{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "fbcdfea8-7241-46d7-a771-c0381a3e7063", "metadata": {}, "outputs": [], "source": [ "# imports\n", "\n", "import os\n", "import re\n", "import math\n", "import json\n", "from tqdm import tqdm\n", "import random\n", "from dotenv import load_dotenv\n", "from huggingface_hub import login\n", "import numpy as np\n", "import pickle\n", "from openai import OpenAI\n", "from sentence_transformers import SentenceTransformer\n", "from datasets import load_dataset\n", "import chromadb\n", "from items import Item\n", "from testing import Tester\n", "from agents.pricer_agent import price\n", "import pandas as pd\n", "import numpy as np\n", "from sklearn.linear_model import LinearRegression\n", "from sklearn.metrics import mean_squared_error, r2_score" ] }, { "cell_type": "code", "execution_count": 2, "id": "e6e88bd1-f89c-4b98-92fa-aa4bc1575bca", "metadata": {}, "outputs": [], "source": [ "# CONSTANTS\n", "\n", "QUESTION = \"How much does this cost to the nearest dollar?\\n\\n\"\n", "DB = \"products_vectorstore\"" ] }, { "cell_type": "code", "execution_count": 3, "id": "98666e73-938e-469d-8987-e6e55ba5e034", "metadata": {}, "outputs": [], "source": [ "# environment\n", "\n", "load_dotenv()\n", "os.environ['OPENAI_API_KEY'] = os.getenv('OPENAI_API_KEY', 'your-key-if-not-using-env')\n", "os.environ['HF_TOKEN'] = os.getenv('HF_TOKEN', 'your-key-if-not-using-env')" ] }, { "cell_type": "code", "execution_count": 4, "id": "9a25a5cf-8f6c-4b5d-ad98-fdd096f5adf8", "metadata": {}, "outputs": [], "source": [ "openai = OpenAI()" ] }, { "cell_type": "code", "execution_count": 5, "id": "dc696493-0b6f-48aa-9fa8-b1ae0ecaf3cd", "metadata": {}, "outputs": [], "source": [ "# Load in the test pickle file:\n", "\n", "with open('test.pkl', 'rb') as file:\n", " test = pickle.load(file)" ] }, { "cell_type": "code", "execution_count": 6, "id": "33d38a06-0c0d-4e96-94d1-35ee183416ce", "metadata": {}, "outputs": [], "source": [ "def make_context(similars, prices):\n", " message = \"To provide some context, here are some other items that might be similar to the item you need to estimate.\\n\\n\"\n", " for similar, price in zip(similars, prices):\n", " message += f\"Potentially related product:\\n{similar}\\nPrice is ${price:.2f}\\n\\n\"\n", " return message" ] }, { "cell_type": "code", "execution_count": null, "id": "61f203b7-63b6-48ed-869b-e393b5bfcad3", "metadata": {}, "outputs": [], "source": [ "def messages_for(item, similars, prices):\n", " system_message = \"You estimate prices of items. Reply only with the price, no explanation\"\n", " user_prompt = make_context(similars, prices)\n", " user_prompt += \"And now the question for you:\\n\\n\"\n", " user_prompt += item.test_prompt().replace(\" to the nearest dollar\",\"\").replace(\"\\n\\nPrice is $\",\"\")\n", " return [\n", " {\"role\": \"system\", \"content\": system_message},\n", " {\"role\": \"user\", \"content\": user_prompt},\n", " {\"role\": \"assistant\", \"content\": \"Price is $\"}\n", " ]" ] }, { "cell_type": "code", "execution_count": null, "id": "b26f405d-6e1f-4caa-b97f-1f62cd9d1ebc", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "d26a1104-cd11-4361-ab25-85fb576e0582", "metadata": {}, "outputs": [], "source": [ "client = chromadb.PersistentClient(path=DB)\n", "collection = client.get_or_create_collection('products')" ] }, { "cell_type": "code", "execution_count": null, "id": "1e339760-96d8-4485-bec7-43fadcd30c4d", "metadata": {}, "outputs": [], "source": [ "def description(item):\n", " text = item.prompt.replace(\"How much does this cost to the nearest dollar?