scoring update

.env

@@ -0,0 +1 @@
+OPENAI_API_KEY=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

.gitignore

@@ -0,0 +1 @@
+/__pycache__

main_router.py → app.py

@@ -1,17 +1,25 @@
 from basic_handler import get_basic_info
 from strategy_handler import get_strategy_info
 from finance_handler import get_financial_info
-from rolling_update import update_financial_info
+from rolling_update import update_financial_info, initial_update_flow
+from select_top_financials import select_by_financials
 from config import Config
+from common_functionality import convert_to_excel
+import os
+from dotenv import load_dotenv
 
 import pandas as pd
 import openai
 import gradio as gr
 
-
+load_dotenv()
+openai.api_key = os.getenv("OPENAI_API_KEY")
+print(f"os: {os.getenv('OPENAI_API_KEY')}")
+print(openai.api_key)
 
 # TODO: add an identifier for financial columns for better repeat parsing
 
+
 def process_file(file) -> dict:
     input_data = pd.read_excel(file.name)
     data_dict = get_formatting_columns(input_data)
@@ -30,51 +38,73 @@ def get_formatting_columns(data: pd.DataFrame) -> dict:
     return data_dict
 
 
-def shape_sector_info(data_dict: dict, processed_data: pd.DataFrame, action) -> pd.DataFrame:
+def shape_sector_info(data_dict: dict, processed_data: pd.DataFrame, action) -> dict[str, pd.DataFrame]:
     if action == Config.ACTION_BASIC_INFO:
         processed_data = get_basic_info(data_dict["sector"], data_dict["num_companies"],
                                         data_dict[Config.ACTION_BASIC_INFO])
+        [processed_data[c].update(get_financial_info(company["ticker"], data_dict[Config.ACTION_FINANCIAL_INFO])
+                                  ) for c, company in enumerate(processed_data)]
         sector_data = pd.DataFrame(processed_data)
-        return sector_data
-    else:
-        processed_data = processed_data.to_dict(orient="records")
+        updated_data = initial_update_flow(sector_data)
+        filename = convert_to_excel(updated_data, filename="basic+financial")
+        return filename
+        # return {Config.MAIN_SHEET_NAME: sector_data}
+    processed_data = processed_data.to_dict(orient="records")
     if action == Config.ACTION_STRATEGY_INFO:
         [processed_data[c].update(get_strategy_info(company["name"], data_dict[Config.ACTION_STRATEGY_INFO])
                                   ) for c, company in enumerate(processed_data)]
-    elif action == Config.ACTION_FINANCIAL_INFO:
-        [processed_data[c].update(get_financial_info(company["ticker"], data_dict[Config.ACTION_FINANCIAL_INFO])
-                                  ) for c, company in enumerate(processed_data)]
     else:
-        return "Invalid action Error"
+        return {"Invalid action Error": pd.DataFrame()}
     sector_data = pd.DataFrame(processed_data)
-    return sector_data
+    return {Config.MAIN_SHEET_NAME: sector_data}
 
 
-def main_interface(requirements_file, processed_file, action):
-    try:
+def main_interface(requirements_file, processed_file, action):  # file, file, str
+    if requirements_file:
         data_dict = process_file(requirements_file)
-    except Exception as e:
+    else:
         data_dict = {}
     if action == Config.ACTION_BASIC_INFO:
-        sector_data = shape_sector_info(data_dict=data_dict, processed_data=None, action=action)
-    elif action in [Config.ACTION_STRATEGY_INFO, Config.ACTION_FINANCIAL_INFO]:
+        filename = shape_sector_info(data_dict=data_dict, processed_data=None, action=action)
+        return filename
+    elif action == Config.ACTION_STRATEGY_INFO:
         sector_data = pd.read_excel(processed_file.name)
         sector_data = shape_sector_info(data_dict=data_dict, processed_data=sector_data, action=action)
         if type(sector_data) == str:
             return sector_data
+    elif action == Config.ACTION_SELECTED_STRATEGY_INFO:
+        processed_data = pd.read_excel(processed_file.name, sheet_name=None)
+        selected_data = processed_data[Config.SELECTED_BY_FINANCIALS]
+        selected_data = selected_data.to_dict(orient="records")
+        [selected_data[c].update(get_strategy_info(company["ticker"], data_dict[Config.ACTION_STRATEGY_INFO])
+                                 ) for c, company in enumerate(selected_data)]
+        processed_data[Config.SELECTED_BY_FINANCIALS] = pd.DataFrame(selected_data)
+        filename = convert_to_excel(processed_data, filename="updated_strategic_data", index=False)
+        return filename
     elif action == Config.ACTION_UPDATE_INFO:
         filename = update_financial_info(processed_file)
         return filename
-    file_name = "processed_file.xlsx"
-    sector_data.to_excel(file_name, index=False)
-    return file_name
+    elif action == Config.ACTION_SELECT_BY_FINANCIALS:
+        filename = select_by_financials(processed_file)
+        return filename
+    processed_dataframes = pd.read_excel(processed_file.name, sheet_name=None)
+    if len(processed_dataframes.keys()) > 1:
+        processed_dataframes.pop(Config.MAIN_SHEET_NAME)
+        processed_dataframes = {**sector_data, **processed_dataframes}
+        filename = convert_to_excel(processed_dataframes, filename="updated_data", index=False)
+    else:
+        filename = convert_to_excel(sector_data, filename="updated_data", index=False)
+    return filename
 
