Is create_openai_data_generator correct for my usecase?

[deepayansur]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import getpass
import os
import pandas as pd

if "OPENAI_API_KEY" not in os.environ:
    os.environ["OPENAI_API_KEY"] = getpass.getpass("Enter your OpenAI API key: ")

from langchain.prompts import FewShotPromptTemplate, PromptTemplate
from langchain_experimental.tabular_synthetic_data.openai import (
    OPENAI_TEMPLATE,
    create_openai_data_generator,
)
from langchain_experimental.tabular_synthetic_data.prompts import (
    SYNTHETIC_FEW_SHOT_PREFIX,
    SYNTHETIC_FEW_SHOT_SUFFIX,
)
from langchain_openai import ChatOpenAI
from pydantic import BaseModel

class L_NL(BaseModel):
    TQTL: str
    Explanation: str

examples = [
    {
        "L":"x>0",
        "Explanation":"x will always be greater than 0"
    },
    {
        "L":"x>a+b",
        "Explanation":"x will always be greater than sum of a and b"    
    }
]

sample_prompt = PromptTemplate( input_variables= ["L","Explanation"],template="L: {L}\n Explanation: {Explanation}")
print(sample_prompt)

tqtl_example =  pd.read_csv("L-GPT\Readable_logic_100.csv")## fetch from readable

# Define the few-shot template
prompt_template = FewShotPromptTemplate(
    prefix="Give a more humanized style translation of this time quality temporal logic sentence.",
    examples=examples,
    suffix="L: {L}\nExplanation:",
    input_variables=["L"],
    example_prompt=sample_prompt,
)

synthetic_data_generator = create_openai_data_generator(
    output_schema=L_NL,
    llm=ChatOpenAI(
        temperature=1,
        model = "gpt-3.5-turbo"
    ),  
    prompt=prompt_template,
)



answers = []
for index, row in tqtl_example.head(10).iterrows():
    print(row['L'])
    
    # prompt = prompt_template.format(TQTL=row['TQTL'])
    # Query the model
    synthetic_results = synthetic_data_generator.generate(
        subject= "Explanation for logic",
        TQTL = row['L'],
        runs=1,
    )
    answers.append((row['L'], synthetic_results[0].Explanation))

Description

so when I run this, during this section:
synthetic_results = synthetic_data_generator.generate(
subject= "Explanation for logic",...
I am getting an issue second loop onwards.
I debugged this and found that the create_openai_data_generator is using generate function from langchain_experimental\tabular_synthetic_data\base.py
over here in the Line 102:
self.results.append(result)
self._update_examples(result)
takes place which leads to synthetic_data_generator.llm_chain.prompt.examples[1]['example'] being changed to
{
'example': {
L: "x is smaller than 1",
Explanation: "x is smaller than 1"
}
}
this causes an error as the examples are not in same format and it looks fo 'L'

Should I do it manually or is there an alternate approach or am I making some mistake?

System Info

aiohappyeyeballs==2.4.3
aiohttp==3.10.10
aiosignal==1.3.1
altair==5.4.1
annotated-types==0.7.0
anyio==4.4.0
asttokens==2.4.1
attrs==24.2.0
blinker==1.8.2
cachetools==5.5.0
certifi==2024.7.4
charset-normalizer==3.4.0
click==8.1.7
colorama==0.4.6
comm==0.2.2
contourpy==1.3.0
cycler==0.12.1
dataclasses-json==0.6.7
debugpy==1.8.7
decorator==5.1.1
distro==1.9.0
executing==2.1.0
fonttools==4.53.1
frozenlist==1.5.0
gitdb==4.0.11
GitPython==3.1.43
graphviz==0.20.3
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.5
httpx==0.27.2
idna==3.8
ipykernel==6.29.5
ipython==8.29.0
jedi==0.19.1
Jinja2==3.1.4
jiter==0.5.0
joblib==1.4.2
jsonpatch==1.33
jsonpointer==3.0.0
jsonschema==4.23.0
jsonschema-specifications==2024.10.1
jupyter_client==8.6.3
jupyter_core==5.7.2
kiwisolver==1.4.5
langchain==0.3.4
langchain-community==0.3.3
langchain-core==0.3.12
langchain-experimental==0.3.2
langchain-openai==0.2.3
langchain-text-splitters==0.3.0
langsmith==0.1.137
markdown-it-py==3.0.0
MarkupSafe==3.0.2
marshmallow==3.23.0
matplotlib==3.9.2
matplotlib-inline==0.1.7
mdurl==0.1.2
multidict==6.1.0
mypy-extensions==1.0.0
narwhals==1.10.0
nest-asyncio==1.6.0
numpy==1.26.4
openai==1.52.2
orjson==3.10.10
packaging==24.1
pandas==2.2.2
parso==0.8.4
pillow==10.4.0
platformdirs==4.3.6
prompt_toolkit==3.0.48
propcache==0.2.0
protobuf==5.28.3
psutil==6.0.0
pure_eval==0.2.3
pyarrow==17.0.0
pydantic==2.8.2
pydantic-settings==2.6.0
pydantic_core==2.20.1
pydeck==0.9.1
Pygments==2.18.0
pyparsing==3.1.4
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
pytz==2024.1
pywin32==308
PyYAML==6.0.2
pyzmq==26.2.0
referencing==0.35.1
regex==2024.9.11
requests==2.32.3
requests-toolbelt==1.0.0
rich==13.9.3
rpds-py==0.20.0
scikit-learn==1.5.1
scipy==1.14.1
sexpdata==1.0.2
six==1.16.0
smmap==5.0.1
sniffio==1.3.1
SQLAlchemy==2.0.36
stack-data==0.6.3
streamlit==1.39.0
tenacity==9.0.0
threadpoolctl==3.5.0
tiktoken==0.8.0
toml==0.10.2
tornado==6.4.1
tqdm==4.66.5
traitlets==5.14.3
typing-inspect==0.9.0
typing_extensions==4.12.2
tzdata==2024.1
urllib3==2.2.3
watchdog==5.0.3
wcwidth==0.2.13
yarl==1.16.0