Hugging Face's logo

Spaces:
nateraw
/
jupyterlab-inference-dev
Runtime error

App Files Community
1
jupyterlab-inference-dev
File size: 8,404 Bytes 
c7a96cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
 
mod health;
/// Text Generation Inference Webserver
mod infer;
mod queue;
pub mod server;
mod validation;

use infer::Infer;
use queue::{Entry, Queue};
use serde::{Deserialize, Serialize};
use utoipa::ToSchema;
use validation::Validation;

/// Hub type
#[derive(Clone, Debug, Deserialize)]
pub struct HubModelInfo {
    #[serde(rename(deserialize = "id"))]
    pub model_id: String,
    pub sha: Option<String>,
    pub pipeline_tag: Option<String>,
}

#[derive(Clone, Debug, Serialize, ToSchema)]
pub struct Info {
    /// Model info
    #[schema(example = "bigscience/blomm-560m")]
    pub model_id: String,
    #[schema(nullable = true, example = "e985a63cdc139290c5f700ff1929f0b5942cced2")]
    pub model_sha: Option<String>,
    #[schema(example = "torch.float16")]
    pub model_dtype: String,
    #[schema(example = "cuda")]
    pub model_device_type: String,
    #[schema(nullable = true, example = "text-generation")]
    pub model_pipeline_tag: Option<String>,
    /// Router Parameters
    #[schema(example = "128")]
    pub max_concurrent_requests: usize,
    #[schema(example = "2")]
    pub max_best_of: usize,
    #[schema(example = "4")]
    pub max_stop_sequences: usize,
    #[schema(example = "1024")]
    pub max_input_length: usize,
    #[schema(example = "2048")]
    pub max_total_tokens: usize,
    #[schema(example = "1.2")]
    pub waiting_served_ratio: f32,
    #[schema(example = "32000")]
    pub max_batch_total_tokens: u32,
    #[schema(example = "20")]
    pub max_waiting_tokens: usize,
    #[schema(example = "2")]
    pub validation_workers: usize,
    /// Router Info
    #[schema(example = "0.5.0")]
    pub version: &'static str,
    #[schema(nullable = true, example = "null")]
    pub sha: Option<&'static str>,
    #[schema(nullable = true, example = "null")]
    pub docker_label: Option<&'static str>,
}

#[derive(Clone, Debug, Deserialize, ToSchema)]
pub(crate) struct GenerateParameters {
    #[serde(default)]
    #[schema(exclusive_minimum = 0, nullable = true, default = "null", example = 1)]
    pub best_of: Option<usize>,
    #[serde(default)]
    #[schema(
        exclusive_minimum = 0.0,
        nullable = true,
        default = "null",
        example = 0.5
    )]
    pub temperature: Option<f32>,
    #[serde(default)]
    #[schema(
        exclusive_minimum = 0.0,
        nullable = true,
        default = "null",
        example = 1.03
    )]
    pub repetition_penalty: Option<f32>,
    #[serde(default)]
    #[schema(exclusive_minimum = 0, nullable = true, default = "null", example = 10)]
    pub top_k: Option<i32>,
    #[serde(default)]
    #[schema(
        exclusive_minimum = 0.0,
        maximum = 1.0,
        nullable = true,
        default = "null",
        example = 0.95
    )]
    pub top_p: Option<f32>,
    #[serde(default)]
    #[schema(
        exclusive_minimum = 0.0,
        maximum = 1.0,
        nullable = true,
        default = "null",
        example = 0.95
    )]
    pub typical_p: Option<f32>,
    #[serde(default)]
    #[schema(default = "false", example = true)]
    pub do_sample: bool,
    #[serde(default = "default_max_new_tokens")]
    #[schema(exclusive_minimum = 0, exclusive_maximum = 512, default = "20")]
    pub max_new_tokens: u32,
    #[serde(default)]
    #[schema(nullable = true, default = "null", example = false)]
    pub return_full_text: Option<bool>,
    #[serde(default)]
    #[schema(inline, max_items = 4, example = json ! (["photographer"]))]
    pub stop: Vec<String>,
    #[serde(default)]
    #[schema(nullable = true, default = "null", example = "null")]
    pub truncate: Option<usize>,
    #[serde(default)]
    #[schema(default = "false", example = true)]
    pub watermark: bool,
    #[serde(default)]
    #[schema(default = "true")]
    pub details: bool,
    #[serde(default)]
    #[schema(
        exclusive_minimum = 0,
        nullable = true,
        default = "null",
        example = "null"
    )]
    pub seed: Option<u64>,
}

fn default_max_new_tokens() -> u32 {
    20
}

fn default_parameters() -> GenerateParameters {
    GenerateParameters {
        best_of: None,
        temperature: None,
        repetition_penalty: None,
        top_k: None,
        top_p: None,
        typical_p: None,
        do_sample: false,
        max_new_tokens: default_max_new_tokens(),
        return_full_text: None,
        stop: Vec::new(),
        truncate: None,
        watermark: false,
        details: false,
        seed: None,
    }
}

