On this tutorial, we exhibit how PraisonAI Agents can elevate your knowledge evaluation from handbook scripting to a totally autonomous, AI-driven pipeline. In a number of natural-language prompts, you’ll be taught to orchestrate each stage of the workflow, loading CSV or Excel recordsdata, filtering rows, summarizing traits, grouping by customized fields, pivoting tables, and exporting outcomes to each CSV and Excel, with out writing conventional Pandas code. On this implementation, beneath the hood, PraisonAI leverages Google Gemini to interpret your directions and invoke the suitable instruments. On the similar time, options akin to self-reflection and verbose logging offer you full visibility into every intermediate reasoning step.
!pip set up "praisonaiagents[llm]"We set up the core PraisonAI Brokers library, together with its LLM integration extras, which herald all vital dependencies (akin to Litellm and Gemini connectors) to drive autonomous workflows with massive language fashions.
import os
os.environ["GEMINI_API_KEY"] = "Use Your API Key"
llm_id = "gemini/gemini-1.5-flash-8b"We configure your atmosphere for Gemini entry by setting your API key, then specify which Gemini mannequin (the “1.5-flash-8b” variant) the PraisonAI Agent ought to use as its LLM backend.
from google.colab import recordsdata
uploaded = recordsdata.add()
csv_path = subsequent(iter(uploaded))
print("Loaded:", csv_path)We leverage Colab’s file‐add widget to allow you to choose a neighborhood CSV, seize its filename into csv_path, and print a affirmation, making it simple to deliver your knowledge into the pocket book interactively.
from praisonaiagents import Agent
from praisonaiagents.instruments import (
read_csv, filter_data, get_summary, group_by, pivot_table, write_csv
)
agent = Agent(
directions="You're a Information Analyst Agent utilizing Google Gemini.",
llm=llm_id,
instruments=[
read_csv, filter_data, get_summary, group_by, pivot_table, write_csv
],
self_reflect=True,
verbose=True
)We instantiate a PraisonAI Agent wired to Google Gemini, equipping it with knowledge‐evaluation instruments (CSV I/O, filtering, summarization, grouping, pivoting, and export). Enabling self-reflect permits the agent to critique its reasoning, whereas verbose mode streams detailed tool-invocation logs for transparency.
end result = agent.begin(f"""
1. read_csv to load knowledge from "{csv_path}"
2. get_summary to stipulate total traits
3. filter_data to maintain rows the place Shut > 800
4. group_by 12 months to common closing value
5. pivot_table to format the output desk
""")
print(end result)We ship a transparent, step-by-step immediate to your PraisonAI Agent, instructing it to load the CSV, summarize total traits, filter for closing costs over $ 800, compute yearly averages, and pivot the desk. The agent then prints out the mixed response (together with any generated abstract or knowledge output).
In conclusion, we have now constructed an end-to-end knowledge pipeline powered by PraisonAI Brokers and Gemini, which matches from uncooked knowledge add to insightful visualizations and downloadable studies in only a few cells. We’ve seen how PraisonAI’s declarative toolset replaces dozens of traces of boilerplate code with concise, human-readable steps, and the way built-in mechanisms, akin to end result caching and dual-mode API invocation, guarantee each effectivity and reliability.
Sources
Additionally, don’t overlook to observe us on Twitter and be a part of our Telegram Channel and LinkedIn Group. Don’t Neglect to affix our 90k+ ML SubReddit.
Asif Razzaq is the CEO of Marktechpost Media Inc.. As a visionary entrepreneur and engineer, Asif is dedicated to harnessing the potential of Synthetic Intelligence for social good. His most up-to-date endeavor is the launch of an Synthetic Intelligence Media Platform, Marktechpost, which stands out for its in-depth protection of machine studying and deep studying information that’s each technically sound and simply comprehensible by a large viewers. The platform boasts of over 2 million month-to-month views, illustrating its recognition amongst audiences.
