np.asarray(img) - _, encoded_img = cv2.imencode(ext, img) - encoded_img = encoded_img.tobytes() - return file_storage.postprocess_file( - encoded_img, config=file_storage_config, key=key - ) + img_bytes = serving_utils.image_array_to_bytes(img, ext=ext) + if file_storage is not None: + file_storage.set(key, img_bytes) + if isinstance(file_storage, SupportsGetURL): + return file_storage.get_url(key) + return serving_utils.base64_encode(img_bytes) def create_pipeline_app( @@ -91,12 +91,10 @@ def create_pipeline_app( pipeline=pipeline, app_config=app_config, app_aiohttp_session=True ) - if "file_storage_config" in ctx.extra: - ctx.extra["file_storage_config"] = file_storage.parse_file_storage_config( - ctx.extra["file_storage_config"] - ) + if ctx.config.extra and "file_storage" in ctx.config.extra: + ctx.extra["file_storage"] = create_storage(ctx.config.extra["file_storage"]) else: - ctx.extra["file_storage_config"] = file_storage.InMemoryStorageConfig() + ctx.extra["file_storage"] = None ctx.extra.setdefault("max_img_size", _DEFAULT_MAX_IMG_SIZE) ctx.extra.setdefault("max_num_imgs", _DEFAULT_MAX_NUM_IMGS) @@ -172,7 +170,7 @@ async def _infer( subitem[label]["img"], request_id=request_id, filename=f"image_{i}_{j}.jpg", - file_storage_config=ctx.extra["file_storage_config"], + file_storage=ctx.extra["file_storage"], ) text = subitem[label]["image_text"] else: diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/multi_label_image_classification.py b/paddlex/inference/pipelines/serving/_pipeline_apps/multi_label_image_classification.py index 5f357c244c..59cf4c9f29 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/multi_label_image_classification.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/multi_label_image_classification.py @@ -72,7 +72,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: result["class_ids"], cat_names, result["scores"] ): categories.append(Category(id=id_, name=name, score=score)) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/object_detection.py b/paddlex/inference/pipelines/serving/_pipeline_apps/object_detection.py index e758b17c73..68c039015d 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/object_detection.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/object_detection.py @@ -72,7 +72,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: score=obj["score"], ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/ocr.py b/paddlex/inference/pipelines/serving/_pipeline_apps/ocr.py index 7aa194abaa..d8f83d5171 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/ocr.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/ocr.py @@ -80,7 +80,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: result["dt_polys"], result["rec_text"], result["rec_score"] ): texts.append(Text(poly=poly, text=text, score=score)) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/pedestrian_attribute_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/pedestrian_attribute_recognition.py index 1b4a2eae60..8f4c42fa2a 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/pedestrian_attribute_recognition.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/pedestrian_attribute_recognition.py @@ -84,7 +84,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: score=obj["det_score"], ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py b/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py new file mode 100644 index 0000000000..b17f7d7ec5 --- /dev/null +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py @@ -0,0 +1,313 @@ +# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +import asyncio +import faiss +import pickle +from typing import Dict, List, Optional + +from fastapi import FastAPI, HTTPException +from pydantic import BaseModel, Field +from typing_extensions import Annotated, TypeAlias + +from .....utils import logging +from ....components.retrieval.faiss import IndexData +from ...pp_shitu_v2 import ShiTuV2Pipeline +from ..storage import create_storage +from .. import utils as serving_utils +from ..app import AppConfig, create_app +from ..models import Response, ResultResponse + + +class ImageLabelPair(BaseModel): + image: str + label: str + + +class BuildIndexRequest(BaseModel): + imageLabelPairs: List[ImageLabelPair] + + +class BuildIndexResult(BaseModel): + indexKey: str + idMap: Dict[int, str] + + +class AddImagesToIndexRequest(BaseModel): + imageLabelPairs: List[ImageLabelPair] + indexKey: str + + +class AddImagesToIndexResult(BaseModel): + idMap: Dict[int, str] + + +class RemoveImagesFromIndexRequest(BaseModel): + ids: List[int] + indexKey: str + + +class RemoveImagesFromIndexResult(BaseModel): + idMap: Dict[int, str] + + +class InferRequest(BaseModel): + image: str + indexKey: Optional[str] = None + + +BoundingBox: TypeAlias = Annotated[List[float], Field(min_length=4, max_length=4)] + + +class RecResult(BaseModel): + label: str + score: float + + +class DetectedObject(BaseModel): + bbox: BoundingBox + recResults: List[RecResult] + score: float + + +class InferResult(BaseModel): + detectedObjects: List[DetectedObject] + image: str + + +# XXX: I have to implement serialization and deserialization functions myself, +# which is fragile. +def _serialize_index_data(index_data: IndexData) -> bytes: + tup = (index_data.index_bytes, index_data.index_info) + return pickle.dumps(tup) + + +def _deserialize_index_data(index_data_bytes: bytes) -> IndexData: + tup = pickle.loads(index_data_bytes) + index = faiss.deserialize_index(tup[0]) + return IndexData(index, tup[1]) + + +def create_pipeline_app(pipeline: ShiTuV2Pipeline, app_config: AppConfig) -> FastAPI: + app, ctx = create_app( + pipeline=pipeline, app_config=app_config, app_aiohttp_session=True + ) + + if ctx.config.extra and "index_storage" in ctx.config.extra: + ctx.extra["index_storage"] = create_storage(ctx.config.extra["index_storage"]) + else: + ctx.extra["index_storage"] = create_storage({"type": "memory"}) + + @app.post( + "/shitu-index-build", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _build_index( + request: BuildIndexRequest, + ) -> ResultResponse[BuildIndexResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + request_id = serving_utils.generate_request_id() + + try: + images = [pair.image for pair in request.imageLabelPairs] + file_bytes_list = await asyncio.gather( + *(serving_utils.get_raw_bytes(img, aiohttp_session) for img in images) + ) + images = [ + serving_utils.image_bytes_to_array(item) for item in file_bytes_list + ] + labels = [pair.label for pair