keras-hub-nightly 0.15.0.dev20240823171555__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- keras_hub/__init__.py +52 -0
- keras_hub/api/__init__.py +27 -0
- keras_hub/api/layers/__init__.py +47 -0
- keras_hub/api/metrics/__init__.py +24 -0
- keras_hub/api/models/__init__.py +249 -0
- keras_hub/api/samplers/__init__.py +29 -0
- keras_hub/api/tokenizers/__init__.py +35 -0
- keras_hub/src/__init__.py +13 -0
- keras_hub/src/api_export.py +53 -0
- keras_hub/src/layers/__init__.py +13 -0
- keras_hub/src/layers/modeling/__init__.py +13 -0
- keras_hub/src/layers/modeling/alibi_bias.py +143 -0
- keras_hub/src/layers/modeling/cached_multi_head_attention.py +137 -0
- keras_hub/src/layers/modeling/f_net_encoder.py +200 -0
- keras_hub/src/layers/modeling/masked_lm_head.py +239 -0
- keras_hub/src/layers/modeling/position_embedding.py +123 -0
- keras_hub/src/layers/modeling/reversible_embedding.py +311 -0
- keras_hub/src/layers/modeling/rotary_embedding.py +169 -0
- keras_hub/src/layers/modeling/sine_position_encoding.py +108 -0
- keras_hub/src/layers/modeling/token_and_position_embedding.py +150 -0
- keras_hub/src/layers/modeling/transformer_decoder.py +496 -0
- keras_hub/src/layers/modeling/transformer_encoder.py +262 -0
- keras_hub/src/layers/modeling/transformer_layer_utils.py +106 -0
- keras_hub/src/layers/preprocessing/__init__.py +13 -0
- keras_hub/src/layers/preprocessing/masked_lm_mask_generator.py +220 -0
- keras_hub/src/layers/preprocessing/multi_segment_packer.py +319 -0
- keras_hub/src/layers/preprocessing/preprocessing_layer.py +62 -0
- keras_hub/src/layers/preprocessing/random_deletion.py +271 -0
- keras_hub/src/layers/preprocessing/random_swap.py +267 -0
- keras_hub/src/layers/preprocessing/start_end_packer.py +219 -0
- keras_hub/src/metrics/__init__.py +13 -0
- keras_hub/src/metrics/bleu.py +394 -0
- keras_hub/src/metrics/edit_distance.py +197 -0
- keras_hub/src/metrics/perplexity.py +181 -0
- keras_hub/src/metrics/rouge_base.py +204 -0
- keras_hub/src/metrics/rouge_l.py +97 -0
- keras_hub/src/metrics/rouge_n.py +125 -0
- keras_hub/src/models/__init__.py +13 -0
- keras_hub/src/models/albert/__init__.py +20 -0
- keras_hub/src/models/albert/albert_backbone.py +267 -0
- keras_hub/src/models/albert/albert_classifier.py +202 -0
- keras_hub/src/models/albert/albert_masked_lm.py +129 -0
- keras_hub/src/models/albert/albert_masked_lm_preprocessor.py +194 -0
- keras_hub/src/models/albert/albert_preprocessor.py +206 -0
- keras_hub/src/models/albert/albert_presets.py +70 -0
- keras_hub/src/models/albert/albert_tokenizer.py +119 -0
- keras_hub/src/models/backbone.py +311 -0
- keras_hub/src/models/bart/__init__.py +20 -0
- keras_hub/src/models/bart/bart_backbone.py +261 -0
- keras_hub/src/models/bart/bart_preprocessor.py +276 -0
- keras_hub/src/models/bart/bart_presets.py +74 -0
- keras_hub/src/models/bart/bart_seq_2_seq_lm.py +490 -0
- keras_hub/src/models/bart/bart_seq_2_seq_lm_preprocessor.py +262 -0
- keras_hub/src/models/bart/bart_tokenizer.py +124 -0
- keras_hub/src/models/bert/__init__.py +23 -0
- keras_hub/src/models/bert/bert_backbone.py +227 -0
- keras_hub/src/models/bert/bert_classifier.py +183 -0
- keras_hub/src/models/bert/bert_masked_lm.py +131 -0
- keras_hub/src/models/bert/bert_masked_lm_preprocessor.py +198 -0
- keras_hub/src/models/bert/bert_preprocessor.py +184 -0
- keras_hub/src/models/bert/bert_presets.py +147 -0
- keras_hub/src/models/bert/bert_tokenizer.py +112 -0
- keras_hub/src/models/bloom/__init__.py +20 -0
- keras_hub/src/models/bloom/bloom_attention.py +186 -0
- keras_hub/src/models/bloom/bloom_backbone.py +173 -0
- keras_hub/src/models/bloom/bloom_causal_lm.py +298 -0
- keras_hub/src/models/bloom/bloom_causal_lm_preprocessor.py +176 -0
- keras_hub/src/models/bloom/bloom_decoder.py +206 -0