 
 iface = gr.Interface(
     fn=main_interface, 
     inputs=[gr.inputs.File(label="Requirements file"), gr.inputs.File(label="(Optional) Populated file"),
-            gr.inputs.Dropdown([Config.ACTION_BASIC_INFO, Config.ACTION_STRATEGY_INFO, Config.ACTION_FINANCIAL_INFO,
-                                Config.ACTION_UPDATE_INFO],
+            gr.inputs.Dropdown([Config.ACTION_BASIC_INFO,
+                                Config.ACTION_STRATEGY_INFO,
+                                Config.ACTION_SELECTED_STRATEGY_INFO,
+                                Config.ACTION_UPDATE_INFO,
+                                Config.ACTION_SELECT_BY_FINANCIALS],
                                label="Select the type of information you want to extract")],
     outputs=gr.outputs.File()
 )

basic_handler.py

@@ -1,5 +1,9 @@
 from prompts import Prompts
 from common_functionality import send_openai_request
+import openai
+import os
+
+openai.api_key = os.getenv("OPENAI_API_KEY")
 
 
 def get_basic_info(sector: str, n_companies: int, required_fields: list) -> list:
@@ -9,18 +13,13 @@ def get_basic_info(sector: str, n_companies: int, required_fields: list) -> list
         ignore_companies = []
         for _ in range(n_companies // max_num_companies):
             prompt = Prompts.BASIC_INFORMATION_PROMPT.format(num_companies=max_num_companies,
-                                                             sector=sector, ignore_companies=ignore_companies,
+                                                             ignore_companies=ignore_companies,
                                                              required_fields=required_fields)
-            data.extend(send_openai_request(prompt))
-            ignore_companies = [i["name"] for i in data]
-        prompt = Prompts.BASIC_INFORMATION_PROMPT.format(num_companies=n_companies, sector=sector,
-                                                         ignore_companies=ignore_companies,
-                                                         required_fields=required_fields)
-        resp = send_openai_request(prompt)
-        if type(resp) == dict:
-            data.append(resp)
-        else:
-            data.extend(resp)
+            resp = send_openai_request(prompt)
+            if type(resp) == dict:
+                data.append(resp)
+            else:
+                data.extend(resp)
     else:
         prompt = Prompts.BASIC_INFORMATION_PROMPT_SINGLE_USE.format(num_companies=n_companies, sector=sector,
                                                                     required_fields=required_fields)

common_functionality.py

@@ -1,5 +1,21 @@
 import json
 import openai
+import pandas as pd
+import yfinance as yf
+
+from config import Config
+
+
+def convert_to_excel(dataframes: dict[pd.DataFrame], filename="updated_data", index=True) -> str:
+    """
+    Accepts a dictionary of dataframes and converts it to an Excel file
+    """
+    filename = f"{filename}.xlsx"
+    writer = pd.ExcelWriter(filename, engine='openpyxl')
+    for ticker, df in dataframes.items():
+        df.to_excel(writer, index=index, sheet_name=ticker)
+    writer.close()
+    return filename
 
 
 def send_openai_request(prompt: str, max_tokens=2000, temperature=0.1) -> dict:
@@ -18,3 +34,10 @@ def send_openai_request(prompt: str, max_tokens=2000, temperature=0.1) -> dict:
         presence_penalty=0
         )
     return json.loads(response.choices[0]["message"]["content"])
+
+
+def get_financial_columns(data: pd.DataFrame) -> list:
+    test = yf.Ticker("AAPL")
+    financial_columns = [x for x in data.columns if x in (test.info.keys() or Config.FINANCE_HISTORICAL_RETURN in x)
+                         and x != "Country"]
+    return financial_columns

config.py

@@ -1,11 +1,17 @@
 class Config:
 
