{ "cells": [ { "cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting sentencepiece\n", " Using cached sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.7 kB)\n", "Downloading sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m49.7 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n", "\u001b[?25hInstalling collected packages: sentencepiece\n", "Successfully installed sentencepiece-0.2.0\n" ] } ], "source": [ "!pip install sentencepiece" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting langsmith\n", " Downloading langsmith-0.1.104-py3-none-any.whl.metadata (13 kB)\n", "Requirement already satisfied: httpx<1,>=0.23.0 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from langsmith) (0.27.0)\n", "Requirement already satisfied: orjson<4.0.0,>=3.9.14 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from langsmith) (3.10.7)\n", "Requirement already satisfied: pydantic<3,>=1 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from langsmith) (2.8.2)\n", "Requirement already satisfied: requests<3,>=2 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from langsmith) (2.32.2)\n", "Requirement already satisfied: anyio in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith) (4.4.0)\n", "Requirement already satisfied: certifi in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith) (2024.7.4)\n", "Requirement already satisfied: httpcore==1.* in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith) (1.0.5)\n", "Requirement already satisfied: idna in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith) (3.7)\n", "Requirement already satisfied: sniffio in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith) (1.3.1)\n", "Requirement already satisfied: h11<0.15,>=0.13 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from httpcore==1.*->httpx<1,>=0.23.0->langsmith) (0.14.0)\n", "Requirement already satisfied: annotated-types>=0.4.0 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from pydantic<3,>=1->langsmith) (0.7.0)\n", "Requirement already satisfied: pydantic-core==2.20.1 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from pydantic<3,>=1->langsmith) (2.20.1)\n", "Requirement already satisfied: typing-extensions>=4.6.1 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from pydantic<3,>=1->langsmith) (4.12.2)\n", "Requirement already satisfied: charset-normalizer<4,>=2 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from requests<3,>=2->langsmith) (2.0.4)\n", "Requirement already satisfied: urllib3<3,>=1.21.1 in /home/sukhera/miniconda3/lib/python3.12/site-packages (from requests<3,>=2->langsmith) (2.2.2)\n", "Downloading langsmith-0.1.104-py3-none-any.whl (149 kB)\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m149.1/149.1 kB\u001b[0m \u001b[31m22.4 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n", "\u001b[?25hInstalling collected packages: langsmith\n", "Successfully installed langsmith-0.1.104\n" ] } ], "source": [ "!pip install -U langsmith" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/sukhera/miniconda3/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", " from .autonotebook import tqdm as notebook_tqdm\n" ] } ], "source": [ "import os\n", "import PyPDF2\n", "from transformers import BertTokenizer, BertModel\n", "from transformers import LongformerModel, LongformerTokenizer\n", "from transformers import BigBirdModel, BigBirdTokenizer\n", "import numpy as np\n", "from groq import Groq\n", "import gradio as gr\n", "from pathlib import Path\n", "import torch\n", "import json\n" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "from langsmith import Client\n", "\n", "# Initialize the LangSmith Client\n", "os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n", "os.environ[\"LANGCHAIN_API_KEY\"] = \"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\"\n", "client = Client()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/sukhera/miniconda3/lib/python3.12/site-packages/transformers/tokenization_utils_base.py:1601: FutureWarning: `clean_up_tokenization_spaces` was not set. It will be set to `True` by default. This behavior will be depracted in transformers v4.45, and will be then set to `False` by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884\n", " warnings.warn(\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n", "A parameter name that contains `beta` will be renamed internally to `bias`. Please use a different name to suppress this warning.\n", "A parameter name that contains `gamma` will be renamed internally to `weight`. Please use a different name to suppress this warning.\n" ] } ], "source": [ "# Load BERT tokenizer and model\n", "tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\n", "model = BertModel.from_pretrained('bert-base-uncased')" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "# Load the BigBird model and tokenizer\n", "tokenizer = BigBirdTokenizer.from_pretrained('google/bigbird-roberta-base')\n", "model = BigBirdModel.from_pretrained('google/bigbird-roberta-base')\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "#longformer\n", "# Load the Longformer model and tokenizer\n", "tokenizer = LongformerTokenizer.from_pretrained('allenai/longformer-base-4096')\n", "model = LongformerModel.from_pretrained('allenai/longformer-base-4096')\n" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [], "source": [ "#longFormer\n", "\n", "def