encoderfile 0.6.2

Distribute and run transformer encoders with a single file.
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
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142

use std::collections::HashMap;

use encoderfile::{
    dev_utils::*,
    generated::{
        embedding::{
            EmbeddingRequest, EmbeddingResponse, embedding_inference_server::EmbeddingInference,
        },
        metadata::{GetModelMetadataRequest, GetModelMetadataResponse},
        sentence_embedding::{
            SentenceEmbeddingRequest, SentenceEmbeddingResponse,
            sentence_embedding_inference_server::SentenceEmbeddingInference,
        },
        sequence_classification::{
            SequenceClassificationRequest, SequenceClassificationResponse,
            sequence_classification_inference_server::SequenceClassificationInference,
        },
        token_classification::{
            TokenClassificationRequest, TokenClassificationResponse,
            token_classification_inference_server::TokenClassificationInference,
        },
    },
    transport::grpc::GrpcService,
};

macro_rules! test_grpc_service {
    (
        $mod_name:ident,
        $create_service:expr,
        $has_labels:expr,
        $predict_request:expr,
        $predict_response_ty:ty
    ) => {
        mod $mod_name {
            use super::*;

            #[tokio::test]
            async fn test_get_model_metadata() {
                let service = $create_service;

                let request = tonic::Request::new(GetModelMetadataRequest {});

                let response: GetModelMetadataResponse = service
                    .get_model_metadata(request)
                    .await
                    .unwrap()
                    .into_inner();

                println!("Model metadata: {:?}", response);

                if $has_labels {
                    assert!(!response.id2label.is_empty(), "id2label is an empty dict")
                } else {
                    assert!(
                        response.id2label.is_empty(),
                        "id2label is not an empty dict"
                    );
                }
            }

            #[tokio::test]
            async fn test_predict() {
                let service = $create_service;
                let n_inps = $predict_request.inputs.len();
                let request = tonic::Request::new($predict_request);

                let response: $predict_response_ty =
                    service.predict(request).await.unwrap().into_inner();

                assert!(
                    response.results.len() == n_inps,
                    "Mismatched number of results"
                );
                assert!(response.metadata.is_empty(), "Metadata isn't empty");
            }

            #[tokio::test]
            async fn test_predict_empty() {
                let service = $create_service;
                let mut inp = $predict_request;
                inp.inputs = vec![];
                let request = tonic::Request::new(inp);

                let response = service.predict(request).await;

                let correct_err = match response {
                    Ok(_) => false,
                    Err(e) => match e.code() {
                        tonic::Code::InvalidArgument => true,
                        _ => false,
                    },
                };

                assert!(correct_err, "Empty input doesn't result in correct code")
            }
        }
    };
}

test_grpc_service!(
    embedding_grpc_tests,
    { GrpcService::new(embedding_state()) },
    false,
    EmbeddingRequest {
        inputs: vec!["hello world".to_string(), "the quick brown fox".to_string()],
        metadata: HashMap::new(),
    },
    EmbeddingResponse
);

test_grpc_service!(
    sequence_classification_tests,
    { GrpcService::new(sequence_classification_state()) },
    true,
    SequenceClassificationRequest {
        inputs: vec!["hello world".to_string(), "the quick brown fox".to_string()],
        metadata: HashMap::new(),
    },
    SequenceClassificationResponse
);

test_grpc_service!(
    token_classification_tests,
    { GrpcService::new(token_classification_state()) },
    true,
    TokenClassificationRequest {
        inputs: vec!["hello world".to_string(), "the quick brown fox".to_string()],
        metadata: HashMap::new(),
    },
    TokenClassificationResponse
);

test_grpc_service!(
    sentence_embedding_tests,
    { GrpcService::new(sentence_embedding_state()) },
    false,
    SentenceEmbeddingRequest {
        inputs: vec!["hello world".to_string(), "the quick brown fox".to_string()],
        metadata: HashMap::new(),
    },
    SentenceEmbeddingResponse
);