-    ACTION_BASIC_INFO = "Basic"
+    ACTION_BASIC_INFO = "Basic+Financial"
     ACTION_STRATEGY_INFO = "Strategy"
-    ACTION_FINANCIAL_INFO = "Financial (Rolling)"
+    ACTION_SELECTED_STRATEGY_INFO = "Selected Companies Strategy"
+    ACTION_FINANCIAL_INFO = "Financial"
     ACTION_UPDATE_INFO = "Update financials"
+    ACTION_SELECT_BY_STRATEGY = "Select top best strategy companies"
+    ACTION_SELECT_BY_FINANCIALS = "Select top best financials companies"
 
     FINANCE_UPDATE_TIME = "last_updated"
     FINANCE_HISTORICAL_RETURN = "historical_return"
 
     MAIN_SHEET_NAME = "Companies"
+
+    SELECTED_BY_FINANCIALS = "Top Financials"
+    TEMP_NUMBER_TO_SELECT = 2

prompts.py

@@ -20,4 +20,10 @@ class Prompts:
                                   "flat JSON with the keys being the provided categories description and score. " \
                                   "Do not nest values within the JSON."
     FINANCIAL_INFORMATION_PROMPT = ""
+    SELECT_TOP_FINANCIALS_PROMPT = "Select the top {num_companies} companies with the best financials out of the " \
+                                   "companies in the provided list based on the financial metrics provided. For each" \
+                                   " company provide a short 2-3 bulletpoints justification why this company was " \
+                                   "selected. Format your answer as the following JSON:\n" \
+                                   "[{{ticker: company_ticker, justification: justification}}, ...]\n" \
+                                   "List of companies with financials: {companies}"
 

rolling_update.py

@@ -2,6 +2,7 @@ import pandas as pd
 import yfinance as yf
 from finance_handler import get_financial_info
 from config import Config
+from common_functionality import convert_to_excel, get_financial_columns
 
 
 def update_financial_info(data) -> str:
@@ -41,8 +42,7 @@ def initial_update_flow(data: pd.DataFrame) -> dict:
     dataframes: dictionary of dataframes with the updated info
     """
     tickers = data["ticker"].unique()
-    test = yf.Ticker("AAPL")
-    financial_columns = [x for x in data.columns if x in test.info.keys() or Config.FINANCE_HISTORICAL_RETURN in x]
+    financial_columns = get_financial_columns(data)
     processed_data = data.to_dict(orient="records")
     current_data = [data[data["ticker"] == ticker][financial_columns + [Config.FINANCE_UPDATE_TIME]
                                                    ].to_dict(orient="records")[0] for ticker in tickers]
@@ -66,7 +66,8 @@ def add_financial_history(dataframes: dict[pd.DataFrame]) -> dict[pd.DataFrame]:
     2. Updates the main tab with the most recent financial info for each company
 
     """
-    financial_columns = dataframes[list(dataframes.keys())[-1]].columns
+    companies_tickers_filtered = [i for i in dataframes.keys() if i not in Config.__dict__.values()][0]
+    financial_columns = dataframes[companies_tickers_filtered].columns
     for name, df in dataframes.items():
         if name == Config.MAIN_SHEET_NAME:
             df_dict = df.to_dict(orient="records")
@@ -74,20 +75,10 @@ def add_financial_history(dataframes: dict[pd.DataFrame]) -> dict[pd.DataFrame]:
              for c, company in enumerate(df_dict)]
             dataframes[name] = pd.DataFrame(df_dict)
             continue
+        if name == Config.SELECTED_BY_FINANCIALS:
+            continue
         df_dict = df.to_dict(orient="records")
         updated_info = dataframes[Config.MAIN_SHEET_NAME][dataframes[Config.MAIN_SHEET_NAME]["ticker"] == name][financial_columns].to_dict(orient="records")[0]
         df_dict.append(updated_info)
         dataframes[name] = pd.DataFrame(df_dict)
     return dataframes
-
-
-def convert_to_excel(dataframes: dict[pd.DataFrame], filename="updated_data", index=True) -> str:
-    """
-    Accepts a dictionary of dataframes and converts it to an Excel file
-    """
-    filename = f"{filename}.xlsx"
-    writer = pd.ExcelWriter(filename, engine='openpyxl')
-    for ticker, df in dataframes.items():
-        df.to_excel(writer, index=index, sheet_name=ticker)
-    writer.close()
-    return filename

