Leveraging UnDatas.io and deepseek to Analyze Tesla Gen Report: A Step-by-Step Guide

This Blog will introduce to you, dear readers, through a notebook example how to use the Undatas.io platform and the deepseek model to analyze the Tesla Gen report. In this Blog, we’ll take you step by step through the process, sharing detailed insights and useful tips that we’ve gathered during our exploration. You’ll get to see exactly how these tools work together to extract valuable information from the Tesla Gen report, and hopefully, it’ll give you a clear understanding of their potential applications in similar scenarios. Stay tuned as we dive deeper into this interesting topic!

1.UnDatas.io: A Game-Changer for Unstructured Data in Finance

UnDatas.io is a highly versatile platform that has found extensive applications across diverse industries. In the financial sector, it stands out with its unique and innovative features, revolutionizing the operations of financial professionals and the management strategies of financial institutions. It equips them with the tools necessary to handle complex unstructured data prevalent in financial reports, contracts, and market analyses, thereby enhancing decision-making processes.

2. Setting Up UnDatas.io and Integrating with OpenAI’s deepseek

To commence using UnDatas.io for math test paper analysis, the first step involves installing the UnDatas.io Python API library. Importing an UnDataIO object requires a token and an optional task name obtained from the UnDatas.io platform.

pip install -U -q openai undatasio

from undatasio.undatasio import UnDatasIO

undatasio_obj = UnDatasIO(UNDATASIO_API_KEY, task_name='PdfParserDemo')

Additionally, the OpenAI Python SDK needs to be installed as it is utilized to call the deepseek model. Initializing the OpenAI object necessitates applying for an API key, ensuring secure and authorized access to the model.

from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("API_KEY"),
    base_url="https://api.deepseek.com"
)

3. Extracting Valuable Information from Tsla-gen-report

The show_version function of the generated UnDatas.io object provides valuable insights by displaying all version information and file lists associated with the current token’s task name. This aids in organizing and understanding the available resources.

version_data = undatasio_obj.show_version()
version_data.data

This blog’s example selects the table in the first Financial Summary of tsla-20241023-gen.pdf as the sample content, as shown in the figure above.

Moreover, the get_result_type method of the UnDatas.io object is a powerful tool for extracting diverse information from PDF files, such as text, images, tables, titles, and inline equation details.

result = undatasio_obj.get_result_type(
    type_info=['table'],
    file_name='tsla-20241023-gen_test.pdf',
    version='v2'
)
print(result.data)

4. Build RAG pipeline over the Parsed Report

Use deepseek-chat model and set the system and user prompts.

We ask a question over the parsed markdown table and get back the right answer! We also ask a question over the table.

query1 = 'The amount of Energy generation and storage revenue in Q1-2024.'

response = client.chat.completions.create(
        model="deepseek-chat",
        messages=[
            {"role": "system", "content": "You are a data analysis expert. Please extract information from the data provided by the user. Note that only the information asked by the user should be returned, and nothing else should be returned. Data: %s" % (result.data, )},
            {"role": "user", "content": query1},
        ],
        stream=False
    )

res_data = response.choices[0].message.content
res_data

Code running result:

Let’s also try asking a question over another piece of the table.

query2 = 'The amount of Total revenues in Q2-2024.'

response = client.chat.completions.create(
        model="deepseek-chat",
        messages=[
            {"role": "system", "content": "You are a data analysis expert. Please extract information from the data provided by the user. Note that only the information asked by the user should be returned, and nothing else should be returned. Data: %s" % (result.data, )},
            {"role": "user", "content": query2},
        ],
        stream=False
    )

res_data = response.choices[0].message.content
res_data

Code running result:

5. Summary

In summary, UnDatas.io has emerged as a powerful tool with immense potential in the financial domain, especially when combined with OpenAI’s deepseek model. It enables efficient handling and analysis of unstructured financial data, as exemplified by our exploration of the Tsla-gen-report. From installation and data extraction to in-depth analysis using the RAG pipeline, this blog has detailed the step-by-step process of leveraging these technologies.

For financial professionals, it offers a streamlined way to access and interpret critical financial information, facilitating more informed decision-making. Institutions can enhance their data management strategies and gain a competitive edge in the market. Looking ahead, as technology continues to evolve, the integration of UnDatas.io and deepseek is likely to play an even more prominent role in shaping the future of financial data analysis, unlocking new levels of insights and driving innovation in the industry.