\\n\\n\", \"\")\n", " return text.split(\"\\n\\nPrice is $\")[0]" ] }, { "cell_type": "code", "execution_count": null, "id": "9f759bd2-7a7e-4c1a-80a0-e12470feca89", "metadata": {}, "outputs": [], "source": [ "model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')" ] }, { "cell_type": "code", "execution_count": null, "id": "e44dbd25-fb95-4b6b-bbbb-8da5fc817105", "metadata": {}, "outputs": [], "source": [ "def vector(item):\n", " return model.encode([description(item)])" ] }, { "cell_type": "code", "execution_count": null, "id": "ffd5ee47-db5d-4263-b0d9-80d568c91341", "metadata": {}, "outputs": [], "source": [ "def find_similars(item):\n", " results = collection.query(query_embeddings=vector(item).astype(float).tolist(), n_results=5)\n", " documents = results['documents'][0][:]\n", " prices = [m['price'] for m in results['metadatas'][0][:]]\n", " return documents, prices" ] }, { "cell_type": "code", "execution_count": null, "id": "d11f1c8d-7480-4d64-a274-b030d701f1b8", "metadata": {}, "outputs": [], "source": [ "def get_price(s):\n", " s = s.replace('$','').replace(',','')\n", " match = re.search(r\"[-+]?\\d*\\.\\d+|\\d+\", s)\n", " return float(match.group()) if match else 0" ] }, { "cell_type": "code", "execution_count": null, "id": "a919cf7d-b3d3-4968-8c96-54a0da0b0219", "metadata": {}, "outputs": [], "source": [ "# The function for gpt-4o-mini\n", "\n", "def gpt_4o_mini_rag(item):\n", " documents, prices = find_similars(item)\n", " response = openai.chat.completions.create(\n", " model=\"gpt-4o-mini\", \n", " messages=messages_for(item, documents, prices),\n", " seed=42,\n", " max_tokens=5\n", " )\n", " reply = response.choices[0].message.content\n", " return get_price(reply)" ] }, { "cell_type": "code", "execution_count": null, "id": "8b918cfc-76c1-442a-8caa-bec500cd504b", "metadata": {}, "outputs": [], "source": [ "gpt_4o_mini_rag(test[1000])" ] }, { "cell_type": "code", "execution_count": null, "id": "c92cfc0b-b36d-456f-94cc-fe3f315cc25e", "metadata": {}, "outputs": [], "source": [ "test[1000]" ] }, { "cell_type": "code", "execution_count": null, "id": "e6d5deb3-6a2a-4484-872c-37176c5e1f07", "metadata": {}, "outputs": [], "source": [ "def proprietary(item):\n", " text = item.prompt.split(\"to the nearest dollar?\\n\\n\")[1].split(\"\\n\\nPrice is $\")[0]\n", " return price(text)" ] }, { "cell_type": "code", "execution_count": null, "id": "bacdf607-37b9-4997-adb1-d63abfb645b1", "metadata": {}, "outputs": [], "source": [ "print(proprietary(test[1]))\n", "print(gpt_4o_mini_rag(test[1]))" ] }, { "cell_type": "code", "execution_count": null, "id": "b35532e7-098a-4ab9-a8f7-8f101b437181", "metadata": {}, "outputs": [], "source": [ "truths = []\n", "proprietaries = []\n", "rags = []\n", "for i in tqdm(range(1000,1250)):\n", " item = test[i]\n", " truths.append(item.price)\n", " proprietaries.append(proprietary(item))\n", " rags.append(gpt_4o_mini_rag(item))" ] }, { "cell_type": "code", "execution_count": null, "id": "e6ae54c7-6e8e-4333-b075-b59978fed560", "metadata": {}, "outputs": [], "source": [ "mins = [min(p,r) for p,r in zip(proprietaries, rags)]\n", "maxes = [max(p,r) for p,r in zip(proprietaries, rags)]\n", "\n", "X = pd.DataFrame({\n", " 'Proprietary': proprietaries,\n", " 'RAG': rags,\n", " 'Min': mins,\n", " 'Max': maxes,\n", "})\n", "\n", "# Convert y to a Series\n", "y = pd.Series(truths)" ] }, { "cell_type": "code", "execution_count": null, "id": "e68684ed-d029-4d95-bb13-eead19b20e49", "metadata": {}, "outputs": [], "source": [ "# Train a Linear Regression\n", "np.random.seed(42)\n", "\n", "lr = LinearRegression()\n", "lr.fit(X, y)\n", "\n", "feature_columns = [\"Proprietary\", \"RAG\", \"Min\", \"Max\"]\n", "\n", "for feature, coef in zip(feature_columns, lr.coef_):\n", " print(f\"{feature}: {coef:.2f}\")\n", "print(f\"Intercept={lr.intercept_:.2f}\")" ] }, { "cell_type": "code", "execution_count": null, "id": "28530362-97b8-42a0-bf89-967539b6f170", "metadata": {}, "outputs": [], "source": [ "def ensemble(item):\n", " prop = proprietary(item)\n", " rag = gpt_4o_mini_rag(item)\n", " Xt = pd.DataFrame({\n", " 'Proprietary': [prop],\n", " 'RAG': [rag],\n", " 'Min': [min(prop,rag)],\n", " 'Max': [max(prop,rag)],\n", " })\n", " yt = lr.predict(Xt)\n", " return yt[0]" ] }, { "cell_type": "code", "execution_count": null, "id": "08021c05-340b-4ee2-9d11-4b280766976f", "metadata": {}, "outputs": [], "source": [ "ensemble(test[0])" ] }, { "cell_type": "code", "execution_count": null, "id": "d8308c74-546f-4fc0-ada4-1974addacfd1", "metadata": {}, "outputs": [], "source": [ "test[0].price" ] }, { "cell_type": "code", "execution_count": null, "id": "80792910-c59f-4d96-aa53-683464a8e60c", "metadata": {}, "outputs": [], "source": [ "Tester.test(ensemble, test)" ] }, { "cell_type": "code", "execution_count": null, "id": "d0c41043-2049-4883-947f-2aad2f6954c2", "metadata": {}, "outputs": [], "source": [ "from sklearn.ensemble import RandomForestRegressor\n", "\n", "result = collection.get(include=['embeddings', 'documents', 'metadatas'])\n", "vectors = np.array(result['embeddings'])\n", "documents = result['documents']\n", "prices = [metadata['price'] for metadata in result['metadatas']]" ] }, { "cell_type": "code", "execution_count": null, "id": "e9c3276f-ae01-478d-bb27-dc73b567b41a", "metadata": {}, "outputs": [], "source": [ "rf_model = RandomForestRegressor(n_estimators=100, random_state=42, n_jobs=8)\n", "rf_model.fit(vectors, prices)" ] }, { "cell_type": "code", "execution_count": null, "id": "3e8f70cd-4147-40c6-9861-a3513b7e5499", "metadata": {}, "outputs": [], "source": [ "def new_rf(item):\n", " text = item.prompt.split(\"to the nearest dollar?\\n\\n\")[1].split(\"\\n\\nPrice is $\")[0]\n", " vector = model.encode([text])\n", " return max(0, rf_model.predict(vector)[0])" ] }, { "cell_type": "code", "execution_count": null, "id": "a2e3340f-7ed4-47eb-a5a9-dff4c0353f58", "metadata": {}, "outputs": [], "source": [ "new_rf(test[0])" ] }, { "cell_type": "code", "execution_count": null, "id": "f91c903b-8db1-4374-807e-3a8ce282ef30", "metadata": {}, "outputs": [], "source": [ "Tester.test(new_rf, test)" ] }, { "cell_type": "code", "execution_count": null, "id": "3c8e23c5-1ed3-4bd1-a3c0-129d4712c93a", "metadata": {}, "outputs": [], "source": [ "forests = []\n", "for i in tqdm(range(1000,1250)):\n", " item = test[i]\n", " forests.append(new_rf(item))" ] }, { "cell_type": "code", "execution_count": null, "id": "8e2eca63-8230-4904-9a79-7e779747479e", "metadata": {}, "outputs": [], "source": [ "truths2 = []\n", "proprietaries2 = []\n", "rags2 = []\n", "forests2 = []\n", "for i in tqdm(range(1000,2000)):\n", " item = test[i]\n", " truths2.append(item.price)\n", " proprietaries2.append(proprietary(item))\n", " rags2.append(gpt_4o_mini_rag(item))\n", " forests2.append(new_rf(item))" ] }, { "cell_type": "code", "execution_count": null, "id": "0a3e057f-05c5-4f8f-8b3b-0afdfccc1412", "metadata": {}, "outputs": [], "source": [ "mins2 = [min(p,r,f) for p,r,f in zip(proprietaries2, rags2, forests2)]\n", "maxes2 = [max(p,r,f) for p,r,f in zip(proprietaries2, rags2, forests2)]\n", "\n", "\n", "\n", "X2 = pd.DataFrame({\n", " 'Proprietary': proprietaries2,\n", " 'RAG': rags2,\n", " 'Forest': forests2,\n", " 'Min': mins2,\n", " 'Max': maxes2,\n", "})\n", "\n", "# Convert y to a Series\n", "y2 = pd.Series(truths2)" ] }, { "cell_type": "code", "execution_count": null, "id": "1ae62175-b955-428e-b077-705c49ee71bd", "metadata": {}, "outputs": [], "source": [ "# Train a Linear Regression\n", "np.random.seed(42)\n", "\n", "lr2 = LinearRegression()\n", "lr2.fit(X2, y2)\n", "\n", "feature_columns = X2.columns.tolist()\n", "\n", "for feature, coef in zip(feature_columns, lr2.coef_):\n", " print(f\"{feature}: {coef:.2f}\")\n", "print(f\"Intercept={lr.intercept_:.2f}\")" ] }, { "cell_type": "code", "execution_count": null, "id": "214a3831-c464-4218-a349-534b6bda7f12", "metadata": {}, "outputs": [], "source": [ "def ensemble2(item):\n", " prop = proprietary(item)\n", " rag = gpt_4o_mini_rag(item)\n", " r_f = new_rf(item)\n", " Xt2 = pd.DataFrame({\n", " 'Proprietary': [prop],\n", " 'RAG': [rag],\n", " 'Forest': [r_f],\n", " 'Min': [min(prop,rag, r_f)],\n", " 'Max': [max(prop,rag, r_f)],\n", " })\n", " yt2 = lr.predict(Xt2)\n", " return yt2[0]" ] }, { "cell_type": "code", "execution_count": null, "id": "b234cb68-af68-4475-ae18-8892aac6b74e", "metadata": {}, "outputs": [], "source": [ "Tester.test(ensemble2, test)" ] }, { "cell_type": "code", "execution_count": null, "id": "10a7275f-1aa9-4446-9100-a7a0ba0215f2", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.11.10" } }, "nbformat": 4, "nbformat_minor": 5 }