Skip to main content

A Python module to explain exceptions in execution-time using a magic cell

Project description

explainable-exceptions

Artificial Intelligence is being used in several fields, including computer engineering and programming.

During this work, we pretend to facilitate the process of programming as much as possible adding the option to use a NLP (Natural Language Processing) model to help the user to understand and solve the exceptions that may appear while programming.

As simple as using a magic cell, the user will be able to send the exception to the huggingface chat and obtain a possible solution to solve the exception. In this process, the user must be logged in his huggingchat account. Credentials can be saved to be requested only once.

After receiving the response, the user will be able to request another response if the given one is not helpful enough.

Install

To install explainable_exceptions you must use the following pip command:

pip install explainable_exceptions

How to use

In order to use the magic cell, we need to import it from the core of explainable_exceptions. The cell is called explain.

After that, we can label cells using the magic cell command (%%). We can provide a huggingchat user or we can just use the command (if there are not credentials, then it will ask to generate them).

If an exception is raised inside a magic cell, it will be sent to the huggingchat model to generate a possible solution. This response will be shown by the jupyter standard output.

from explainable_exceptions.core import explain

A usage example could be:

%%explain `username`

import random
random.choice([])

Where username must be replaced (if provided) by the username of the huggingchat account that will be used to request information.

#%%explain ruescog

import random
random.choice([])
CRITICAL:root:Traceback (most recent call last):
  File "/home/ruescog/explainable-exceptions/explainable_exceptions/core.py", line 81, in explain
    exec(cell, globals(), local_ns)
  File "<string>", line 5, in <module>
  File "/grupoa/config/miniconda3/lib/python3.8/random.py", line 290, in choice
    raise IndexError('Cannot choose from an empty sequence') from None
IndexError: Cannot choose from an empty sequence

WARNING:root:Be careful, the following response has been generated automatically by a Natural Language Processing Model, so the answer may be incorrect or false.

Huggingchat response (online version):

That error is raised by random.choice() function which is used for generating random elements from iterables like lists, strings etc., and it seems that there might be no element available to pick from. To fix the issue you need to ensure that either pass some argument to the function or check if any item exists before calling the function on that object. You can refer to python documentation for more details : Random Functions . In case you face similar issues please feel free to ask me , i am here to help you :) . Also I am just starting so would love if you could rate interact with me as well.


Do you need another answer? (y/n) n

IndexError: Cannot choose from an empty sequence

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

explainable-exceptions-1.0.2.tar.gz (9.0 kB view hashes)

Uploaded Source

Built Distribution

explainable_exceptions-1.0.2-py3-none-any.whl (8.7 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page