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
|
// Copyright Vespa.ai. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
package ai.vespa.embedding.huggingface;
import ai.vespa.embedding.PoolingStrategy;
import ai.vespa.modelintegration.evaluator.OnnxEvaluator;
import ai.vespa.modelintegration.evaluator.OnnxEvaluatorOptions;
import ai.vespa.modelintegration.evaluator.OnnxRuntime;
import com.yahoo.api.annotations.Beta;
import com.yahoo.component.AbstractComponent;
import com.yahoo.component.annotation.Inject;
import com.yahoo.embedding.huggingface.HuggingFaceEmbedderConfig;
import com.yahoo.language.huggingface.HuggingFaceTokenizer;
import com.yahoo.language.process.Embedder;
import com.yahoo.tensor.IndexedTensor;
import com.yahoo.tensor.Tensor;
import com.yahoo.tensor.TensorAddress;
import com.yahoo.tensor.TensorType;
import java.nio.file.Paths;
import java.util.BitSet;
import java.util.List;
import java.util.Map;
import java.util.logging.Logger;
import static com.yahoo.language.huggingface.ModelInfo.TruncationStrategy.LONGEST_FIRST;
@Beta
public class HuggingFaceEmbedder extends AbstractComponent implements Embedder {
private static final Logger log = Logger.getLogger(HuggingFaceEmbedder.class.getName());
private final Embedder.Runtime runtime;
private final String inputIdsName;
private final String attentionMaskName;
private final String tokenTypeIdsName;
private final String outputName;
private final boolean normalize;
private final HuggingFaceTokenizer tokenizer;
private final OnnxEvaluator evaluator;
private final PoolingStrategy poolingStrategy;
@Inject
public HuggingFaceEmbedder(OnnxRuntime onnx, Embedder.Runtime runtime, HuggingFaceEmbedderConfig config) {
this.runtime = runtime;
inputIdsName = config.transformerInputIds();
attentionMaskName = config.transformerAttentionMask();
tokenTypeIdsName = config.transformerTokenTypeIds();
outputName = config.transformerOutput();
normalize = config.normalize();
var tokenizerPath = Paths.get(config.tokenizerPath().toString());
var builder = new HuggingFaceTokenizer.Builder()
.addSpecialTokens(true)
.addDefaultModel(tokenizerPath)
.setPadding(false);
var info = HuggingFaceTokenizer.getModelInfo(tokenizerPath);
log.fine(() -> "'%s' has info '%s'".formatted(tokenizerPath, info));
if (info.maxLength() == -1 || info.truncation() != LONGEST_FIRST) {
// Force truncation to max token vector length accepted by model if tokenizer.json contains no valid truncation configuration
int maxLength = info.maxLength() > 0 && info.maxLength() <= config.transformerMaxTokens()
? info.maxLength()
: config.transformerMaxTokens();
builder.setTruncation(true).setMaxLength(maxLength);
}
this.tokenizer = builder.build();
poolingStrategy = PoolingStrategy.fromString(config.poolingStrategy().toString());
var onnxOpts = new OnnxEvaluatorOptions();
if (config.transformerGpuDevice() >= 0)
onnxOpts.setGpuDevice(config.transformerGpuDevice());
onnxOpts.setExecutionMode(config.transformerExecutionMode().toString());
onnxOpts.setThreads(config.transformerInterOpThreads(), config.transformerIntraOpThreads());
evaluator = onnx.evaluatorOf(config.transformerModel().toString(), onnxOpts);
validateModel();
}
public void validateModel() {
Map<String, TensorType> inputs = evaluator.getInputInfo();
validateName(inputs, inputIdsName, "input");
validateName(inputs, attentionMaskName, "input");
if (!tokenTypeIdsName.isEmpty()) validateName(inputs, tokenTypeIdsName, "input");
Map<String, TensorType> outputs = evaluator.getOutputInfo();
validateName(outputs, outputName, "output");
}
private void validateName(Map<String, TensorType> types, String name, String type) {
if ( ! types.containsKey(name)) {
throw new IllegalArgumentException("Model does not contain required " + type + ": '" + name + "'. " +
"Model contains: " + String.join(",", types.keySet()));
}
}
@Override
public List<Integer> embed(String s, Context context) {
var start = System.nanoTime();
var tokens = tokenizer.embed(s, context);
runtime.sampleSequenceLength(tokens.size(), context);
runtime.sampleEmbeddingLatency((System.nanoTime() - start)/1_000_000d, context);
return tokens;
}
@Override
public void deconstruct() {
evaluator.close();
tokenizer.close();
}
@SuppressWarnings("unchecked")
@Override
public Tensor embed(String s, Context context, TensorType tensorType) {
var start = System.nanoTime();
var encoding = tokenizer.encode(s, context.getLanguage());
runtime.sampleSequenceLength(encoding.ids().size(), context);
Tensor inputSequence = createTensorRepresentation(encoding.ids(), "d1");
