Example code for accomplishing common tasks with the LangChain library
Settings in the TripleA are very important. Before running long and important processes, make sure that the configuration variables are correct. The check_config.py
is a simple example of checking settings
After performing various steps on the collection of articles in TripleA, a lot of information is collected. If we want to get an output from this information that includes information related to questions and answers from artificial intelligence, the export_engine function should be used. This function allows you to write the functions related to filtering and transformation of the data model and outputs as you wish. and provides a list of normalized dictionaries. You can convert this list to csv files by using convert_unified2csv_dynamically function in Class Converter
.
sample-export-engine-advanced.py
sample_reset_llm_flag_with_fx.py
[sample_update_response.py
]
sample_update_response_jsondecodererror.py
In this method, special methods are used, which can be used in general for any structured LLM answer.
The remodel_llm_response
function is produced as a dynamic engine that allows changing the LLM response model using another customer order function such as fx_response_remodel
.