oxirs-vec 0.2.4

Vector index abstractions for semantic similarity and AI-augmented querying
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
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97

//! # SentenceTransformerGenerator - model_details_group Methods
//!
//! This module contains method implementations for `SentenceTransformerGenerator`.
//!
//! 🤖 Generated with [SplitRS](https://github.com/cool-japan/splitrs)

use super::types::{ModelDetails, TransformerModelType};

use super::sentencetransformergenerator_type::SentenceTransformerGenerator;

impl SentenceTransformerGenerator {
    /// Get detailed information about the current model
    pub fn model_details(&self) -> ModelDetails {
        self.get_model_details()
    }
    /// Get model-specific vocabulary size and training details
    pub(super) fn get_model_details(&self) -> ModelDetails {
        match &self.model_type {
            TransformerModelType::BERT => ModelDetails {
                vocab_size: 30522,
                num_layers: 12,
                num_attention_heads: 12,
                hidden_size: 768,
                intermediate_size: 3072,
                max_position_embeddings: 512,
                supports_languages: vec!["en".to_string()],
                model_size_mb: 440,
                typical_inference_time_ms: 50,
            },
            TransformerModelType::RoBERTa => ModelDetails {
                vocab_size: 50265,
                num_layers: 12,
                num_attention_heads: 12,
                hidden_size: 768,
                intermediate_size: 3072,
                max_position_embeddings: 514,
                supports_languages: vec!["en".to_string()],
                model_size_mb: 470,
                typical_inference_time_ms: 55,
            },
            TransformerModelType::DistilBERT => ModelDetails {
                vocab_size: 30522,
                num_layers: 6,
                num_attention_heads: 12,
                hidden_size: 384,
                intermediate_size: 1536,
                max_position_embeddings: 512,
                supports_languages: vec!["en".to_string()],
                model_size_mb: 250,
                typical_inference_time_ms: 25,
            },
            TransformerModelType::MultiBERT => ModelDetails {
                vocab_size: 120000,
                num_layers: 12,
                num_attention_heads: 12,
                hidden_size: 768,
                intermediate_size: 3072,
                max_position_embeddings: 512,
                supports_languages: vec![
                    "en".to_string(),
                    "de".to_string(),
                    "fr".to_string(),
                    "es".to_string(),
                    "it".to_string(),
                    "pt".to_string(),
                    "ru".to_string(),
                    "zh".to_string(),
                    "ja".to_string(),
                    "ko".to_string(),
                    "ar".to_string(),
                    "hi".to_string(),
                    "th".to_string(),
                    "tr".to_string(),
                    "pl".to_string(),
                    "nl".to_string(),
                    "sv".to_string(),
                    "da".to_string(),
                    "no".to_string(),
                    "fi".to_string(),
                ],
                model_size_mb: 670,
                typical_inference_time_ms: 70,
            },
            TransformerModelType::Custom(_path) => ModelDetails {
                vocab_size: 50000,
                num_layers: 12,
                num_attention_heads: 12,
                hidden_size: self.config.dimensions,
                intermediate_size: self.config.dimensions * 4,
                max_position_embeddings: self.config.max_sequence_length,
                supports_languages: vec!["unknown".to_string()],
                model_size_mb: 500,
                typical_inference_time_ms: 60,
            },
        }
    }
}