kproc 0.7.0

Knowledge Processing library.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59

//! Decoders, from graph to text.

use crate::{
  processor,
  processors::rag::{self, graph::DecoderOutputs},
  Result,
};
use kproc_llm::prelude::*;
use kproc_pmacros::Processor;

/// Decode a graph into a text.
#[derive(Processor)]
#[streams(rag::graph::DecoderInputs, rag::graph::DecoderOutputs)]
pub struct Decoder<T: LargeLanguageModel>
{
  model: T,
}

impl<T: LargeLanguageModel + Sync + Send> Decoder<T>
{
  /// Create a new text encoder for the model
  pub fn new(model: T) -> Self
  {
    Self { model }
  }
}

impl<T: LargeLanguageModel + Sync + Send> processor::ValueProcessor for Decoder<T>
{
  #[allow(clippy::manual_async_fn)] // https://github.com/rust-lang/rust-clippy/issues/12664
  fn process_one(
    &self,
    inputs: Self::Inputs,
  ) -> impl std::future::Future<Output = Result<Self::Outputs>> + Send
  {
    async move {
      let system =
        r#"You are a data scientist working for a company that is building a graph database. Your task is to take a graph from a database and turn it into a sentence that can be understood by humans. The Nodes are provided in the form [ENTITY_ID, TYPE, PROPERTIES] and the relationships are in the form [ENTITY_ID_1, RELATIONSHIP, ENTITY_ID_2, PROPERTIES].

Example:
Data:
{
"nodes": [["alice", "Person", {"age": 25, "occupation": "lawyer", "name":"Alice"}], ["bob", "Person", {"occupation": "journalist", "name": "Bob"}], ["alice.com", "Webpage", {"url": "www.alice.com"}], ["bob.com", "Webpage", {"url": "www.bob.com"}]],
"relationships": [["alice", "roommate", "bob", {"start": 2021}], ["alice", "owns", "alice.com", {}], ["bob", "owns", "bob.com", {}]]
}
Output: Alice lawyer and is 25 years old and Bob is her roommate since 2001. Bob works as a journalist. Alice owns a the webpage www.alice.com and Bob owns the webpage www.bob.com.

Generate a text for the following graph."#
            .to_string();

      let prompt = serde_json::to_string(&inputs.graph)?;
      let prompt = GenerationPrompt::prompt(prompt).system(system.to_string());

      let res = self.model.generate(prompt)?.await?;

      Ok(DecoderOutputs { text: res })
    }
  }
}