Machine Learning-Assisted Code Generation in Directed Acyclic Graph-Driven Notebook Environment

A non-transitory computer-readable medium comprising memory with instructions encoded thereon that, when executed, cause one or more processors to perform operations, the instructions comprising instructions to:

• receive, in a code cell connected to a plurality of cells in a graph structure, a natural language command to generate code;
• determine, using directed edges of the graph structure, a set of precedent cells from which the code cell depends;
• input into a machine learning model the natural language command and values from the set of precedent cells pertaining to the code;
• receive, as output from the machine learning model, generated code;
• and
• update the code cell to include the generated code.

A system comprising:

• memory with instructions encoded thereon;
• and
• one or more processors that, when executing the instructions, are caused to perform operations comprising:
• receiving, in a code cell connected to a plurality of cells in a graph structure, a natural language command to generate code;
• determining, using directed edges of the graph structure, a set of precedent cells from which the code cell depends;
• inputting into a machine learning model the natural language command and values from the set of precedent cells pertaining to the code;
• receiving, as output from the machine learning model, generated code;
• and
• updating the code cell to include the generated code.