select_top_financials.py

@@ -0,0 +1,33 @@
+import openai
+import pandas as pd
+
+from config import Config
+from common_functionality import convert_to_excel, get_financial_columns, send_openai_request
+from prompts import Prompts
+
+
+def select_by_financials(processed_file):
+    dataframes = pd.read_excel(processed_file.name, sheet_name=None)
+    selected_companies_df = get_selected_companies(dataframes[Config.MAIN_SHEET_NAME])
+    joined_data = {**dataframes, **selected_companies_df}
+    filename = convert_to_excel(joined_data, filename="selected_companies", index=False)
+    return filename
+
+
+def get_selected_companies(data: pd.DataFrame) -> dict:
+    financial_columns = get_financial_columns(data)
+    if Config.FINANCE_UPDATE_TIME in financial_columns:
+        financial_columns.remove(Config.FINANCE_UPDATE_TIME)
+    if "country" in financial_columns:
+        financial_columns.remove("country")
+    tickers = list(data["ticker"])
+    financial_data = data[financial_columns].to_dict(orient="records")
+    for c, company in enumerate(financial_data):
+        company.update({"ticker": tickers[c]})
+    prompt = Prompts.SELECT_TOP_FINANCIALS_PROMPT.format(num_companies=Config.TEMP_NUMBER_TO_SELECT,
+                                                         companies=financial_data)
+    response = send_openai_request(prompt=prompt)
+    selected_df = pd.DataFrame(response)
+    selected_df["justification"] = selected_df["justification"].apply(lambda x: " ".join(x))
+    selected_df = pd.merge(selected_df, data[financial_columns + ["ticker"]], on="ticker")
+    return {Config.SELECTED_BY_FINANCIALS: selected_df}