get_longformer_embedding(text):\n", " # Tokenize the text\n", " inputs = tokenizer(text, return_tensors='pt', truncation=True, padding=True, max_length=4096)\n", " \n", " # Get the embeddings from Longformer\n", " with torch.no_grad():\n", " outputs = model(**inputs)\n", " \n", " # Use the [CLS] token's embedding as the aggregate representation\n", " cls_embedding = outputs.last_hidden_state[:, 0, :].numpy()\n", " \n", " return cls_embedding" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "# BIGBIRD\n", "def get_bigbird_embedding(text):\n", " # Tokenize the text\n", " inputs = tokenizer(text, return_tensors='pt', truncation=True, padding=True, max_length=4096)\n", " \n", " # Get the embeddings from BigBird\n", " with torch.no_grad():\n", " outputs = model(**inputs)\n", " \n", " # Use the [CLS] token's embedding as the aggregate representation\n", " cls_embedding = outputs.last_hidden_state[:, 0, :].numpy()\n", " \n", " return cls_embedding" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "def get_bert_embedding(text):\n", " # Tokenize the text\n", " inputs = tokenizer(text, return_tensors='pt', truncation=True, padding=True, max_length=512)\n", " \n", " # Get the embeddings from BERT\n", " with torch.no_grad():\n", " outputs = model(**inputs)\n", " \n", " # Use the [CLS] token's embedding as the aggregate representation\n", " cls_embedding = outputs.last_hidden_state[:, 0, :].numpy()\n", " \n", " return cls_embedding\n" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "def process_folder(file):\n", " folder_path = os.path.dirname(file.name) # Get the directory of the selected file\n", " files = os.listdir(folder_path) # List all files in the directory\n", " file_paths = [os.path.join(folder_path, f) for f in files] # Get full paths of all files\n", " return f\"Files in folder: {', '.join(files)}\"" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "# Function to extract text from a PDF\n", "def extract_text_from_pdf(pdf_file):\n", " text = ''\n", " with open(pdf_file, 'rb') as file:\n", " reader = PyPDF2.PdfReader(file)\n", " for page in reader.pages:\n", " text += page.extract_text() or ''\n", " return text\n" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "cluster_emb={}" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "\n", "def calculate_cosine(embedding1, embedding2):\n", " # Calculate the dot product and magnitudes of the embeddings\n", " dot_product = np.dot(embedding1, embedding2)\n", " magnitude1 = np.linalg.norm(embedding1)\n", " magnitude2 = np.linalg.norm(embedding2)\n", " \n", " # Calculate cosine similarity\n", " similarity = dot_product / (magnitude1 * magnitude2)\n", " return similarity" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [], "source": [ "def foo(files, JD):\n", " # Extract text and compute embeddings for job description using different models\n", " text_jd = extract_text_from_pdf(JD) \n", " JD_embedding_bert = get_bert_embedding(text_jd).flatten() # Flatten to match the dimension\n", " JD_embedding_longformer = get_longformer_embedding(text_jd).flatten()\n", " JD_embedding_bigbird = get_bigbird_embedding(text_jd).flatten()\n", "\n", " sim = []\n", " \n", " for d in files:\n", " text = extract_text_from_pdf(d)\n", " # Compute embeddings for the resume using different models\n", " resume_embedding_bert = get_bert_embedding(text).flatten() # Fixed function call\n", " resume_embedding_longformer = get_longformer_embedding(text).flatten()\n", " resume_embedding_bigbird = get_bigbird_embedding(text).flatten()\n", " # Calculate cosine similarity for each model\n", " similarity_bert = calculate_cosine(resume_embedding_bert, JD_embedding_bert)\n", " similarity_longformer = calculate_cosine(resume_embedding_longformer, JD_embedding_longformer)\n", " similarity_bigbird = calculate_cosine(resume_embedding_bigbird, JD_embedding_bigbird)\n", " # Append the results to the array\n", " sim.append(f\"\\nFile: {d.name:}\\n\"\n", " f\"Bert Similarity: {similarity_bert:.4f}\\n\"\n", " f\"Longformer Similarity: {similarity_longformer:.4f}\\n\"\n", " f\"BigBird Similarity: {similarity_bigbird:.4f}\\n\")\n", " \n", " \n", " \n", " return \"\\n\".join(sim) # Join the list into a single string for Gradio output\n" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/sukhera/miniconda3/lib/python3.12/site-packages/transformers/tokenization_utils_base.py:2888: UserWarning: `max_length` is ignored when `padding`=`True` and there is no truncation strategy. To pad to max length, use `padding='max_length'`.\n", " warnings.warn(\n", "/home/sukhera/miniconda3/lib/python3.12/site-packages/transformers/tokenization_utils_base.py:2888: UserWarning: `max_length` is ignored when `padding`=`True` and there is no truncation strategy. To pad to max length, use `padding='max_length'`.\n", " warnings.warn(\n" ] } ], "source": [ "\n", "with gr.Blocks() as func:\n", " inputs = [gr.File(file_count=\"multiple\", label=\"Upload Resume Files\"), gr.File(label=\"Upload Job Description\")]\n", " outputs = gr.Textbox(label=\"Similarity Scores\")\n", " show = gr.Button(value=\"Calculate Similarity\")\n", " show.click(foo, inputs, outputs)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Rerunning server... use `close()` to stop if you need to change `launch()` parameters.\n", "----\n", "\n", "To create a public link, set `share=True` in `launch()`.\n" ] }, { "data": { "text/html": [ "
" ], "text/plain": [ "