in request.imageLabelPairs] + + index_data = await pipeline.call( + pipeline.pipeline.build_index, images, labels + ) + + index_storage = ctx.extra["index_storage"] + index_key = request_id + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, index_key, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=BuildIndexResult(indexKey=index_key, idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/shitu-index-add", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _add_images_to_index( + request: AddImagesToIndexRequest, + ) -> ResultResponse[AddImagesToIndexResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + try: + images = [pair.image for pair in request.imageLabelPairs] + file_bytes_list = await asyncio.gather( + *(serving_utils.get_raw_bytes(img, aiohttp_session) for img in images) + ) + images = [ + serving_utils.image_bytes_to_array(item) for item in file_bytes_list + ] + labels = [pair.label for pair in request.imageLabelPairs] + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + + index_data = await pipeline.call( + pipeline.pipeline.append_index, images, labels, index_data + ) + + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, request.indexKey, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=AddImagesToIndexResult(idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/shitu-index-remove", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _remove_images_from_index( + request: RemoveImagesFromIndexRequest, + ) -> ResultResponse[RemoveImagesFromIndexResult]: + pipeline = ctx.pipeline + + try: + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + + index_data = await pipeline.call( + pipeline.pipeline.remove_index, request.ids, index_data + ) + + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, request.indexKey, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=RemoveImagesFromIndexResult(idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/shitu-infer", + operation_id="infer", + responses={422: {"model": Response}}, + ) + async def _infer(request: InferRequest) -> ResultResponse[InferResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + try: + image_bytes = await serving_utils.get_raw_bytes( + request.image, aiohttp_session + ) + image = serving_utils.image_bytes_to_array(image_bytes) + + if request.indexKey is not None: + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + else: + index_data = None + + result = list( + await pipeline.call(pipeline.pipeline.predict, image, index=index_data) + )[0] + + objects: List[DetectedObject] = [] + for obj in result["boxes"]: + rec_results: List[RecResult] = [] + if obj["rec_scores"] is not None: + for label, score in zip(obj["labels"], obj["rec_scores"]): + rec_results.append( + RecResult( + label=label, + score=score, + ) + ) + objects.append( + DetectedObject( + bbox=obj["coordinate"], + recResults=rec_results, + score=obj["det_score"], + ) + ) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=InferResult(detectedObjects=objects, image=output_image_base64), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + return app diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/ppchatocrv3.py b/paddlex/inference/pipelines/serving/_pipeline_apps/ppchatocrv3.py index 87f2241ba4..747652478d 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/ppchatocrv3.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/ppchatocrv3.py @@ -16,7 +16,6 @@ import os from typing import Awaitable, Final, List, Literal, Optional, Tuple, Union -import cv2 import numpy as np from fastapi import FastAPI, HTTPException from numpy.typing import ArrayLike @@ -26,7 +25,7 @@ from .....utils import logging from .... import results from ...ppchatocrv3 import PPChatOCRPipeline -from .. import file_storage +from ..storage import SupportsGetURL, Storage, create_storage from .. import utils as serving_utils from ..app import AppConfig, create_app from ..models import Response, ResultResponse @@ -169,16 +168,17 @@ def _postprocess_image( img: ArrayLike, request_id: str, filename: str, - file_storage_config: file_storage.FileStorageConfig, + file_storage: Optional[Storage], ) -> str: key = f"{request_id}/{filename}" ext = os.path.splitext(filename)[1] img = np.asarray(img) - _, encoded_img = cv2.imencode(ext, img) - encoded_img = encoded_img.tobytes() - return file_storage.postprocess_file( - encoded_img, config=file_storage_config, key=key - ) + img_bytes = serving_utils.image_array_to_bytes(img, ext=ext) + if file_storage is not None: + file_storage.set(key, img_bytes) + if isinstance(file_storage, SupportsGetURL): + return file_storage.get_url(key) + return serving_utils.base64_encode(img_bytes) def create_pipeline_app(pipeline: PPChatOCRPipeline, app_config: AppConfig) -> FastAPI: @@ -186,12 +186,10 @@ def create_pipeline_app(pipeline: PPChatOCRPipeline, app_config: AppConfig) -> F pipeline=pipeline, app_config=app_config, app_aiohttp_session=True ) - if "file_storage_config" in ctx.extra: - ctx.extra["file_storage_config"] = file_storage.parse_file_storage_config( - ctx.extra["file_storage_config"] - ) + if ctx.config.extra and "file_storage" in ctx.config.extra: + ctx.extra["file_storage"] = create_storage(ctx.config.extra["file_storage"]) else: - ctx.extra["file_storage_config"] = file_storage.InMemoryStorageConfig() + ctx.extra["file_storage"] = None ctx.extra.setdefault("max_img_size", _DEFAULT_MAX_IMG_SIZE) ctx.extra.setdefault("max_num_imgs", _DEFAULT_MAX_NUM_IMGS) @@ -259,7 +257,7 @@ async def _analyze_images( img, request_id=request_id, filename=f"input_image_{i}.jpg", - file_storage_config=ctx.extra["file_storage_config"], + file_storage=ctx.extra["file_storage"], ) pp_img_futures.append(future) future = serving_utils.call_async( @@ -267,7 +265,7 @@ async def _analyze_images( item["ocr_result"].img, request_id=request_id, filename=f"ocr_image_{i}.jpg", - file_storage_config=ctx.extra["file_storage_config"], + file_storage=ctx.extra["file_storage"], ) pp_img_futures.append(future) future = serving_utils.call_async( @@ -275,7 +273,7 @@ async def _analyze_images( item["layout_result"].img, request_id=request_id, filename=f"layout_image_{i}.jpg", - file_storage_config=ctx.extra["file_storage_config"], + file_storage=ctx.extra["file_storage"], ) pp_img_futures.append(future) texts: List[Text] = [] diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/seal_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/seal_recognition.py index 9a746a7955..7d9ec20fcc 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/seal_recognition.