CCFlows / CF_CodeDebugCritic.yaml
yeeef's picture
propagate flows changes
a899312
raw
history blame
4.26 kB
name: "CodeDebugCritic_Flow"
description: "Given a problem description and a candidate solution, test the code and provide useful feedback concerning the correctness of the solution and the potential mistakes."
# ~~~ Input interface specification ~~~
input_interface:
- "problem_description"
- "input_description"
- "output_description"
- "io_examples_and_explanation"
- "public_tests_individual_io"
- "code"
# ~~~ Output interface specification ~~~
output_interface:
- "testing_results_summary"
- "all_tests_passed"
- "code_feedback"
# ~~~ Flow specification ~~~
public_tests_key: "public_tests_individual_io"
subflows_config:
CodeTestingCritic:
_target_: .CF_CodeTesting.instantiate_from_default_config
CodeCriticWrongAttempt:
_target_: .CF_CodeCriticWrongAttempt.instantiate_from_default_config
topology:
# ~~~ Code Testing Critic ~~~
- goal: "Test the code on the public tests and provide a results summary."
### Input Interface
input_interface:
_target_: flows.interfaces.KeyInterface
additional_transformations:
- _target_: flows.data_transformations.KeyMatchInput
### Flow Specification
flow: CodeTestingCritic
### Output Interface
output_interface:
_target_: flows.interfaces.KeyInterface
additional_transformations:
- _target_: .src.data_transformations.CorrectnessFlag
input_key: "public_tests_results"
output_key: "all_tests_passed"
- _target_: .src.data_transformations.TestingResultsSummaryGeneration
output_key: "testing_results_summary"
single_test_error_message: True
no_error_template: |2-
${.issue_title}
All of the executed tests passed.
compilation_error_template: |2-
${.issue_title}
The execution resulted in a compilation error.
## Compilation error message:
{{error_message}}
timeout_error_template: |2-
${.issue_title}
The execution timed out, the solution is not efficient enough.
runtime_error_template: |2-
${.issue_title}
The execution resulted in a runtime error on the following test.
## [Failed test] Input
```
{{test_input}}
```
## [Failed test] Runtime error message
{{error_message}}
single_test_error_template: |2-
${.issue_title}
The Python code does not solve the problem in the problem description due to logical errors. It fails the following test:
## [Failed test] Input
```
{{test_input}}
```
## [Failed test] Expected output
```
{{expected_output}}
```
## [Failed test] Generated output
```
{{generated_output}}
```
all_tests_header: |2-
${.issue_title}
The Python code does not solve the problem in the problem description due to logical errors. It fails on the following tests.
test_error_template: |2-
## [Failed test {{idx}}]
### [Failed test {{idx}}] Input
```
{{test_input}}
```
### [Failed test {{idx}}] Expected output
```
{{expected_output}}
```
### [Failed test {{idx}}] Generated output
```
{{generated_output}}
```
tests_separator: "\n\n"
issue_title: "# Issue with the last proposed solution"
# ~~~ Feedback Generator ~~~
- goal: "Generate feedback grounded in the test results summary."
### Input Interface
input_interface:
_target_: flows.interfaces.KeyInterface
additional_transformations:
- _target_: flows.data_transformations.KeyMatchInput
### Flow Specification
flow: CodeCriticWrongAttempt
### Output Interface
output_interface:
_target_: flows.interfaces.KeyInterface
additional_transformations:
- _target_: flows.data_transformations.KeyRename
old_key2new_key:
api_output: "code_feedback"
reset: true