- keras_hub/src/models/bloom/bloom_preprocessor.py +185 -0
- keras_hub/src/models/bloom/bloom_presets.py +121 -0
- keras_hub/src/models/bloom/bloom_tokenizer.py +116 -0
- keras_hub/src/models/causal_lm.py +383 -0
- keras_hub/src/models/classifier.py +109 -0
- keras_hub/src/models/csp_darknet/__init__.py +13 -0
- keras_hub/src/models/csp_darknet/csp_darknet_backbone.py +410 -0
- keras_hub/src/models/csp_darknet/csp_darknet_image_classifier.py +133 -0
- keras_hub/src/models/deberta_v3/__init__.py +24 -0
- keras_hub/src/models/deberta_v3/deberta_v3_backbone.py +210 -0
- keras_hub/src/models/deberta_v3/deberta_v3_classifier.py +228 -0
- keras_hub/src/models/deberta_v3/deberta_v3_masked_lm.py +135 -0
- keras_hub/src/models/deberta_v3/deberta_v3_masked_lm_preprocessor.py +191 -0
- keras_hub/src/models/deberta_v3/deberta_v3_preprocessor.py +206 -0
- keras_hub/src/models/deberta_v3/deberta_v3_presets.py +82 -0
- keras_hub/src/models/deberta_v3/deberta_v3_tokenizer.py +155 -0
- keras_hub/src/models/deberta_v3/disentangled_attention_encoder.py +227 -0
- keras_hub/src/models/deberta_v3/disentangled_self_attention.py +412 -0
- keras_hub/src/models/deberta_v3/relative_embedding.py +94 -0
- keras_hub/src/models/densenet/__init__.py +13 -0
- keras_hub/src/models/densenet/densenet_backbone.py +210 -0
- keras_hub/src/models/densenet/densenet_image_classifier.py +131 -0
- keras_hub/src/models/distil_bert/__init__.py +26 -0
- keras_hub/src/models/distil_bert/distil_bert_backbone.py +187 -0
- keras_hub/src/models/distil_bert/distil_bert_classifier.py +208 -0
- keras_hub/src/models/distil_bert/distil_bert_masked_lm.py +137 -0
- keras_hub/src/models/distil_bert/distil_bert_masked_lm_preprocessor.py +194 -0
- keras_hub/src/models/distil_bert/distil_bert_preprocessor.py +175 -0
- keras_hub/src/models/distil_bert/distil_bert_presets.py +57 -0
- keras_hub/src/models/distil_bert/distil_bert_tokenizer.py +114 -0
- keras_hub/src/models/electra/__init__.py +20 -0
- keras_hub/src/models/electra/electra_backbone.py +247 -0
- keras_hub/src/models/electra/electra_preprocessor.py +154 -0
- keras_hub/src/models/electra/electra_presets.py +95 -0
- keras_hub/src/models/electra/electra_tokenizer.py +104 -0
- keras_hub/src/models/f_net/__init__.py +20 -0
- keras_hub/src/models/f_net/f_net_backbone.py +236 -0
- keras_hub/src/models/f_net/f_net_classifier.py +154 -0
- keras_hub/src/models/f_net/f_net_masked_lm.py +132 -0
- keras_hub/src/models/f_net/f_net_masked_lm_preprocessor.py +196 -0
- keras_hub/src/models/f_net/f_net_preprocessor.py +177 -0
- keras_hub/src/models/f_net/f_net_presets.py +43 -0
- keras_hub/src/models/f_net/f_net_tokenizer.py +95 -0
- keras_hub/src/models/falcon/__init__.py +20 -0
- keras_hub/src/models/falcon/falcon_attention.py +156 -0
- keras_hub/src/models/falcon/falcon_backbone.py +164 -0
- keras_hub/src/models/falcon/falcon_causal_lm.py +291 -0
- keras_hub/src/models/falcon/falcon_causal_lm_preprocessor.py +173 -0
- keras_hub/src/models/falcon/falcon_preprocessor.py +187 -0
- keras_hub/src/models/falcon/falcon_presets.py +30 -0
- keras_hub/src/models/falcon/falcon_tokenizer.py +110 -0
- keras_hub/src/models/falcon/falcon_transformer_decoder.py +255 -0
- keras_hub/src/models/feature_pyramid_backbone.py +73 -0
- keras_hub/src/models/gemma/__init__.py +20 -0
- keras_hub/src/models/gemma/gemma_attention.py +250 -0
- keras_hub/src/models/gemma/gemma_backbone.py +316 -0
- keras_hub/src/models/gemma/gemma_causal_lm.py +448 -0
- keras_hub/src/models/gemma/gemma_causal_lm_preprocessor.py +167 -0
- keras_hub/src/models/gemma/gemma_decoder_block.py +241 -0
- keras_hub/src/models/gemma/gemma_preprocessor.py +191 -0
- keras_hub/src/models/gemma/gemma_presets.py +248 -0
- keras_hub/src/models/gemma/gemma_tokenizer.py +103 -0
- keras_hub/src/models/gemma/rms_normalization.py +40 -0