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/seal_recognition.py @@ -89,8 +89,8 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: ): texts.append(Text(poly=poly, text=text, score=score)) seal_impressions.append(SealImpression(texts=texts)) - layout_image_base64 = serving_utils.image_to_base64( - result["layout_result"].img + layout_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result["layout_result"].img) ) # TODO: OCR image diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/semantic_segmentation.py b/paddlex/inference/pipelines/serving/_pipeline_apps/semantic_segmentation.py index 5a412475e2..7870eecbb7 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/semantic_segmentation.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/semantic_segmentation.py @@ -62,8 +62,8 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: pred = result["pred"][0].tolist() size = [len(pred), len(pred[0])] label_map = [item for sublist in pred for item in sublist] - output_image_base64 = serving_utils.image_to_base64( - result.img.convert("RGB") + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img.convert("RGB")) ) return ResultResponse( diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py b/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py index b6c5c5288c..fe67240312 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py @@ -72,7 +72,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: score=obj["score"], ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/table_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/table_recognition.py index 32c5c9008f..36e221efff 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/table_recognition.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/table_recognition.py @@ -85,10 +85,12 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: html=item["html"], ) ) - layout_image_base64 = serving_utils.image_to_base64( - result["layout_result"].img + layout_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result["layout_result"].img) + ) + ocr_iamge_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result["ocr_result"].img) ) - ocr_iamge_base64 = serving_utils.image_to_base64(result["ocr_result"].img) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/ts_ad.py b/paddlex/inference/pipelines/serving/_pipeline_apps/ts_ad.py index bc36c55f66..cd5710d71f 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/ts_ad.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/ts_ad.py @@ -50,7 +50,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: result = (await pipeline.infer(df))[0] - output_csv = serving_utils.data_frame_to_base64(result["anomaly"]) + output_csv = serving_utils.base64_encode( + serving_utils.data_frame_to_bytes(result["anomaly"]) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/ts_fc.py b/paddlex/inference/pipelines/serving/_pipeline_apps/ts_fc.py index d840184516..a7cb80c368 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/ts_fc.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/ts_fc.py @@ -50,7 +50,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: result = (await pipeline.infer(df))[0] - output_csv = serving_utils.data_frame_to_base64(result["forecast"]) + output_csv = serving_utils.base64_encode( + serving_utils.data_frame_to_bytes(result["forecast"]) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/vehicle_attribute_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/vehicle_attribute_recognition.py index db59313edb..0c9f485982 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/vehicle_attribute_recognition.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/vehicle_attribute_recognition.py @@ -84,7 +84,9 @@ async def _infer(request: InferRequest) -> ResultResponse[InferResult]: score=obj["det_score"], ) ) - output_image_base64 = serving_utils.image_to_base64(result.img) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) return ResultResponse( logId=serving_utils.generate_log_id(), diff --git a/paddlex/inference/pipelines/serving/app.py b/paddlex/inference/pipelines/serving/app.py index 96b3eb105f..bbe44ddec6 100644 --- a/paddlex/inference/pipelines/serving/app.py +++ b/paddlex/inference/pipelines/serving/app.py @@ -131,6 +131,7 @@ async def _app_lifespan(app: fastapi.FastAPI) -> AsyncGenerator[None, None]: app = fastapi.FastAPI(lifespan=_app_lifespan) ctx = AppContext[_PipelineT](config=app_config) + app.state.context = ctx @app.get("/health", operation_id="checkHealth") async def _check_health() -> Response: diff --git a/paddlex/inference/pipelines/serving/file_storage.py b/paddlex/inference/pipelines/serving/file_storage.py deleted file mode 100644 index 9104afb45e..0000000000 --- a/paddlex/inference/pipelines/serving/file_storage.py +++ /dev/null @@ -1,80 +0,0 @@ -# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -import base64 -import uuid -from typing import Any, Dict, Literal, Optional, Union - -from baidubce.auth.bce_credentials import BceCredentials -from baidubce.bce_client_configuration import BceClientConfiguration -from baidubce.services.bos.bos_client import BosClient -from pydantic import BaseModel, Discriminator, SecretStr, TypeAdapter -from typing_extensions import Annotated, assert_never - - -class InMemoryStorageConfig(BaseModel): - type: Literal["memory"] = "memory" - - -class BOSConfig(BaseModel): - endpoint: str - ak: SecretStr - sk: SecretStr - bucket_name: str - key_prefix: Optional[str] = None - connection_timeout_in_mills: Optional[int] = None - - type: Literal["bos"] = "bos" - - -FileStorageConfig = Union[InMemoryStorageConfig, BOSConfig] - - -def parse_file_storage_config(dic: Dict[str, Any]) -> FileStorageConfig: - # XXX: mypy deduces a wrong type - return TypeAdapter( - Annotated[FileStorageConfig, Discriminator("type")] - ).validate_python( - dic - ) # type: ignore[return-value] - - -def postprocess_file( - file: bytes, config: FileStorageConfig, key: Optional[str] = None -) -> str: - if config.type == "memory": - return base64.b64encode(file).decode("ascii") - elif config.type == "bos": - # TODO: Currently BOS clients are created on the fly since they are not - # thread-safe. Should we use a background thread with a queue or use a - # dedicated thread? - bos_cfg = BceClientConfiguration( - credentials=BceCredentials( - config.ak.get_secret_value(), config.sk.get_secret_value() - ), - endpoint=config.endpoint, - connection_timeout_in_mills=config.connection_timeout_in_mills, - ) - client = BosClient(bos_cfg) - if