workflow.ipynb

@@ -0,0 +1,599 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-27T02:30:24.971742Z",
+     "start_time": "2023-09-27T02:30:24.766127Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import json\n",
+    "import datetime\n",
+    "import yfinance as yf\n",
+    "import openai\n",
+    "openai.api_key = \"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\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:11:08.848554Z",
+     "start_time": "2023-09-21T02:11:08.820339Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "                                              Sector\n0                                      Cybersecurity\n1                                                  2\n2     Name, ticker, subsector, country, description \n3  currentPrice, trailingPE, forwardPE, marketCap...\n4  Leadership and Governance, Market Position and...",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Sector</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Cybersecurity</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Name, ticker, subsector, country, description</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>currentPrice, trailingPE, forwardPE, marketCap...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Leadership and Governance, Market Position and...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 41,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "with open(\"/Users/artemkalyta/Downloads/input_v0.1-example.xlsx\", 'rb') as f:\n",
+    "    temp_data = pd.read_excel(f)\n",
+    "temp_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:11:15.777916Z",
+     "start_time": "2023-09-21T02:11:15.763160Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "                                              Sector\n0                                      Cybersecurity\n1                                                  2\n2     Name, ticker, subsector, country, description \n3  currentPrice, trailingPE, forwardPE, marketCap...\n4  Leadership and Governance, Market Position and...",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Sector</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Cybersecurity</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Name, ticker, subsector, country, description</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>currentPrice, trailingPE, forwardPE, marketCap...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Leadership and Governance, Market Position and...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 42,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data = pd.read_excel(\"/Users/artemkalyta/Downloads/input_v0.1-example.xlsx\")\n",
+    "data.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:16:38.142654Z",
+     "start_time": "2023-09-21T02:16:17.490453Z"
+    }
+   },
+   "outputs": [],
+   "source": [
+    "def get_formatting_columns(data: pd.DataFrame) -> dict:\n",
+    "    data_dict = {\"sector\": None, \"num_companies\": None, \"basic_info\": None, \"financial_info\": None, \"strategy_info\": None}\n",
+    "    data_dict[\"sector\"] = data[\"Sector\"].values.tolist()[0]\n",
+    "    data_dict[\"num_companies\"] = data[\"Sector\"].values.tolist()[1]\n",
+    "    basic_info = [x.strip().lower() for x in data[\"Sector\"].values.tolist()[2].split(\", \")]\n",
+    "    financial_info = [x.strip() for x in data[\"Sector\"].values.tolist()[3].split(\", \")]\n",
+    "    strategy_info = [x.strip() for x in data[\"Sector\"].values.tolist()[4].split(\", \")]\n",
+    "    data_dict[\"basic_info\"] = basic_info\n",
+    "    data_dict[\"financial_info\"] = financial_info\n",
+    "    data_dict[\"strategy_info\"] = strategy_info\n",
+    "    return data_dict\n",
+    "\n",
+    "def send_openai_request(prompt, max_tokens=2000):\n",
+    "    response = openai.ChatCompletion.create(\n",
+    "        model=\"gpt-4\",\n",
+    "        messages=[\n",
+    "            {\n",
+    "                \"role\": \"user\",\n",
+    "                \"content\": f\"{prompt}\"\n",
+    "            }\n",
+    "        ],\n",
+    "        temperature=0.1,\n",
+    "        max_tokens=max_tokens,\n",
+    "        top_p=1,\n",
+    "        frequency_penalty=0,\n",
+    "        presence_penalty=0\n",
+    "        )\n",
+    "    return json.loads(response.choices[0][\"message\"][\"content\"])\n",
+    "\n",
+    "\n",
+    "def get_basic_info(sector: str, n_companies: int, required_fields: list) -> list:\n",
+    "    max_num_companies = 20\n",
+    "    data = []\n",
+    "    if n_companies > max_num_companies:\n",
+    "        ignore_companies = []\n",
+    "        for _ in range(n_companies // max_num_companies):   \n",
+    "            prompt = f\"\"\"Provide a list of the {max_num_companies} most prominent companies in {sector} with the best potential growth. Companies that you should not mention: {ignore_companies}. Include all the relevant, concise, and descriptive information and format the response as the valid JSON with the following keys: {required_fields}\"\"\"\n",
+    "            data.extend(send_openai_request(prompt))\n",
+    "            ignore_companies = [i[\"name\"] for i in data]\n",
+    "        prompt = f\"\"\"Provide a list of the {n_companies % max_num_companies} most prominent companies in {sector} with the best potential growth. Companies that you should not mention: {ignore_companies}. Include all the relevant, concise, and descriptive information and format the response as the valid JSON with the following keys: {required_fields}\"\"\"\n",
+    "        resp = send_openai_request(prompt)\n",
+    "        if type(resp) == dict:\n",
+    "            data.append(resp)\n",
+    "        else:\n",
+    "            data.extend(resp)\n",
+    "    else:\n",
