date-a-scientist

Query dataframes, find issue with your notebook snippets as if a professional data scientist was pair coding with you.

Currently just a thin wrapper around an amazing library called pandas-ai by sinaptik-ai!

How to use it?

from date_a_scientist import DateAScientist
import pandas as pd

df = pd.DataFrame(
    [
        {"name": "Alice", "age": 25, "city": "New York"},
        {"name": "Bob", "age": 30, "city": "Los Angeles"},
        {"name": "Charlie", "age": 35, "city": "Chicago"},
    ]
)
ds = DateAScientist(
    df=df,
    llm_openai_api_token=...,  # your OpenAI API token goes here
    llm_model_name="gpt-3.5-turbo",  # by default, it uses "gpt-4o"
)

# should return "Alice"
ds.chat("What is the name of the first person?")

Additionally we can pass a description of fields, so that more meaningful questions can be asked:

ds = DateAScientist(
    df=df,
    llm_openai_api_token=...,  # your OpenAI API token goes here
    llm_model_name="gpt-3.5-turbo",  # by default, it uses "gpt-4o"
    column_descriptions={
        "name": "The name of the person",
        "age": "The age of the person",
        "city": "The city where the person lives",
    },
)

ds = DateAScientist(
    df=df,
    llm_openai_api_token=...,  # your OpenAI API token goes here
    llm_model_name="gpt-3.5-turbo",  # by default, it uses "gpt-4o"
)

# should return DataFrame with Chicago rows
ds.chat("Who lives in Chicago?")

Finally if you want to get the code that was generated, you can use ds.code():

ds.code("Who lives in Chicago?")

which will return monokai styled code. If you want to return plain code, you can use:

ds.code("Who lives in Chicago?", return_as_string=True)

Inspirations

Metadata

Release files for date-a-scientist 0.1.19

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for date-a-scientist 0.1.19
File	Size	Uploaded
date_a_scientist-0.1.19.tar.gz	6.7 kB	Jan 10, 2025	Details

Built distribution (wheel)

Table of built distributions (wheels) for date-a-scientist 0.1.19
File	Interpreter	ABI	Platform	Reset
date_a_scientist-0.1.19-py3-none-any.whl 7.2 kB Jan 10, 2025	Python 3	none	any	Details

Total release size: 13.8 kB

Release files / date_a_scientist-0.1.19.tar.gz

Download URL	date_a_scientist-0.1.19.tar.gz
Size	6.7 kB
Tags	Source
SHA-256 checksum How to use checksums	`6bfbbc83a33c32b6b04c4399de41b109b0c6cecc651fa27c936ec45cde1c9006`
BLAKE2b-256 checksum How to use checksums	`2f3ee39992e5386c9d409fe1dd2d9a517612a5efd118acec4420fe348d921d2d`
Upload date	Jan 10, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`poetry/1.8.3 CPython/3.11.3 Darwin/24.0.0`

Release files / date_a_scientist-0.1.19-py3-none-any.whl

Download URL	date_a_scientist-0.1.19-py3-none-any.whl
Size	7.2 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`bc1f9815ca8911f624649e5283464f2d86a6a7332b515e59e8239f2abca53efb`
BLAKE2b-256 checksum How to use checksums	`67089be71365a391f04c04f4038482a8cfca57b7dc0d1903a59d558f262995fe`
Upload date	Jan 10, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`poetry/1.8.3 CPython/3.11.3 Darwin/24.0.0`