Create job_to_cv_openai.ipynb

cv generation

Function 01/Function 01/Function1/March_Update/OpenAI/job_to_cv_openai.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Test with OpenAI Server"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from openai import OpenAI\n",
+    "import os\n",
+    "from dotenv import load_dotenv\n",
+    "load_dotenv()\n",
+    "\n",
+    "# Point to the OpenAI server\n",
+    "client = OpenAI(api_key=os.environ.get(\"OPENAI_API_KEY\"))\n",
+    "\n",
+    "# Point to the local server\n",
+    "# client = OpenAI(base_url=\"http://localhost:1234/v1\", api_key=\"not-needed\")\n",
+    "\n",
+    "def generate_resumes(job_description):\n",
+    "\n",
+    "    # Create completions\n",
+    "    completion = client.chat.completions.create(\n",
+    "        model=\"gpt-3.5-turbo\",\n",
+    "        messages=[\n",
+    "            {\"role\": \"system\", \"content\": \"You need to create 100% matching 3 different separate BEST RESUMES for the following job description. Number the Resumes as Resume 1, Resume 2, Resume 3. Don't Include like Personal Details Sections.\"},\n",
+    "            {\"role\": \"user\", \"content\": job_description}\n",
+    "        ]\n",
+    "    )\n",
+    "\n",
+    "    cvs = completion.choices[0].message.content\n",
+    "    return cvs"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import re\n",
+    "import pandas as pd\n",
+    "\n",
+    "def preprocess_text(text):\n",
+    "    # Remove characters like * - + # _\n",
+    "    processed_text = re.sub(r'[\\*\\-\\+#_]', '', text)\n",
+    "    # Replace '\\n\\n' with a space, and '\\n' with a single newline\n",
+    "    processed_text = re.sub(r'\\n\\n', ' ', processed_text)\n",
+    "    processed_text = re.sub(r'\\n', ' ', processed_text)\n",
+    "    return processed_text\n",
+    "\n",
+    "def separate_resumes(cvs):\n",
+    "    # Try the first option: Resume 1:\n",
+    "    resume_match = re.search(r'Resume 1:(.*?)Resume 2', cvs, re.DOTALL)\n",
+    "    resume1 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "\n",
+    "    # If the first option is not found, try the second option: Resume 1\n",
+    "    if not resume1:\n",
+    "        resume_match = re.search(r'Resume 1(.*?)Resume 2', cvs, re.DOTALL)\n",
+    "        resume1 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "\n",
+    "    resume1 = preprocess_text(resume1)\n",
+    "\n",
+    "    # Repeat the process for Resume 2 and Resume 3\n",
+    "    resume_match = re.search(r'Resume 2:(.*?)Resume 3', cvs, re.DOTALL)\n",
+    "    resume2 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "    if not resume2:\n",
+    "        resume_match = re.search(r'Resume 2(.*?)Resume 3', cvs, re.DOTALL)\n",
+    "        resume2 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "    resume2 = preprocess_text(resume2)\n",
+    "\n",
+    "    resume_match = re.search(r'Resume 3:(.*)', cvs, re.DOTALL)\n",
+    "    resume3 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "    if not resume3:\n",
+    "        resume_match = re.search(r'Resume 3(.*)', cvs, re.DOTALL)\n",
+    "        resume3 = resume_match.group(1).strip() if resume_match else \"\"\n",
+    "    resume3 = preprocess_text(resume3)\n",
+    "\n",
+    "    data = {'Resume 1': [resume1], 'Resume 2': [resume2], 'Resume 3': [resume3]}\n",
+    "    df = pd.DataFrame(data)\n",
+    "    return df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load the job data\n",
+    "job_data = pd.read_csv('Job_ad.csv')\n",
+    "\n",
+    "# Create new columns 'Resume 1', 'Resume 2', 'Resume 3' in the job data\n",
+    "for index, row in job_data.iterrows():\n",
+    "    job_description = f\"{row['jobTitle']} {row['jobSummary']} {row['keyResponsibilities']} {row['requiredQualification']}\"\n",
+    "    cvs_result = generate_resumes(job_description)\n",
+    "    df_separate_result = separate_resumes(cvs_result)\n",
+    "\n",
+    "    # Add new columns to the job data\n",
+    "    job_data.loc[index, 'Resume 1'] = df_separate_result['Resume 1'][0]\n",
+    "    job_data.loc[index, 'Resume 2'] = df_separate_result['Resume 2'][0]\n",
+    "    job_data.loc[index, 'Resume 3'] = df_separate_result['Resume 3'][0]\n",
+    "\n",
+    "# Save the updated job data to a new CSV file\n",
+    "job_data.to_csv('Job_ad_with_resumes.csv', index=False)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "resume-ranker",
+   "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": "3.10.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}