added touchpoint

.ipynb_checkpoints/Untitled-checkpoint.ipynb

@@ -1,6 +0,0 @@
-{
- "cells": [],
- "metadata": {},
- "nbformat": 4,
- "nbformat_minor": 5
-}

Untitled.ipynb

@@ -1,110 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "id": "8e91c310-bc69-4a28-9197-e180aaaa491f",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "utils imported!\n"
-     ]
-    }
-   ],
-   "source": [
-    "import re\n",
-    "import pickle\n",
-    "import numpy as np\n",
-    "import pandas as pd\n",
-    "\n",
-    "\n",
-    "tfidf = pickle.load(open('models/tfidf.sav', 'rb'))\n",
-    "svc_sentiment = pickle.load(open('models/sentiment_model.sav', 'rb'))\n",
-    "tfidf_sentiment = pickle.load(open('models/tfidf_sentiment.sav', 'rb'))\n",
-    "svc = pickle.load(open('models/svc_model.sav', 'rb'))\n",
-    "\n",
-    "labels = [\n",
-    "       'Product quality', 'Knowledge',\n",
-    "       'Appointment', 'Service etiquette', 'Waiting time',\n",
-    "       'Repair speed', 'Repair cost', 'Repair quality', 'Warranty',\n",
-    "       'Product replacement', 'Loan sets']\n",
-    "\n",
-    "sample_file = pd.read_csv('sample.csv').to_csv(index=False).encode('utf-8')\n",
-    "\n",
-    "print('utils imported!')\n",
-    "\n",
-    "def get_single_prediction(text):\n",
-    "    \n",
-    "    # manipulate data into a format that we pass to our model\n",
-    "    text = text.lower().strip() #lower case\n",
-    "    print(list(text))\n",
-    "    # Vectorise text and store in new dataframe. Sentence vector = average of word vectors\n",
-    "    text_vectors = tfidf.transform([text])\n",
-    "    print(text_vectors.shape)\n",
-    "    # Make topic predictions\n",
-    "    results = svc.predict_proba(text_vectors).squeeze().round(2)\n",
-    "    print(results.shape)\n",
-    "\n",
-    "    pred_prob = pd.DataFrame({'topic': labels, 'probability': results}).sort_values('probability', ascending=True)\n",
-    "\n",
-    "    # Make sentiment predictions\n",
-    "    text_vectors_sentiment = tfidf_sentiment.transform(list(text))\n",
-    "\n",
-    "    results_sentiment = svc_sentiment.predict_proba(text_vectors).squeeze().round(2)\n",
-    "    pred_prob_sentiment = pd.DataFrame({'sentiment': ['Negative', 'Positive'], 'probability': results_sentiment}).sort_values('probability', ascending=True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "id": "b940cb16-c287-4fef-accc-fa54f20c3864",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "['h', 'e', 'l', 'l', 'o']\n",
-      "(1, 500)\n",
-      "(11,)\n"
-     ]
-    }
-   ],
-   "source": [
-    "get_single_prediction('hello')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "1f68efcf-479d-41de-be1e-e94d00b58fab",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "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": "3.8.8"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}