{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multi-Single label text classification\n",
    "\n",
    "*Note: This example builds upon multi-label classification. If you are unfamiliar with it, refer to the previous examples.*\n",
    "\n",
    "Multi-single label classification is a type of hierarchical classification: First, a sample is classified with multi-label classification, yielding a mumber of classes. Then, a single prediction (in a second dimension) is given for each of those classes.  \n",
    "We showcase this with an ABSA (aspect-based sentiment analysis) example, using reviews of hotels. Just like a multi-label dataset, each sample has multiple aspects, but then each aspect additionally has a sentiment. For example, \"I liked the breakfast, althought the staff doesn’t even bother to greet you when you enter\" has the aspects \"food\" with positive sentiment, and the aspect \"staff\" with a negative sentiment. The label would thus be `[[\"Food\", \"positive\"], [\"Staff\", \"negative\"]]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from autotransformers import AutoTransformer, DatasetLoader"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below is an example dataset for multi-single-label tasks. Additionally to the \"classes\", we now have to configure a list of \"inner classes\" for the inner classification task. Note the `None` inner class, which is used when a sample is not labeled with the corresponding outer class.  \n",
    "The samples' values are lists of pairs in multi-single classification: Each pair specified an outer and its corresponding inner class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = {\n",
    "    \"meta\": {\n",
    "        \"name\": \"example_multisingle_label\",\n",
    "        \"version\": \"1.0.0\",\n",
    "        \"created_with\": \"wizard\"\n",
    "    },\n",
    "    \"config\": [\n",
    "        {\n",
    "            \"domain\": \"text\",\n",
    "            \"type\": \"IText\"\n",
    "        },\n",
    "        {\n",
    "            \"task_id\": \"task1\",\n",
    "            \"classes\": [\"Room\", \"Staff\", \"Cleanliness\"],\n",
    "            \"inner_classes\": [\"None\", \"positive\", \"neutral\", \"negative\"],\n",
    "            \"none_inner_class\": \"None\",\n",
    "            \"type\": \"TMultiSingleClassification\"\n",
    "        }\n",
    "    ],\n",
    "    \"train\": [\n",
    "        [\n",
    "            {\"value\": \"the room was very spacious.\"},\n",
    "            {\"value\": [[\"Room\", \"positive\"]]},\n",
    "        ],\n",
    "        [\n",
    "            {\"value\": \"Everything was spotless, bed lines were indeed heavenly and the bed was super comfortable with wonderful pillows.\"},\n",
    "            {\"value\": [[\"Cleanliness\", \"positive\"]]},\n",
    "        ],\n",
    "        [\n",
    "            {\"value\": \"In the room there is a tiny bathroom with shower and seperate toilet.\"},\n",
    "            {\"value\": [[\"Room\", \"neutral\"]]},\n",
    "        ],\n",
    "        [\n",
    "            {\"value\": \"Our agent had booked the wrong room type and the hotel room we were given was poor.\"},\n",
    "            {\"value\": [[\"Room\", \"negative\"]]},\n",
    "        ],\n",
    "    ],\n",
    "    \"eval\": [\n",
    "        [\n",
    "            {\"value\": \"Really clean right by a canal staff really friendly and helpful.\"},\n",
    "            {\"value\": [[\"Staff\", \"positive\"], [\"Cleanliness\", \"positive\"]]},\n",
    "        ],\n",
    "    ],\n",
    "    \"test\": [\n",
    "        [\n",
    "            {\"value\": \"The concierge was also very helpful in organising some train tickets\"},\n",
    "            {\"value\": [[\"Staff\", \"positive\"]]},\n",
    "        ]\n",
    "    ]\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As before, we simply create a DatasetLoader from the dataset and start training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dl = DatasetLoader(dataset)\n",
    "\n",
    "# Or create a DatasetLoader from a file\n",
    "# dl = DatasetLoader(\"path/to/my-dataset.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# In this example, we only train for one epoch to finish fast. \n",
    "# In reality, you want to set this to a higher value for better results.\n",
    "config = [\n",
    "    (\"engine/stop_condition/type\", \"MaxEpochs\"),\n",
    "    (\"engine/stop_condition/value\", 1),\n",
    "]\n",
    "at = AutoTransformer(config)\n",
    "\n",
    "at.init(dataset_loader=dl, path=\".models/example03\")\n",
    "at.train(dl)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And check our result:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "at(\"I liked the breakfast, althought the staff doesn't even bother to greet you when you go anywhere.\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "83ee97f1e4ad98a710574577955c6720418d3d8f987616cd4f238f891737d017"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}