key is None: - key = str(uuid.uuid4()) - if config.key_prefix: - key = f"{config.key_prefix}{key}" - client.put_object_from_string(bucket=config.bucket_name, key=key, data=file) - url = client.generate_pre_signed_url( - config.bucket_name, key, expiration_in_seconds=-1 - ).decode("ascii") - return url - else: - assert_never(config.type) diff --git a/paddlex/inference/pipelines/serving/storage.py b/paddlex/inference/pipelines/serving/storage.py new file mode 100644 index 0000000000..544e3da8f5 --- /dev/null +++ b/paddlex/inference/pipelines/serving/storage.py @@ -0,0 +1,161 @@ +# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +import abc +from os import PathLike +from pathlib import Path +from typing import Any, Dict, Literal, Optional, Union, Protocol, runtime_checkable + +from baidubce.auth.bce_credentials import BceCredentials +from baidubce.bce_client_configuration import BceClientConfiguration +from baidubce.services.bos.bos_client import BosClient +from pydantic import BaseModel, Discriminator, SecretStr, TypeAdapter +from typing_extensions import Annotated, assert_never + + +class InMemoryStorageConfig(BaseModel): + type: Literal["memory"] = "memory" + + +class FileSystemStorageConfig(BaseModel): + directory: Union[str, PathLike] + + type: Literal["file_system"] = "file_system" + + +class BOSConfig(BaseModel): + endpoint: str + ak: SecretStr + sk: SecretStr + bucket_name: str + key_prefix: Optional[str] = None + connection_timeout_in_mills: Optional[int] = None + + type: Literal["bos"] = "bos" + + +FileStorageConfig = Union[InMemoryStorageConfig, FileSystemStorageConfig, BOSConfig] + + +@runtime_checkable +class SupportsGetURL(Protocol): + def get_url(self, key: str) -> str: ... + + +class Storage(metaclass=abc.ABCMeta): + @abc.abstractmethod + def get(self, key: str) -> bytes: + raise NotImplementedError + + @abc.abstractmethod + def set(self, key: str, value: bytes) -> None: + raise NotImplementedError + + @abc.abstractmethod + def delete(self, key: str) -> None: + raise NotImplementedError + + +class InMemoryStorage(Storage): + def __init__(self, config: InMemoryStorageConfig) -> None: + super().__init__() + self._data: Dict[str, bytes] = {} + + def get(self, key: str) -> bytes: + return self._data[key] + + def set(self, key: str, value: bytes) -> None: + self._data[key] = value + + def delete(self, key: str) -> None: + del self._data[key] + + +class FileSystemStorage(Storage): + def __init__(self, config: FileSystemStorageConfig) -> None: + super().__init__() + self._directory = Path(config.directory) + self._directory.mkdir(exist_ok=True) + + def get(self, key: str) -> bytes: + with open(self._get_file_path(key), "rb") as f: + contents = f.read() + return contents + + def set(self, key: str, value: bytes) -> None: + path = self._get_file_path(key) + path.parent.mkdir(exist_ok=True) + with open(path, "wb") as f: + f.write(value) + + def delete(self, key: str) -> None: + file_path = self._get_file_path(key) + file_path.unlink(missing_ok=True) + + def _get_file_path(self, key: str) -> Path: + return self._directory / key + + +class BOS(Storage): + def __init__(self, config: BOSConfig) -> None: + super().__init__() + bos_cfg = BceClientConfiguration( + credentials=BceCredentials( + config.ak.get_secret_value(), config.sk.get_secret_value() + ), + endpoint=config.endpoint, + connection_timeout_in_mills=config.connection_timeout_in_mills, + ) + self._client = BosClient(bos_cfg) + self._bucket_name = config.bucket_name + self._key_prefix = config.key_prefix + + def get(self, key: str) -> bytes: + key = self._get_full_key(key) + return self._client.get_object_as_string(bucket_name=self._bucket_name, key=key) + + def set(self, key: str, value: bytes) -> None: + key = self._get_full_key(key) + self._client.put_object_from_string( + bucket=self._bucket_name, key=key, data=value + ) + + def delete(self, key: str) -> None: + key = self._get_full_key(key) + self._client.delete_object(bucket_name=self._bucket_name, key=key) + + def get_url(self, key: str) -> str: + key = self._get_full_key(key) + return self._client.generate_pre_signed_url( + self._bucket_name, key, expiration_in_seconds=-1 + ).decode("ascii") + + def _get_full_key(self, key: str) -> str: + if self._key_prefix: + return f"{self._key_prefix}/{key}" + return key + + +def create_storage(dic: Dict[str, Any], /) -> Storage: + config = TypeAdapter( + Annotated[FileStorageConfig, Discriminator("type")] + ).validate_python(dic) + if config.type == "memory": + return InMemoryStorage(config) + elif config.type == "file_system": + return FileSystemStorage(config) + elif config.type == "bos": + return BOS(config) + else: + assert_never(config) diff --git a/paddlex/inference/pipelines/serving/utils.py b/paddlex/inference/pipelines/serving/utils.py index 32b2bdd7a1..f960f3e63a 100644 --- a/paddlex/inference/pipelines/serving/utils.py +++ b/paddlex/inference/pipelines/serving/utils.py @@ -101,11 +101,20 @@ def image_bytes_to_array(data: bytes) -> np.ndarray: return cv2.imdecode(np.frombuffer(data, np.uint8), cv2.IMREAD_COLOR) -def image_to_base64(image: Image.Image) -> str: +def image_bytes_to_image(data: bytes) -> Image.Image: + return Image.open(io.BytesIO(data)) + + +def image_to_bytes(image: Image.Image, format: str = "JPEG") -> bytes: with io.BytesIO() as f: - image.save(f, format="JPEG") - image_base64 = base64.b64encode(f.getvalue()).decode("ascii") - return image_base64 + image.save(f, format=format) + img_bytes = f.getvalue() + return img_bytes + + +def image_array_to_bytes(image: np.ndarray, ext: str = ".jpg") -> str: + image = cv2.imencode(ext, image)[1] + return image.tobytes() def csv_bytes_to_data_frame(data: bytes) -> pd.DataFrame: @@ -114,8 +123,12 @@ def csv_bytes_to_data_frame(data: bytes) -> pd.DataFrame: return df -def data_frame_to_base64(df: str) -> str: - return base64.b64encode(df.to_csv().encode("utf-8")).decode("ascii") +def data_frame_to_bytes(df: str) -> str: + return df.to_csv().encode("utf-8") + + +def base64_encode(data: bytes) -> str: + return base64.b64encode(data).decode("ascii") def read_pdf( From 27adcdeee87ddc0d2b3b287ebc81fa04864bee69 Mon Sep 17 00:00:00 2001 From: Bobholamovic Date: Fri, 15 Nov 2024 14:54:17 +0800 Subject: [PATCH 2/5] Fix index_type and metric_type --- .../general_image_recognition.en.md | 18 ++++++++++++++---- .../cv_pipelines/general_image_recognition.md | 10 ++++++++++ .../serving/_pipeline_apps/pp_shitu_v2.py | 8 +++++++- 3 files changed, 31 insertions(+), 5 