+    "        prompt = f\"\"\"Provide a list of the {n_companies} most prominent companies in {sector} with the best potential growth. Include all the relevant, concise, and descriptive information and format the response as the valid JSON with the following keys: {required_fields}\"\"\"\n",
+    "        data.extend(send_openai_request(prompt))\n",
+    "    return data\n",
+    "\n",
+    "\n",
+    "def get_strategy_info(company_name: str, categories: list) -> dict:\n",
+    "    prompt = f\"\"\"Please analyze {company_name} and give a descriptive, detailed, and very specific analysis (at least 5-6 sentences or more) using numbers when appropriate (only if you're absolutely confident in the numbers) on each of the following categories: {categories}. Each category should be described in detail. All of the information should be useful to a potential investor and should describe the advantages and drawbacks of investing in the company, you should approach every category only from a business and potential profit perspective. Please format your answer as a valid JSON with the keys being the provided categories.\"\"\"\n",
+    "    response = send_openai_request(prompt)\n",
+    "    return response\n",
+    "\n",
+    "def historical_return(company: yf.Ticker, history_years: int) -> float:\n",
+    "    history = company.history(period=f\"{history_years}y\")\n",
+    "    return (history[\"Close\"].iloc[-1] - history[\"Close\"].iloc[0]) / history[\"Close\"].iloc[0]\n",
+    "\n",
+    "def get_financial_info(ticker: str, metrics: list) -> dict:\n",
+    "    company = yf.Ticker(ticker)\n",
+    "    financial_info = {}\n",
+    "    for metric in metrics:\n",
+    "        if company.info.get(metric, None):\n",
+    "            financial_info.update({metric: company.info.get(metric)})\n",
+    "            continue\n",
+    "        if \"historical_return\" in metric:\n",
+    "            financial_info.update({metric: historical_return(company, metric[-1])})\n",
+    "    return financial_info\n",
+    "\n",
+    "def shape_sector_info(data_dict: dict) -> pd.DataFrame:\n",
+    "    sector_data = get_basic_info(data_dict[\"sector\"], data_dict[\"num_companies\"], data_dict[\"basic_info\"])\n",
+    "    for c, company in enumerate(sector_data):\n",
+    "        #sector_data[c].update(get_strategy_info(company[\"name\"], data_dict[\"strategy_info\"]))\n",
+    "        sector_data[c].update(get_financial_info(company[\"ticker\"], data_dict[\"financial_info\"]))\n",
+    "    sector_data = pd.DataFrame(sector_data)\n",
+    "    return sector_data\n",
+    "\n",
+    "\n",
+    "\n",
+    "data_dict = get_formatting_columns(data)\n",
+    "sector_data = shape_sector_info(data_dict)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "                        name ticker                subsector        country  \\\n0  CrowdStrike Holdings Inc.   CRWD  Software-Infrastructure  United States   \n1    Palo Alto Networks Inc.   PANW  Software-Infrastructure  United States   \n\n                                         description  currentPrice  forwardPE  \\\n0  CrowdStrike is a leading cybersecurity company...        163.89  46.825714   \n1  Palo Alto Networks is a global cybersecurity l...        235.30  36.765625   \n\n     marketCap      beta  recommendationMean  historical_return1  \\\n0  39134965760  0.959392                 1.8           -0.031669   \n1  72880881664  1.191908                 1.8            0.375058   \n\n   historical_return5  trailingPE  \n0            1.825690         NaN  \n1            2.075549      188.24  ",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>name</th>\n      <th>ticker</th>\n      <th>subsector</th>\n      <th>country</th>\n      <th>description</th>\n      <th>currentPrice</th>\n      <th>forwardPE</th>\n      <th>marketCap</th>\n      <th>beta</th>\n      <th>recommendationMean</th>\n      <th>historical_return1</th>\n      <th>historical_return5</th>\n      <th>trailingPE</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>CrowdStrike Holdings Inc.</td>\n      <td>CRWD</td>\n      <td>Software-Infrastructure</td>\n      <td>United States</td>\n      <td>CrowdStrike is a leading cybersecurity company...</td>\n      <td>163.89</td>\n      <td>46.825714</td>\n      <td>39134965760</td>\n      <td>0.959392</td>\n      <td>1.8</td>\n      <td>-0.031669</td>\n      <td>1.825690</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Palo Alto Networks Inc.</td>\n      <td>PANW</td>\n      <td>Software-Infrastructure</td>\n      <td>United States</td>\n      <td>Palo Alto Networks is a global cybersecurity l...</td>\n      <td>235.30</td>\n      <td>36.765625</td>\n      <td>72880881664</td>\n      <td>1.191908</td>\n      <td>1.8</td>\n      <td>0.375058</td>\n      <td>2.075549</td>\n      <td>188.24</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 53,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sector_data"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:17:24.897424Z",
+     "start_time": "2023-09-21T02:17:24.888427Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:17:28.507658Z",
+     "start_time": "2023-09-21T02:17:28.488571Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "                        name ticker                subsector        country  \\\n0  CrowdStrike Holdings Inc.   CRWD  Software-Infrastructure  United States   \n1    Palo Alto Networks Inc.   PANW  Software-Infrastructure  United States   \n\n                                         description  currentPrice  forwardPE  \\\n0  CrowdStrike is a leading cybersecurity company...        