diffai-core 0.5.1

Core library for AI/ML diff analysis - PyTorch, Safetensors, tensor statistics
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

use serde_json::Value;

use crate::types::DiffResult;

// Submodules
mod impact;
mod methods;
mod precision;
mod types;

// Internal use only - no re-exports needed as types are used within submodules
use impact::analyze_quantization_impact;
use methods::analyze_quantization_methods;
use precision::analyze_quantization_precision;

// A5-3: QUANTIZATION ANALYSIS - Low Priority ML Feature
// ============================================================================

// A5-3: QuantizationAnalysis - Model quantization and precision analysis
pub fn analyze_quantization_patterns(
    old_model: &Value,
    new_model: &Value,
    results: &mut Vec<DiffResult>,
) {
    if let (Value::Object(old_obj), Value::Object(new_obj)) = (old_model, new_model) {
        // Quantization precision analysis
        if let Some((old_prec, new_prec)) = analyze_quantization_precision(old_obj, new_obj) {
            results.push(DiffResult::ModelArchitectureChanged(
                "quantization_precision".to_string(),
                old_prec,
                new_prec,
            ));
        }

        // Quantization method analysis
        if let Some((old_method, new_method)) = analyze_quantization_methods(old_obj, new_obj) {
            results.push(DiffResult::ModelArchitectureChanged(
                "quantization_methods".to_string(),
                old_method,
                new_method,
            ));
        }

        // Quantization impact analysis
        if let Some((old_impact, new_impact)) = analyze_quantization_impact(old_obj, new_obj) {
            results.push(DiffResult::ModelArchitectureChanged(
                "quantization_impact".to_string(),
                old_impact,
                new_impact,
            ));
        }
    }
}