Skip to content

Latest commit

 

History

History
259 lines (214 loc) · 8.4 KB

README.restful.md

File metadata and controls

259 lines (214 loc) · 8.4 KB
Raw

Table of Contents

Jina ❤️ Docker and RESTful APIs

Run the Docker

We have put the south park example in a Docker image and make it ready to search.

docker run -p 45678:45678 jinaai/hub.app.distilbert-southpark:latest

Now you can check out the results via HTTP POST request. Jina supports RESTful APIs, and you can simply send a POST request to http://0.0.0.0:45678/api/search, e.g.

curl --request POST \
     -d '{"top_k": 10, "mode": "search",  "data": ["text:You are damn right"]}' \
     -H 'Content-Type: application/json' \
     'http://0.0.0.0:45678/api/search'

You can find more details of the Jina RESTful APIs at docs.

The response is organized in the format of json and the matched results is stored in topkResults.

Click here to see the console output 
{
  "search": {
    "docs": [
      {
        "weight": 1.0,
        "length": 1,
        "topkResults": [
          {
            "matchDoc": {
              "docId": 9328,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Stan[SEP]Oh thanks, dude.\n"
            },
            "score": {
              "value": 0.2900288,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 7600,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "StephenAbootman[SEP]Don't call me your guy!\n"
            },
            "score": {
              "value": 0.28131312,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 7497,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "MrNelson[SEP]Excuse me: over here, please?\n"
            },
            "score": {
              "value": 0.2740146,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 8814,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Kyle[SEP]Thanks, dude.\n"
            },
            "score": {
              "value": 0.27255633,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 9957,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Kyle[SEP]Thanks, dude!\n"
            },
            "score": {
              "value": 0.27255633,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 7868,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Token[SEP]Thanks, dude.\n"
            },
            "score": {
              "value": 0.27255633,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 239,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Randy[SEP]Hey, it's my boss.\n"
            },
            "score": {
              "value": 0.27224803,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 5389,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Cartman[SEP]I got your time phone!\n"
            },
            "score": {
              "value": 0.2721367,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 5878,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Kyle[SEP]Wow, dude, check it out!\n"
            },
            "score": {
              "value": 0.2709449,
              "opName": "MinRanker"
            }
          },
          {
            "matchDoc": {
              "docId": 4388,
              "weight": 1.0,
              "mimeType": "text/plain",
              "text": "Steve[SEP]Please hang on, I'm going to call Customer Service.\n"
            },
            "score": {
              "value": 0.27075276,
              "opName": "MinRanker"
            }
          }
        ],
        "mimeType": "text/plain",
        "text": "text:hey, dude"
      }
    ],
    "topK": 10
  }
}

Index and Search By Yourself

Index

You can give it a try and index the data by yourself as well. We've built another docker image for playing around.

docker run -v "$(pwd)/workspace:/workspace" -v "$(pwd)/data:/data" -e "JINA_LOG_PROFILING=1" -p 5000:5000 -e "MAX_NUM_DOCS=100" jinaai/hub.app.distilbert index

Command args explained

  • $(pwd)/data is the path of the data, running bash get_data.sh will store the data at $(pwd)/data by default
  • $(pwd)/workspace is where the Jina index will be stored.
  • -e "JINA_LOG_PROFILING=1" -p 5000:5000 are just for dashboard monitoring. They are optional. By setting these args, you can monitor the logs at Jina dashboard.
  • -e "MAX_NUM_DOCS=100" is to set up the maximal number of documents to stored. For illustation purpose, we set it to 100.

Search

After building the index, you can use the following command to search,

docker run -v "$(pwd)/workspace:/workspace" -p 45678:45678 jinaai/hub.app.distilbert search

Command args explained

  • $(pwd)/workspace is where the Jina index will be stored.
  • -p 45678:45678 is for mapping the RESTful APIs to the local machine.

Now you can search with the RESTful APIs as before.

Index your own data

Of course, you can use this docker image to index your own data as well. Simply save your text data at /your/favourite/path with the name my_awesome_data.csv and keep each Document you want to index at one line in that file. Run the following command,

docker run -v "$(pwd)/workspace:/workspace" -v "/your/favourite/path:/data" -e "DATA_FILE= my_awesome_data.csv" jinaai/hub.app.distilbert index

Next Steps

  • Check out the details of this demo at README.md

Documentation

The best way to learn Jina in depth is to read our documentation. Documentation is built on every push, merge, and release event of the master branch. You can find more details about the following topics in our documentation.

Stay tuned

  • Slack chanel - a communication platform for developers to discuss Jina
  • Community newsletter - subscribe to the latest update, release and event news of Jina
  • LinkedIn - get to know Jina AI as a company
  • Twitter Follow - follow us and interact with us using hashtag #JinaSearch
  • Join Us - want to work full-time with us at Jina? We are hiring!
  • Company - know more about our company, we are fully committed to open-source!

License

Copyright (c) 2020 Jina AI Limited. All rights reserved.

Jina is licensed under the Apache License, Version 2.0. See LICENSE for the full license text.