163.89  46.825714   \n1  Palo Alto Networks is a global cybersecurity l...        235.30  36.765625   \n\n     marketCap      beta  recommendationMean  historical_return1  \\\n0  39134965760  0.959392                 1.8           -0.031669   \n1  72880881664  1.191908                 1.8            0.375058   \n\n   historical_return5  trailingPE  \n0            1.825690         NaN  \n1            2.075549      188.24  ",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>name</th>\n      <th>ticker</th>\n      <th>subsector</th>\n      <th>country</th>\n      <th>description</th>\n      <th>currentPrice</th>\n      <th>forwardPE</th>\n      <th>marketCap</th>\n      <th>beta</th>\n      <th>recommendationMean</th>\n      <th>historical_return1</th>\n      <th>historical_return5</th>\n      <th>trailingPE</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>CrowdStrike Holdings Inc.</td>\n      <td>CRWD</td>\n      <td>Software-Infrastructure</td>\n      <td>United States</td>\n      <td>CrowdStrike is a leading cybersecurity company...</td>\n      <td>163.89</td>\n      <td>46.825714</td>\n      <td>39134965760</td>\n      <td>0.959392</td>\n      <td>1.8</td>\n      <td>-0.031669</td>\n      <td>1.825690</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Palo Alto Networks Inc.</td>\n      <td>PANW</td>\n      <td>Software-Infrastructure</td>\n      <td>United States</td>\n      <td>Palo Alto Networks is a global cybersecurity l...</td>\n      <td>235.30</td>\n      <td>36.765625</td>\n      <td>72880881664</td>\n      <td>1.191908</td>\n      <td>1.8</td>\n      <td>0.375058</td>\n      <td>2.075549</td>\n      <td>188.24</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 54,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sector_data.to_excel(\"/Users/artemkalyta/Downloads/output_v0.1-example.xlsx\")\n",
+    "sector_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-27T02:13:50.891551Z",
+     "start_time": "2023-09-27T02:13:50.886535Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['__class__', '__delattr__', '__delete__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__get__', '__getattribute__', '__getstate__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__isabstractmethod__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__set__', '__set_name__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'deleter', 'fdel', 'fget', 'fset', 'getter', 'setter']\n"
+     ]
+    }
+   ],
+   "source": [
+    "import yfinance as yf\n",
+    "print(dir(yf.ticker.Ticker.financials))\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T01:56:55.816136Z",
+     "start_time": "2023-09-21T01:56:55.777294Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "0.3457653580944472"
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def historical_return(ticker: str, history_years: int) -> float:\n",
+    "    company = yf.Ticker(ticker)\n",
+    "    history = company.history(period=f\"{history_years}y\")\n",
+    "    return (history[\"Close\"].iloc[-1] - history[\"Close\"].iloc[0]) / history[\"Close\"].iloc[0]\n",
+    "\n",
+    "historical_return(\"GOOG\", 1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:15:43.563900Z",
+     "start_time": "2023-09-21T02:15:43.372194Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "1.8"
+     },
+     "execution_count": 51,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tick = yf.Ticker(\"PANW\").info[\"recommendationMean\"]\n",
+    "tick"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T01:29:10.489157Z",
+     "start_time": "2023-09-21T01:29:10.032458Z"
+    }
+   },
+   "outputs": [
+    {
+     "ename": "KeyError",
+     "evalue": "'regularMarketPrice'",
+     "output_type": "error",
+     "traceback": [
+      "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m",
+      "\u001B[0;31mKeyError\u001B[0m                                  Traceback (most recent call last)",
+      "Cell \u001B[0;32mIn[4], line 3\u001B[0m\n\u001B[1;32m      1\u001B[0m \u001B[38;5;28;01mimport\u001B[39;00m \u001B[38;5;21;01myfinance\u001B[39;00m \u001B[38;5;28;01mas\u001B[39;00m \u001B[38;5;21;01myf\u001B[39;00m\n\u001B[1;32m      2\u001B[0m ticker \u001B[38;5;241m=\u001B[39m yf\u001B[38;5;241m.\u001B[39mTicker(\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mGOOGL\u001B[39m\u001B[38;5;124m'\u001B[39m)\u001B[38;5;241m.\u001B[39minfo\n\u001B[0;32m----> 3\u001B[0m marketPrice \u001B[38;5;241m=\u001B[39m \u001B[43mticker\u001B[49m\u001B[43m[\u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43mregularMarketPrice\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m]\u001B[49m\n\u001B[1;32m      4\u001B[0m previousClosePrice \u001B[38;5;241m=\u001B[39m ticker[\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mregularMarketPreviousClose\u001B[39m\u001B[38;5;124m'\u001B[39m]\n\u001B[1;32m      5\u001B[0m \u001B[38;5;28mprint\u001B[39m(\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mTicker Value: GOOGL\u001B[39m\u001B[38;5;124m'\u001B[39m)\n",
+      "\u001B[0;31mKeyError\u001B[0m: 'regularMarketPrice'"
+     ]
+    }
+   ],
+   "source": [
+    "[\"currentPrice\", \"trailingPE\", \"forwardPE\", \"marketCap\", \"beta\", \"recommendationMean\", \"historical_return1\", \"historical_return5\"]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-09-21T02:04:30.806580Z",
+     "start_time": "2023-09-21T02:04:30.802510Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "dict"
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "a = ticker.get(\"ebitdaa\", None)\n",
+    "print(a)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "{'sector': 'Cybersecurity',\n 'num_companies': 2,\n 'basic_info': ['name', 'ticker', 'subsector', 'country', 'description'],\n 'financian_info': ['P/E',\n  'Market Cap',\n  'Beta',\n  'Analyst Recommendation',\n  '1-year Return',\n  '5-year Return'],\n 'strategy_info': ['Leadership and Governance',\n  'Market Position and Competitive Advantage',\n  'R&D and Innovation',\n  'ESG Factors']}"
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_dict"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-21T01:36:45.859093Z",
