Wolfram Language Paclet Repository
Community-contributed installable additions to the Wolfram Language
Code generation by filling in templates using NLP techniques
Contributed by: Anton Antonov
Code generation by filling in templates using NLP techniques.
To install this paclet in your Wolfram Language environment,
evaluate this code:
PacletInstall["AntonAntonov/NLPTemplateEngine"]
To load the code after installation, evaluate this code:
Needs["AntonAntonov`NLPTemplateEngine`"]
Here we generate classifier creation code:
In[1]:= |
Out[1]= |
Here we generate quantile regression pipeline code:
In[2]:= |
Out[2]= |
Here we get an association result:
In[3]:= |
Out[3]= |
The pre-loaded templates are used to produce Machine Learning (ML) computational workflow specifications for: Classification, Latent Semantic Analysis, Quantile Regression, Recommendations, Random tabular data generation. The implementations are in Python, R, and WL. Here is example of Python recommender creation spec:
In[4]:= |
Out[4]= |
Here is an example of an R pipeline generation for Quantile Regression:
In[5]:= |
Out[5]= |
The function Concretize has Method option that can be given the values FindTextualAnswer, LLMTextualAnswer, or Automatic. Here an LLM is used to both classify the workflow spec and to find the parameters for the corresponding template:
In[6]:= |
Out[6]= |
Concretize can answer "introspection" questions. For example, here we get the WL template for recommendations:
In[7]:= |
Out[7]= |
Here we query the number of templates (known):
In[8]:= |
Out[8]= |
Load the NLP-Template-Engine:
In[9]:= |
Get the "training" templates data (from CSV file you have created or changed) for a new workflow ("SendMail"):
In[10]:= |
Out[11]= |
Add the ingested data for the new workflow (from the CSV file) into the NLP-Template-Engine:
In[12]:= |
Out[12]= |
Parse natural language specification with the newly ingested and on-boarded workflow ("SendMail"):
In[13]:= |
Out[13]= |
Here is a screenshot with an example of using Concretize for random tabular data generation:
Wolfram Language Version 13.2