deletions(-) diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md index 3093c219a1..d2aed56ebe 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.en.md @@ -738,13 +738,14 @@ import requests API_BASE_URL = "http://0.0.0.0:8080" base_image_label_pairs = [ - {"image": "./demo0.jpg", "label": "rabbit"}, - {"image": "./demo1.jpg", "label": "rabbit"}, - {"image": "./demo2.jpg", "label": "puppy"}, + {"image": "./demo0.jpg", "label": "兔子"}, + {"image": "./demo1.jpg", "label": "兔子"}, + {"image": "./demo2.jpg", "label": "小狗"}, ] image_label_pairs_to_add = [ - {"image": "./demo3.jpg", "label": "puppy"}, + {"image": "./demo3.jpg", "label": "小狗"}, ] +ids_to_remove = [1] infer_image_path = "./demo4.jpg" output_image_path = "./out.jpg" @@ -778,6 +779,15 @@ if resp_index_add.status_code != 200: result_index_add = resp_index_add.json()["result"] print(f"Number of images indexed: {len(result_index_add['idMap'])}") +payload = {"ids": ids_to_remove, "indexKey": result_index_build["indexKey"]} +resp_index_remove = requests.post(f"{API_BASE_URL}/shitu-index-remove", json=payload) +if resp_index_remove.status_code != 200: + print(f"Request to shitu-index-remove failed with status code {resp_index_remove}.") + pprint.pp(resp_index_remove.json()) + sys.exit(1) +result_index_remove = resp_index_remove.json()["result"] +print(f"Number of images indexed: {len(result_index_remove['idMap'])}") + with open(infer_image_path, "rb") as file: image_bytes = file.read() image_data = base64.b64encode(image_bytes).decode("ascii") diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md index 961a83873d..b27ae78244 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/general_image_recognition.md @@ -775,6 +775,7 @@ base_image_label_pairs = [ image_label_pairs_to_add = [ {"image": "./demo3.jpg", "label": "小狗"}, ] +ids_to_remove = [1] infer_image_path = "./demo4.jpg" output_image_path = "./out.jpg" @@ -808,6 +809,15 @@ if resp_index_add.status_code != 200: result_index_add = resp_index_add.json()["result"] print(f"Number of images indexed: {len(result_index_add['idMap'])}") +payload = {"ids": ids_to_remove, "indexKey": result_index_build["indexKey"]} +resp_index_remove = requests.post(f"{API_BASE_URL}/shitu-index-remove", json=payload) +if resp_index_remove.status_code != 200: + print(f"Request to shitu-index-remove failed with status code {resp_index_remove}.") + pprint.pp(resp_index_remove.json()) + sys.exit(1) +result_index_remove = resp_index_remove.json()["result"] +print(f"Number of images indexed: {len(result_index_remove['idMap'])}") + with open(infer_image_path, "rb") as file: image_bytes = file.read() image_data = base64.b64encode(image_bytes).decode("ascii") diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py b/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py index b17f7d7ec5..84bae79207 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/pp_shitu_v2.py @@ -132,8 +132,14 @@ async def _build_index( ] labels = [pair.label for pair in request.imageLabelPairs] + # TODO: Support specifying `index_type` and `metric_type` in the + # request index_data = await pipeline.call( - pipeline.pipeline.build_index, images, labels + pipeline.pipeline.build_index, + images, + labels, + index_type="Flat", + metric_type="IP", ) index_storage = ctx.extra["index_storage"] From e235a2b336d4f659e42a50226085513b25d2a9a3 Mon Sep 17 00:00:00 2001 From: Bobholamovic Date: Fri, 15 Nov 2024 16:47:19 +0800 Subject: [PATCH 3/5] Add face recognition app --- .../serving/_pipeline_apps/__init__.py | 8 + .../_pipeline_apps/face_recognition.py | 317 ++++++++++++++++++ 2 files changed, 325 insertions(+) create mode 100644 paddlex/inference/pipelines/serving/_pipeline_apps/face_recognition.py diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py b/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py index 9083e946b3..2ef035becd 100644 --- a/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/__init__.py @@ -21,6 +21,7 @@ VehicleAttributeRecPipeline, ) from ...base import BasePipeline +from ...face_recognition import FaceRecPipeline from ...formula_recognition import FormulaRecognitionPipeline from ...layout_parsing import LayoutParsingPipeline from ...ocr import OCRPipeline @@ -42,6 +43,7 @@ from ...table_recognition import TableRecPipeline from ..app import create_app_config from .anomaly_detection import create_pipeline_app as create_anomaly_detection_app +from .face_recognition import create_pipeline_app as create_face_recognition_app from .formula_recognition import create_pipeline_app as create_formula_recognition_app from .layout_parsing import create_pipeline_app as create_layout_parsing_app from .image_classification import create_pipeline_app as create_image_classification_app @@ -188,6 +190,12 @@ def create_pipeline_app( "Expected `pipeline` to be an instance of `ShiTuV2Pipeline`." ) return create_pp_shitu_v2_app(pipeline, app_config) + elif pipeline_name == "face_recognition": + if not isinstance(pipeline, FaceRecPipeline): + raise TypeError( + "Expected `pipeline` to be an instance of `FaceRecPipeline`." + ) + return create_face_recognition_app(pipeline, app_config) else: if BasePipeline.get(pipeline_name): raise ValueError( diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/face_recognition.py b/paddlex/inference/pipelines/serving/_pipeline_apps/face_recognition.py new file mode 100644 index 0000000000..b7a47aae30 --- /dev/null +++ b/paddlex/inference/pipelines/serving/_pipeline_apps/face_recognition.py @@ -0,0 +1,317 @@ +# copyright (c) 2024 PaddlePaddle Authors. All Rights Reserve. