phowell
/
pocketflow
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update doc

This commit is contained in:
zachary62 2025-01-02 00:25:37 +00:00
parent c39cc22820
commit 04261bdade
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
docs/structure.md
View File

@ -16,9 +16,9 @@ There are several approaches to achieve a structured output:
In practice, **Prompting** is simple and reliable for modern LLMs. In practice, **Prompting** is simple and reliable for modern LLMs.
## Example Use Cases ### Example Use Cases
1. **Extracting Key Information** - Extracting Key Information
```yaml ```yaml
product: product:
@ -29,7 +29,7 @@ product:
Recommended for advanced users. Recommended for advanced users.
``` ```
2. **Summarizing Documents into Bullet Points** - Summarizing Documents into Bullet Points
```yaml ```yaml
summary: summary:
@ -38,7 +38,7 @@ summary:
- Suitable for all skill levels. - Suitable for all skill levels.
``` ```
3. **Generating Configuration Files** - Generating Configuration Files
```yaml ```yaml
server: server:
@ -50,8 +50,8 @@ server:
## Prompt Engineering ## Prompt Engineering
When prompting the LLM to produce **structured** output: When prompting the LLM to produce **structured** output:
1. **Wrap** the structure in code fences (e.g., ```yaml). 1. **Wrap** the structure in code fences (e.g., ` ```yaml`).
2. **Validate** that all required fields exist (and retry if necessary). 2. **Validate** that all required fields exist (and let `Node` handles retry).
### Example Text Summarization ### Example Text Summarization
@ -73,7 +73,7 @@ summary:
```""" ```"""
response = call_llm(prompt) response = call_llm(prompt)
yaml_str = response.split("```yaml")[1].split("```")[0].strip() yaml_str = response.split("```yaml")[1].split("```")[0].strip()
import yaml import yaml
structured_result = yaml.safe_load(yaml_str) structured_result = yaml.safe_load(yaml_str)