QAEngine

A machine learning engine for ktor chatbot service

.

How to use

---

First, add repository to your build.gradle file

   Kotlin Gradle Script & Gradle
   maven{
       url = "https://s01.oss.sonatype.org/content/repositories/snapshots/"
   }

---

Second, add QA Engine dependency to build.gradle

1. Gradle

   implementation "io.github.ssu-isteam:QAEngine:1.0-SNAPSHOT"
   

2. Kotlin Gradle Script

   implementation("io.github.ssu-isteam:QAEngine:1.0-SNAPSHOT")
   

---

Initialize QA Engine

All examples are coded in Kotlin

1. Create dataset loader :Currently supported dataset loader is MINDsLabDataSetLoader

   var loader = MINDsLabDataSetLoader("your_mindslab_dataset.json")

   //Also, you can add mutliple dataset loader. var loader2 = MINDsLabDataSetLoader("dataset2.json")

2. Prepare your network model and vocab cache path for TfidVectorizer

3. Initialize QAEngine and invoke initEngine() method with parameters prepared above.

   var engine = QAEngine()
   //initEngine has data set loader array parameters to load multiple datasets and merge them to one dataset.
   engine.initEngine(model_path,vocab_cache_path, loader,loader2....)
   

4. Well done! It's now time to get the answer of a question

   var answerText = engine.getAnswer(questionText)