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +import asyncio +import faiss +import pickle +from typing import Dict, List, Optional + +from fastapi import FastAPI, HTTPException +from pydantic import BaseModel, Field +from typing_extensions import Annotated, TypeAlias + +from .....utils import logging +from ....components.retrieval.faiss import IndexData +from ...face_recognition import FaceRecPipeline +from ..storage import create_storage +from .. import utils as serving_utils +from ..app import AppConfig, create_app +from ..models import Response, ResultResponse + + +class ImageLabelPair(BaseModel): + image: str + label: str + + +class BuildIndexRequest(BaseModel): + imageLabelPairs: List[ImageLabelPair] + + +class BuildIndexResult(BaseModel): + indexKey: str + idMap: Dict[int, str] + + +class AddImagesToIndexRequest(BaseModel): + imageLabelPairs: List[ImageLabelPair] + indexKey: str + + +class AddImagesToIndexResult(BaseModel): + idMap: Dict[int, str] + + +class RemoveImagesFromIndexRequest(BaseModel): + ids: List[int] + indexKey: str + + +class RemoveImagesFromIndexResult(BaseModel): + idMap: Dict[int, str] + + +class InferRequest(BaseModel): + image: str + indexKey: Optional[str] = None + + +BoundingBox: TypeAlias = Annotated[List[float], Field(min_length=4, max_length=4)] + + +class RecResult(BaseModel): + label: str + score: float + + +class Face(BaseModel): + bbox: BoundingBox + recResults: List[RecResult] + score: float + + +class InferResult(BaseModel): + faces: List[Face] + image: str + + +def _serialize_index_data(index_data: IndexData) -> bytes: + tup = (index_data.index_bytes, index_data.index_info) + return pickle.dumps(tup) + + +def _deserialize_index_data(index_data_bytes: bytes) -> IndexData: + tup = pickle.loads(index_data_bytes) + index = faiss.deserialize_index(tup[0]) + return IndexData(index, tup[1]) + + +def create_pipeline_app(pipeline: FaceRecPipeline, app_config: AppConfig) -> FastAPI: + app, ctx = create_app( + pipeline=pipeline, app_config=app_config, app_aiohttp_session=True + ) + + if ctx.config.extra and "index_storage" in ctx.config.extra: + ctx.extra["index_storage"] = create_storage(ctx.config.extra["index_storage"]) + else: + ctx.extra["index_storage"] = create_storage({"type": "memory"}) + + @app.post( + "/face-recognition-index-build", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _build_index( + request: BuildIndexRequest, + ) -> ResultResponse[BuildIndexResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + request_id = serving_utils.generate_request_id() + + try: + images = [pair.image for pair in request.imageLabelPairs] + file_bytes_list = await asyncio.gather( + *(serving_utils.get_raw_bytes(img, aiohttp_session) for img in images) + ) + images = [ + serving_utils.image_bytes_to_array(item) for item in file_bytes_list + ] + labels = [pair.label for pair in request.imageLabelPairs] + + # TODO: Support specifying `index_type` and `metric_type` in the + # request + index_data = await pipeline.call( + pipeline.pipeline.build_index, + images, + labels, + index_type="Flat", + metric_type="IP", + ) + + index_storage = ctx.extra["index_storage"] + index_key = request_id + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, index_key, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=BuildIndexResult(indexKey=index_key, idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/face-recognition-index-add", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _add_images_to_index( + request: AddImagesToIndexRequest, + ) -> ResultResponse[AddImagesToIndexResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + try: + images = [pair.image for pair in request.imageLabelPairs] + file_bytes_list = await asyncio.gather( + *(serving_utils.get_raw_bytes(img, aiohttp_session) for img in images) + ) + images = [ + serving_utils.image_bytes_to_array(item) for item in file_bytes_list + ] + labels = [pair.label for pair in request.imageLabelPairs] + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + + index_data = await pipeline.call( + pipeline.pipeline.append_index, images, labels, index_data + ) + + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, request.indexKey, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=AddImagesToIndexResult(idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/face-recognition-index-remove", + operation_id="buildIndex", + responses={422: {"model": Response}}, + ) + async def _remove_images_from_index( + request: RemoveImagesFromIndexRequest, + ) -> ResultResponse[RemoveImagesFromIndexResult]: + pipeline = ctx.pipeline + + try: + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + + index_data = await pipeline.call( + pipeline.pipeline.remove_index, request.ids, index_data + ) + + index_data_bytes = await serving_utils.call_async( + _serialize_index_data, index_data + ) + await serving_utils.call_async( + index_storage.set, request.indexKey, index_data_bytes + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=RemoveImagesFromIndexResult(idMap=index_data.id_map), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + @app.post( + "/face-recognition-infer", + operation_id="infer", + responses={422: {"model": Response}}, + ) + async def _infer(request: InferRequest) -> ResultResponse[InferResult]: + pipeline = ctx.pipeline + aiohttp_session = ctx.aiohttp_session + + try: + image_bytes = await serving_utils.get_raw_bytes( + request.image, aiohttp_session + ) + image = serving_utils.image_bytes_to_array(image_bytes) + + if request.indexKey is not None: + index_storage = ctx.extra["index_storage"] + index_data_bytes = await serving_utils.call_async( + index_storage.get, request.indexKey + ) + index_data = await serving_utils.call_async( + _deserialize_index_data, index_data_bytes + ) + else: + index_data = None + + result = list( + await pipeline.call(pipeline.pipeline.predict, image, index=index_data) + )[0] + + faces: List[Face] = [] + for face in result["boxes"]: + rec_results: List[RecResult] = [] + if face["rec_scores"] is not None: + for label, score in zip(face["labels"], face["rec_scores"]): + rec_results.append( + RecResult( + label=label, + score=score, + ) + ) + faces.append( + Face( + bbox=face["coordinate"], + recResults=rec_results, + score=face["det_score"], + ) + ) + output_image_base64 = serving_utils.base64_encode( + serving_utils.image_to_bytes(result.img) + ) + + return ResultResponse( + logId=serving_utils.generate_log_id(), + errorCode=0, + errorMsg="Success", + result=InferResult(faces=faces, image=output_image_base64), + ) + + except Exception as e: + logging.exception(e) + raise HTTPException(status_code=500, detail="Internal server error") + + return app From 886f5e414badde333e2017e48e2c89f9dba66b50 Mon Sep 17 00:00:00 2001 From: Bobholamovic Date: Fri, 15 Nov 2024 16:47:56 +0800 Subject: [PATCH 4/5] Change small object detection endpoint --- .../cv_pipelines/small_object_detection.en.md | 16 ++++++++-------- .../cv_pipelines/small_object_detection.md | 16 ++++++++-------- .../_pipeline_apps/small_object_detection.py | 4 +++- 3 files changed, 19 insertions(+), 17 deletions(-) diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md index ba5db67ce7..0e17a3c965 100644 --- a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md +++ b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.en.md @@ -297,7 +297,7 @@ Below are the API references and multi-language service invocation examples:
  • infer

