wehub-resource-sync
e768098d0e
tools_continuous_delivery / Private PyPI non-main branch release (push) Has been skipped
tools_continuous_delivery / Private PyPI main branch release (push) Failing after 2m42s
Publish Promptflow Doc / Build (push) Has been cancelled
Publish Promptflow Doc / Deploy (push) Has been cancelled
Flake8 Lint / flake8 (push) Has been cancelled
Spell check CI / Spell_Check (push) Has been cancelled
63 lines
2.5 KiB
Markdown
63 lines
2.5 KiB
Markdown
# Generate Python docstring
|
||
This example can help you automatically generate Python code's docstring and return the modified code.
|
||
|
||
Tools used in this flow：
|
||
- `load_code` tool, it can load code from a file path.
|
||
- Load content from a local file.
|
||
- Loading content from a remote URL, currently loading HTML content, not just code.
|
||
- `divide_code` tool, it can divide code into code blocks.
|
||
- To avoid files that are too long and exceed the token limit, it is necessary to split the file.
|
||
- Avoid using the same function (such as __init__(self)) to generate docstrings in the same one file, which may cause confusion when adding docstrings to the corresponding functions in the future.
|
||
- `generate_docstring` tool, it can generate docstring for a code block, and merge docstring into origin code.
|
||
|
||
## What you will learn
|
||
|
||
In this flow, you will learn
|
||
- How to compose an auto generate docstring flow.
|
||
- How to use different LLM APIs to request LLM, including synchronous/asynchronous APIs, chat/completion APIs.
|
||
- How to use asynchronous multiple coroutine approach to request LLM API.
|
||
- How to construct a prompt.
|
||
|
||
## Prerequisites
|
||
|
||
### Install promptflow sdk and other dependencies:
|
||
```bash
|
||
pip install -r requirements.txt
|
||
```
|
||
|
||
### Create connection for LLM to use
|
||
```bash
|
||
# Override keys with --set to avoid yaml file changes
|
||
pf connection create --file ../../../connections/azure_openai.yml --set api_key=<your_api_key> api_base=<your_api_base>
|
||
```
|
||
Note:
|
||
The [azure_openai.yml](../../../connections/azure_openai.yml) file is located in connections folder.
|
||
We are using connection named `open_ai_connection`in [flow.dag.yaml](flow.dag.yaml).
|
||
|
||
## Execute with Promptflow
|
||
### Execute with SDK
|
||
`python main.py --source <your_file_path>`
|
||
**Note**: the file path should be a python file path, default is `./azure_open_ai.py`.
|
||
|
||
A webpage will be generated, displaying diff:
|
||
|
||
|
||
|
||
### Execute with CLI
|
||
```bash
|
||
# run flow with default file path in flow.dag.yaml
|
||
pf flow test --flow .
|
||
|
||
# run flow with file path
|
||
pf flow test --flow . --inputs source="./azure_open_ai.py"
|
||
```
|
||
|
||
```bash
|
||
# run flow with batch data
|
||
pf run create --flow . --data ./data.jsonl --name auto_generate_docstring --column-mapping source='${data.source}'
|
||
```
|
||
Output the code after add the docstring.
|
||
|
||
You can also skip providing `column-mapping` if provided data has same column name as the flow.
|
||
Reference [here](https://aka.ms/pf/column-mapping) for default behavior when `column-mapping` not provided in CLI.
|