+     "start_time": "2023-09-21T01:36:45.852532Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "dict_keys(['__module__', 'ACTION_BASIC_INFO', 'ACTION_STRATEGY_INFO', 'ACTION_FINANCIAL_INFO', 'ACTION_UPDATE_INFO', 'ACTION_SELECT_BY_STRATEGY', 'ACTION_SELECT_BY_FINANCIALS', 'FINANCE_UPDATE_TIME', 'FINANCE_HISTORICAL_RETURN', 'MAIN_SHEET_NAME', 'SELECTED_BY_FINANCIALS', '__dict__', '__weakref__', '__doc__', '__annotations__'])"
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from config import Config\n",
+    "\n",
+    "Config.__dict__.keys()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-02T03:08:06.743908Z",
+     "start_time": "2023-10-02T03:08:06.740531Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "['a', 'b']"
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "a = [\"a\", \"b\", \"c\"]\n",
+    "a.remove(\"c\")\n",
+    "a"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-02T02:42:04.990078Z",
+     "start_time": "2023-10-02T02:42:04.987609Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "1.9"
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ticker[\"recommendationMean\"]"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-21T01:51:38.725734Z",
+     "start_time": "2023-09-21T01:51:38.723808Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Select the top 5 companies with the best financials out of the companies in the provided list based on the financial metrics provided. For each company provide a short 2-3 bulletpoint justification why this company was selected. Format your answer as the following JSON:[{company_name: justification}, {company_name2: justification}]\n"
+     ]
+    }
+   ],
+   "source": [
+    "SELECT_TOP_FINANCIALS_PROMPT = \"Select the top {num_companies} companies with the best financials out of the \" \\\n",
+    "                                   \"companies in the provided list based on the financial metrics provided. For each\" \\\n",
+    "                                   \" company provide a short 2-3 bulletpoint justification why this company was \" \\\n",
+    "                                   \"selected. Format your answer as the following JSON:\" \\\n",
+    "                                   \"[{{company_name: justification}}, {{company_name2: justification}}]\"\n",
+    "print(SELECT_TOP_FINANCIALS_PROMPT.format(num_companies=5))"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-02T02:24:51.683836Z",
+     "start_time": "2023-10-02T02:24:51.679391Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "False"
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import yfinance as yf\n",
+    "\n",
+    "test = yf.Ticker(\"AAPL\")\n",
+    "\"ticker\" in test.info.keys()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-02T02:39:36.762388Z",
+     "start_time": "2023-10-02T02:39:36.559113Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "[{'name': 'CrowdStrike Holdings Inc.',\n  'ticker': 'CRWD',\n  'subsector': 'Software-Infrastructure',\n  'country': 'United States',\n  'description': 'CrowdStrike is a leading cybersecurity company that provides cloud-delivered solutions for endpoint protection, threat intelligence, and cyber attack response. Their AI-powered technology is designed to stop breaches and their Falcon platform offers instant visibility and protection across the enterprise and prevents attacks on endpoints on or off the network.',\n  'currentPrice': 163.89,\n  'forwardPE': 46.825714,\n  'marketCap': 39134965760,\n  'beta': 0.959392,\n  'recommendationMean': 1.8,\n  'historical_return1': -0.03166913211433715,\n  'historical_return5': 1.825689644649111,\n  'trailingPE': nan},\n {'name': 'Palo Alto Networks Inc.',\n  'ticker': 'PANW',\n  'subsector': 'Software-Infrastructure',\n  'country': 'United States',\n  'description': 'Palo Alto Networks is a global cybersecurity leader known for always challenging the security status quo. Their mission is to protect the digital way of life by preventing successful cyberattacks. They have pioneered the next generation of network security with innovative products that deliver cybersecurity across clouds, networks, and mobile devices.',\n  'currentPrice': 235.3,\n  'forwardPE': 36.765625,\n  'marketCap': 72880881664,\n  'beta': 1.191908,\n  'recommendationMean': 1.8,\n  'historical_return1': 0.3750584955932131,\n  'historical_return5': 2.075548954405327,\n  'trailingPE': 188.24}]"
+     },
+     "execution_count": 61,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "temp = sector_data.to_dict(orient='records')\n",
+    "temp"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-26T18:55:54.094789Z",
+     "start_time": "2023-09-26T18:55:54.094116Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "            price  market_cap\nupdated                      \n2021-09-01     90       90000\n2021-09-02    100      100000",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>price</th>\n      <th>market_cap</th>\n    </tr>\n    <tr>\n      <th>updated</th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2021-09-01</th>\n      <td>90</td>\n      <td>90000</td>\n    </tr>\n    <tr>\n      <th>2021-09-02</th>\n      <td>100</td>\n      <td>100000</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "a = {\"price\": 90, \"market_cap\": 90000, \"updated\": \"2021-09-01\"}\n",
+    "b = {\"price\": 100, \"market_cap\": 100000, \"updated\": \"2021-09-02\"}\n",
+    "df = pd.DataFrame([a, b]).set_index(\"updated\")\n",
+    "df"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-27T02:34:11.587757Z",
+     "start_time": "2023-09-27T02:34:11.581806Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "{'price': {'2021-09-01': 90, '2021-09-02': 100},\n 'market_cap': {'2021-09-01': 90000, '2021-09-02': 100000}}"
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.to_dict()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-09-27T18:19:53.124974Z",
+     "start_time": "2023-09-27T18:19:53.114800Z"
+    }
+   }
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "investor-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": "3.11.5"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}