rbert 0.4.0

A simple interface for Bert embeddings
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
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246

use kalosm_model_types::FileSource;

const SNOWFLAKE_EMBEDDING_PREFIX: &str =
    "Represent this sentence for searching relevant passages: ";

/// A the source of a [`crate::Bert`] model
pub struct BertSource {
    pub(crate) search_embedding_prefix: Option<String>,
    pub(crate) config: FileSource,
    pub(crate) tokenizer: FileSource,
    pub(crate) model: FileSource,
}

impl BertSource {
    /// Create a new [`BertSource`] for embedding plain text
    pub fn new() -> Self {
        Self::default()
    }

    /// Create a new [`BertSource`] for embedding text for search
    pub fn new_for_search() -> Self {
        Self::snowflake_arctic_embed_extra_small()
    }

    /// Set the model to use, check out available models: <https://huggingface.co/models?library=sentence-transformers&sort=trending>
    pub fn with_model(mut self, model: FileSource) -> Self {
        self.model = model;
        self
    }

    /// Set the tokenizer to use
    pub fn with_tokenizer(mut self, tokenizer: FileSource) -> Self {
        self.tokenizer = tokenizer;
        self
    }

    /// Set the config to use
    pub fn with_config(mut self, config: FileSource) -> Self {
        self.config = config;
        self
    }

    /// Set the prefix to use when embedding search queries
    pub(crate) fn with_search_embedding_prefix(
        mut self,
        prefix: impl Into<Option<String>>,
    ) -> Self {
        self.search_embedding_prefix = prefix.into();
        self
    }

    /// Create a new [`BertSource`] with the BGE large english preset
    pub fn bge_large_en() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "BAAI/bge-large-en-v1.5".to_string(),
                "refs/pr/5".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "BAAI/bge-large-en-v1.5".to_string(),
                "refs/pr/5".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "BAAI/bge-large-en-v1.5".to_string(),
                "refs/pr/5".to_string(),
                "config.json".to_string(),
            ))
    }

    /// Create a new [`BertSource`] with the BGE base english preset
    pub fn bge_base_en() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "BAAI/bge-base-en-v1.5".to_string(),
                "refs/pr/1".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "BAAI/bge-base-en-v1.5".to_string(),
                "refs/pr/1".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "BAAI/bge-base-en-v1.5".to_string(),
                "refs/pr/1".to_string(),
                "config.json".to_string(),
            ))
    }

    /// Create a new [`BertSource`] with the BGE small english preset
    pub fn bge_small_en() -> Self {
        Self {
            config: FileSource::huggingface(
                "BAAI/bge-small-en-v1.5".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ),
            tokenizer: FileSource::huggingface(
                "BAAI/bge-small-en-v1.5".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ),
            model: FileSource::huggingface(
                "BAAI/bge-small-en-v1.5".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ),
            search_embedding_prefix: None,
        }
    }

    /// Create a new [`BertSource`] with the MiniLM-L6-v2 preset
    pub fn mini_lm_l6_v2() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "sentence-transformers/all-MiniLM-L6-v2".to_string(),
                "refs/pr/21".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "sentence-transformers/all-MiniLM-L6-v2".to_string(),
                "refs/pr/21".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "sentence-transformers/all-MiniLM-L6-v2".to_string(),
                "refs/pr/21".to_string(),
                "config.json".to_string(),
            ))
    }

    /// Create a new [`BertSource`] with the [snowflake-arctic-embed-xs](https://huggingface.co/Snowflake/snowflake-arctic-embed-xs) model
    pub fn snowflake_arctic_embed_extra_small() -> Self {
        Self::default()
            .with_config(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-xs".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-xs".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_model(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-xs".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_search_embedding_prefix(SNOWFLAKE_EMBEDDING_PREFIX.to_string())
    }

    /// Create a new [`BertSource`] with the [snowflake-arctic-embed-s](https://huggingface.co/Snowflake/snowflake-arctic-embed-s) model
    pub fn snowflake_arctic_embed_small() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-s".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-s".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-s".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ))
            .with_search_embedding_prefix(SNOWFLAKE_EMBEDDING_PREFIX.to_string())
    }

    /// Create a new [`BertSource`] with the [snowflake-arctic-embed-m](https://huggingface.co/Snowflake/snowflake-arctic-embed-m) model
    pub fn snowflake_arctic_embed_medium() -> Self {
        Self::default()
            .with_config(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_model(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_search_embedding_prefix(SNOWFLAKE_EMBEDDING_PREFIX.to_string())
    }

    /// Create a new [`BertSource`] with the [snowflake-arctic-embed-m-long](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-long) model
    ///
    /// This model is slightly larger than [`Self::snowflake_arctic_embed_medium`] and supports longer contexts (up to 2048 tokens).
    pub fn snowflake_arctic_embed_medium_long() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m-long".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m-long".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-m-long".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ))
            .with_search_embedding_prefix(SNOWFLAKE_EMBEDDING_PREFIX.to_string())
    }

    /// Create a new [`BertSource`] with the [snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l) model
    pub fn snowflake_arctic_embed_large() -> Self {
        Self::default()
            .with_model(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-l".to_string(),
                "main".to_string(),
                "model.safetensors".to_string(),
            ))
            .with_tokenizer(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-l".to_string(),
                "main".to_string(),
                "tokenizer.json".to_string(),
            ))
            .with_config(FileSource::huggingface(
                "Snowflake/snowflake-arctic-embed-l".to_string(),
                "main".to_string(),
                "config.json".to_string(),
            ))
            .with_search_embedding_prefix(SNOWFLAKE_EMBEDDING_PREFIX.to_string())
    }
}

impl Default for BertSource {
    fn default() -> Self {
        Self::bge_small_en()
    }
}