Tensor attentionMask = createTensorRepresentation(encoding.attentionMask(), "d1");
Tensor tokenTypeIds = tokenTypeIdsName.isEmpty() ? null : createTensorRepresentation(encoding.typeIds(), "d1");
Map<String, Tensor> inputs;
if (tokenTypeIdsName.isEmpty() || tokenTypeIds.isEmpty()) {
inputs = Map.of(inputIdsName, inputSequence.expand("d0"),
attentionMaskName, attentionMask.expand("d0"));
} else {
inputs = Map.of(inputIdsName, inputSequence.expand("d0"),
attentionMaskName, attentionMask.expand("d0"),
tokenTypeIdsName, tokenTypeIds.expand("d0"));
}
IndexedTensor tokenEmbeddings = (IndexedTensor) evaluateIfNotPresent(inputs,context,s).get(outputName);
long[] resultShape = tokenEmbeddings.shape();
//shape batch, sequence, embedding dimensionality
if (resultShape.length != 3) {
throw new IllegalArgumentException("" +
"Expected 3 output dimensions for output name '" +
outputName + "': [batch, sequence, embedding], got " + resultShape.length);
}
Tensor result;
if (tensorType.valueType() == TensorType.Value.INT8) { // binary quantization
long outputDimensions = resultShape[2];
long targetDim = tensorType.dimensions().get(0).size().get();
//🪆 flexibility - packing only the first 8*targetDim float values from the model output
long floatDimensions = 8 * targetDim;
if(floatDimensions > outputDimensions) {
throw new IllegalArgumentException("Cannot pack " + outputDimensions + " into " + targetDim + " int8s");
}
//perform pooling and normalizing using float version before binary quantization
TensorType poolingType = new TensorType.Builder(TensorType.Value.FLOAT).
indexed(tensorType.indexedSubtype().dimensions().get(0).name(),
floatDimensions).build();
result = poolingStrategy.toSentenceEmbedding(poolingType, tokenEmbeddings, attentionMask);
result = normalize? normalize(result, poolingType) : result;
result = binarize((IndexedTensor) result, tensorType);
} else { // regular float embeddings up to the target dimensionality
result = poolingStrategy.toSentenceEmbedding(tensorType, tokenEmbeddings, attentionMask);
result = normalize ? normalize(result, tensorType) : result;
}
runtime.sampleEmbeddingLatency((System.nanoTime() - start)/1_000_000d, context);
return result;
}
Tensor normalize(Tensor embedding, TensorType tensorType) {
double sumOfSquares = 0.0;
Tensor.Builder builder = Tensor.Builder.of(tensorType);
for (int i = 0; i < tensorType.dimensions().get(0).size().get(); i++) {
double item = embedding.get(TensorAddress.of(i));
sumOfSquares += item * item;
}
double magnitude = Math.sqrt(sumOfSquares);
for (int i = 0; i < tensorType.dimensions().get(0).size().get(); i++) {
double value = embedding.get(TensorAddress.of(i));
builder.cell(value / magnitude, i);
}
return builder.build();
}
/**
* Evaluate the model if the result is not present in the context cache.
* @param inputs the tensor inputs
* @param context the context accompanying the request, a singleton per embedder instance and request
* @param hashKey the key to the cached value
* @return the model output
*/
@SuppressWarnings("unchecked")
protected Map<String, Tensor> evaluateIfNotPresent(Map<String, Tensor> inputs, Context context, String hashKey) {
if (context.getCachedValue(hashKey) == null) {
Map<String, Tensor> outputs = evaluator.evaluate(inputs);
context.putCachedValue(hashKey, outputs);
return outputs;
} else {
return (Map<String, Tensor>) context.getCachedValue(hashKey);
}
}
/**
* Binary quantization of the embedding into a tensor of type int8 with the specified dimensions.
* @param embedding
* @param tensorType
* @return
*/
static public Tensor binarize(IndexedTensor embedding, TensorType tensorType) {
Tensor.Builder builder = Tensor.Builder.of(tensorType);
BitSet bitSet = new BitSet(8);
int index = 0;
for (int d = 0; d < embedding.sizeAsInt(); d++) {
var value = embedding.get(d);
int bitIndex = 7 - (d % 8);
if (value > 0.0) {
bitSet.set(bitIndex);
} else {
bitSet.clear(bitIndex);
}
if ((d + 1) % 8 == 0) {
byte[] bytes = bitSet.toByteArray();
byte packed = (bytes.length == 0) ? 0 : bytes[0];
builder.cell(TensorAddress.of(index), packed);
index++;
bitSet = new BitSet(8);
}
}
return builder.build();
}
private IndexedTensor createTensorRepresentation(List<Long> input, String dimension) {
int size = input.size();
TensorType type = new TensorType.Builder(TensorType.Value.FLOAT).indexed(dimension, size).build();
IndexedTensor.Builder builder = IndexedTensor.Builder.of(type);
for (int i = 0; i < size; ++i) {
builder.cell(input.get(i), i);
}
return builder.build();
}
}
|