    • Performs object detection on an image.

    -

    POST /object-detection

    +

    POST /small-object-detection

    • The request body properties are as follows:
    @@ -407,7 +407,7 @@ Below are the API references and multi-language service invocation examples: 
    import base64
 import requests
 
-API_URL = "http://localhost:8080/object-detection"
+API_URL = "http://localhost:8080/small-object-detection"
 image_path = "./demo.jpg"
 output_image_path = "./out.jpg"
 
@@ -460,7 +460,7 @@ int main() {
     jsonObj["image"] = encodedImage;
     std::string body = jsonObj.dump();
 
-    auto response = client.Post("/object-detection", headers, body, "application/json");
+    auto response = client.Post("/small-object-detection", headers, body, "application/json");
     if (response && response->status == 200) {
         nlohmann::json jsonResponse = nlohmann::json::parse(response->body);
         auto result = jsonResponse["result"];
@@ -505,7 +505,7 @@ import java.util.Base64;
 
 public class Main {
     public static void main(String[] args) throws IOException {
-        String API_URL = "http://localhost:8080/object-detection";
+        String API_URL = "http://localhost:8080/small-object-detection";
         String imagePath = "./demo.jpg";
         String outputImagePath = "./out.jpg";
 
@@ -561,7 +561,7 @@ import (
 )
 
 func main() {
-    API_URL := "http://localhost:8080/object-detection"
+    API_URL := "http://localhost:8080/small-object-detection"
     imagePath := "./demo.jpg"
     outputImagePath := "./out.jpg"
 
@@ -641,7 +641,7 @@ using Newtonsoft.Json.Linq;
 
 class Program
 {
-    static readonly string API_URL = "http://localhost:8080/object-detection";
+    static readonly string API_URL = "http://localhost:8080/small-object-detection";
     static readonly string imagePath = "./demo.jpg";
     static readonly string outputImagePath = "./out.jpg";
 
@@ -677,7 +677,7 @@ class Program
 
    const axios = require('axios');
 const fs = require('fs');
 