#[derive(Clone, Debug, Deserialize, ToSchema)]
pub(crate) struct GenerateRequest {
    #[schema(example = "My name is Olivier and I")]
    pub inputs: String,
    #[serde(default = "default_parameters")]
    pub parameters: GenerateParameters,
}

#[derive(Clone, Debug, Deserialize, ToSchema)]
pub(crate) struct CompatGenerateRequest {
    #[schema(example = "My name is Olivier and I")]
    pub inputs: String,
    #[serde(default = "default_parameters")]
    pub parameters: GenerateParameters,
    #[serde(default)]
    #[allow(dead_code)]
    pub stream: bool,
}

impl From<CompatGenerateRequest> for GenerateRequest {
    fn from(req: CompatGenerateRequest) -> Self {
        Self {
            inputs: req.inputs,
            parameters: req.parameters,
        }
    }
}

#[derive(Debug, Serialize, ToSchema)]
pub struct PrefillToken {
    #[schema(example = 0)]
    id: u32,
    #[schema(example = "test")]
    text: String,
    #[schema(nullable = true, example = - 0.34)]
    logprob: f32,
}

#[derive(Debug, Serialize, ToSchema)]
pub struct Token {
    #[schema(example = 0)]
    id: u32,
    #[schema(example = "test")]
    text: String,
    #[schema(nullable = true, example = - 0.34)]
    logprob: f32,
    #[schema(example = "false")]
    special: bool,
}

#[derive(Serialize, ToSchema)]
#[serde(rename_all(serialize = "snake_case"))]
pub(crate) enum FinishReason {
    #[schema(rename = "length")]
    Length,
    #[serde(rename = "eos_token")]
    #[schema(rename = "eos_token")]
    EndOfSequenceToken,
    #[schema(rename = "stop_sequence")]
    StopSequence,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct BestOfSequence {
    #[schema(example = "test")]
    pub generated_text: String,
    #[schema(example = "length")]
    pub finish_reason: FinishReason,
    #[schema(example = 1)]
    pub generated_tokens: u32,
    #[schema(nullable = true, example = 42)]
    pub seed: Option<u64>,
    pub prefill: Vec<PrefillToken>,
    pub tokens: Vec<Token>,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct Details {
    #[schema(example = "length")]
    pub finish_reason: FinishReason,
    #[schema(example = 1)]
    pub generated_tokens: u32,
    #[schema(nullable = true, example = 42)]
    pub seed: Option<u64>,
    pub prefill: Vec<PrefillToken>,
    pub tokens: Vec<Token>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub best_of_sequences: Option<Vec<BestOfSequence>>,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct GenerateResponse {
    #[schema(example = "test")]
    pub generated_text: String,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub details: Option<Details>,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct StreamDetails {
    #[schema(example = "length")]
    pub finish_reason: FinishReason,
    #[schema(example = 1)]
    pub generated_tokens: u32,
    #[schema(nullable = true, example = 42)]
    pub seed: Option<u64>,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct StreamResponse {
    pub token: Token,
    #[schema(nullable = true, default = "null", example = "test")]
    pub generated_text: Option<String>,
    #[schema(nullable = true, default = "null")]
    pub details: Option<StreamDetails>,
}

#[derive(Serialize, ToSchema)]
pub(crate) struct ErrorResponse {
    pub error: String,
    pub error_type: String,
}

#[cfg(test)]
mod tests {
    use std::io::Write;
    use tokenizers::Tokenizer;

    pub(crate) async fn get_tokenizer() -> Tokenizer {
        if !std::path::Path::new("tokenizer.json").exists() {
            let content = reqwest::get("https://huggingface.co./gpt2/raw/main/tokenizer.json")
                .await
                .unwrap()
                .bytes()
                .await
                .unwrap();
            let mut file = std::fs::File::create("tokenizer.json").unwrap();
            file.write_all(&content).unwrap();
        }
        Tokenizer::from_file("tokenizer.json").unwrap()
    }
}