-const API_URL = 'http://localhost:8080/object-detection'
+const API_URL = 'http://localhost:8080/small-object-detection'
 const imagePath = './demo.jpg'
 const outputImagePath = "./out.jpg";
 
@@ -715,7 +715,7 @@ axios.request(config)
 
 
    <?php
 
-$API_URL = "http://localhost:8080/object-detection";
+$API_URL = "http://localhost:8080/small-object-detection";
 $image_path = "./demo.jpg";
 $output_image_path = "./out.jpg";
 
diff --git a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md
index 05ee161f76..41daf78d2e 100644
--- a/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md
+++ b/docs/pipeline_usage/tutorials/cv_pipelines/small_object_detection.md
@@ -294,7 +294,7 @@ for res in output:
 
  • infer
    • 
 
 
    对图像进行目标检测。
    
-
    POST /object-detection
    
+
    POST /small-object-detection
    
 
      
 
    • 请求体的属性如下：
      • 
 
    
@@ -404,7 +404,7 @@ for res in output:
 
    import base64
 import requests
 
-API_URL = "http://localhost:8080/object-detection" # 服务URL
+API_URL = "http://localhost:8080/small-object-detection" # 服务URL
 image_path = "./demo.jpg"
 output_image_path = "./out.jpg"
 
@@ -462,7 +462,7 @@ int main() {
     std::string body = jsonObj.dump();
 
     // 调用API
-    auto response = client.Post("/object-detection", headers, body, "application/json");
+    auto response = client.Post("/small-object-detection", headers, body, "application/json");
     // 处理接口返回数据
     if (response && response->status == 200) {
         nlohmann::json jsonResponse = nlohmann::json::parse(response->body);
@@ -508,7 +508,7 @@ import java.util.Base64;
 
 public class Main {
     public static void main(String[] args) throws IOException {
-        String API_URL = "http://localhost:8080/object-detection"; // 服务URL
+        String API_URL = "http://localhost:8080/small-object-detection"; // 服务URL
         String imagePath = "./demo.jpg"; // 本地图像
         String outputImagePath = "./out.jpg"; // 输出图像
 
@@ -567,7 +567,7 @@ import (
 )
 
 func main() {
-    API_URL := "http://localhost:8080/object-detection"
+    API_URL := "http://localhost:8080/small-object-detection"
     imagePath := "./demo.jpg"
     outputImagePath := "./out.jpg"
 
@@ -650,7 +650,7 @@ using Newtonsoft.Json.Linq;
 
 class Program
 {
-    static readonly string API_URL = "http://localhost:8080/object-detection";
+    static readonly string API_URL = "http://localhost:8080/small-object-detection";
     static readonly string imagePath = "./demo.jpg";
     static readonly string outputImagePath = "./out.jpg";
 
@@ -689,7 +689,7 @@ class Program
 
    const axios = require('axios');
 const fs = require('fs');
 
-const API_URL = 'http://localhost:8080/object-detection'
+const API_URL = 'http://localhost:8080/small-object-detection'
 const imagePath = './demo.jpg'
 const outputImagePath = "./out.jpg";
 
@@ -730,7 +730,7 @@ axios.request(config)
 
 
    <?php
 
-$API_URL = "http://localhost:8080/object-detection"; // 服务URL
+$API_URL = "http://localhost:8080/small-object-detection"; // 服务URL
 $image_path = "./demo.jpg";
 $output_image_path = "./out.jpg";
 
diff --git a/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py b/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py
index fe67240312..5287c567ff 100644
--- a/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py
+++ b/paddlex/inference/pipelines/serving/_pipeline_apps/small_object_detection.py
@@ -49,7 +49,9 @@ def create_pipeline_app(pipeline: SmallObjDet, app_config: AppConfig) -> FastAPI
     )
 
     @app.post(
-        "/object-detection", operation_id="infer", responses={422: {"model": Response}}
+        "/small-object-detection",
+        operation_id="infer",
+        responses={422: {"model": Response}},
     )
     async def _infer(request: InferRequest) -> ResultResponse[InferResult]:
         pipeline = ctx.pipeline

From d351079b3bd104e17e6cc4906b1926f326055d07 Mon Sep 17 00:00:00 2001
From: Bobholamovic 
Date: Fri, 15 Nov 2024 17:01:38 +0800
Subject: [PATCH 5/5] Fix face recognition bugs

---
 paddlex/inference/pipelines/face_recognition.py | 2 ++
 paddlex/pipelines/face_recognition.yaml         | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/paddlex/inference/pipelines/face_recognition.py b/paddlex/inference/pipelines/face_recognition.py
index a925dbb006..e7a55f7854 100644
--- a/paddlex/inference/pipelines/face_recognition.py
+++ b/paddlex/inference/pipelines/face_recognition.py
@@ -21,6 +21,8 @@ class FaceRecPipeline(ShiTuV2Pipeline):
     entities = "face_recognition"
 
     def get_rec_result(self, det_res, indexer):
+        if len(det_res["boxes"]) == 0:
+            return {"label": [], "score": []}
         subs_of_img = list(self._crop_by_boxes(det_res))
         img_list = [img["img"] for img in subs_of_img]
         all_rec_res = list(self.rec_model(img_list))
diff --git a/paddlex/pipelines/face_recognition.yaml b/paddlex/pipelines/face_recognition.yaml
index ce1cccb27b..5da78fecc0 100644
--- a/paddlex/pipelines/face_recognition.yaml
+++ b/paddlex/pipelines/face_recognition.yaml
@@ -8,6 +8,6 @@ Pipeline:
   det_batch_size: 1
   rec_batch_size: 1
   device: gpu
-  index_dir: None
+  index: None
   score_thres: 0.4
   return_k: 5