- keras_hub/src/models/gpt2/__init__.py +20 -0
- keras_hub/src/models/gpt2/gpt2_backbone.py +199 -0
- keras_hub/src/models/gpt2/gpt2_causal_lm.py +437 -0
- keras_hub/src/models/gpt2/gpt2_causal_lm_preprocessor.py +173 -0
- keras_hub/src/models/gpt2/gpt2_preprocessor.py +187 -0
- keras_hub/src/models/gpt2/gpt2_presets.py +82 -0
- keras_hub/src/models/gpt2/gpt2_tokenizer.py +110 -0
- keras_hub/src/models/gpt_neo_x/__init__.py +13 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_attention.py +251 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_backbone.py +175 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_causal_lm.py +201 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_causal_lm_preprocessor.py +141 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_decoder.py +258 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_preprocessor.py +145 -0
- keras_hub/src/models/gpt_neo_x/gpt_neo_x_tokenizer.py +88 -0
- keras_hub/src/models/image_classifier.py +90 -0
- keras_hub/src/models/llama/__init__.py +20 -0
- keras_hub/src/models/llama/llama_attention.py +225 -0
- keras_hub/src/models/llama/llama_backbone.py +188 -0
- keras_hub/src/models/llama/llama_causal_lm.py +327 -0
- keras_hub/src/models/llama/llama_causal_lm_preprocessor.py +170 -0
- keras_hub/src/models/llama/llama_decoder.py +246 -0
- keras_hub/src/models/llama/llama_layernorm.py +48 -0
- keras_hub/src/models/llama/llama_preprocessor.py +189 -0
- keras_hub/src/models/llama/llama_presets.py +80 -0
- keras_hub/src/models/llama/llama_tokenizer.py +84 -0
- keras_hub/src/models/llama3/__init__.py +20 -0
- keras_hub/src/models/llama3/llama3_backbone.py +84 -0
- keras_hub/src/models/llama3/llama3_causal_lm.py +46 -0
- keras_hub/src/models/llama3/llama3_causal_lm_preprocessor.py +173 -0
- keras_hub/src/models/llama3/llama3_preprocessor.py +21 -0
- keras_hub/src/models/llama3/llama3_presets.py +69 -0
- keras_hub/src/models/llama3/llama3_tokenizer.py +63 -0
- keras_hub/src/models/masked_lm.py +101 -0
- keras_hub/src/models/mistral/__init__.py +20 -0
- keras_hub/src/models/mistral/mistral_attention.py +238 -0
- keras_hub/src/models/mistral/mistral_backbone.py +203 -0
- keras_hub/src/models/mistral/mistral_causal_lm.py +328 -0
- keras_hub/src/models/mistral/mistral_causal_lm_preprocessor.py +175 -0
- keras_hub/src/models/mistral/mistral_layer_norm.py +48 -0
- keras_hub/src/models/mistral/mistral_preprocessor.py +190 -0
- keras_hub/src/models/mistral/mistral_presets.py +48 -0
- keras_hub/src/models/mistral/mistral_tokenizer.py +82 -0
- keras_hub/src/models/mistral/mistral_transformer_decoder.py +265 -0
- keras_hub/src/models/mix_transformer/__init__.py +13 -0
- keras_hub/src/models/mix_transformer/mix_transformer_backbone.py +181 -0
- keras_hub/src/models/mix_transformer/mix_transformer_classifier.py +133 -0
- keras_hub/src/models/mix_transformer/mix_transformer_layers.py +300 -0
- keras_hub/src/models/opt/__init__.py +20 -0
- keras_hub/src/models/opt/opt_backbone.py +173 -0
- keras_hub/src/models/opt/opt_causal_lm.py +301 -0
- keras_hub/src/models/opt/opt_causal_lm_preprocessor.py +177 -0
- keras_hub/src/models/opt/opt_preprocessor.py +188 -0
- keras_hub/src/models/opt/opt_presets.py +72 -0
- keras_hub/src/models/opt/opt_tokenizer.py +116 -0
- keras_hub/src/models/pali_gemma/__init__.py +23 -0
- keras_hub/src/models/pali_gemma/pali_gemma_backbone.py +277 -0
- keras_hub/src/models/pali_gemma/pali_gemma_causal_lm.py +313 -0
- keras_hub/src/models/pali_gemma/pali_gemma_causal_lm_preprocessor.py +147 -0
- keras_hub/src/models/pali_gemma/pali_gemma_decoder_block.py +160 -0
- keras_hub/src/models/pali_gemma/pali_gemma_presets.py +78 -0
- keras_hub/src/models/pali_gemma/pali_gemma_tokenizer.py +79 -0
- keras_hub/src/models/pali_gemma/pali_gemma_vit.py +566 -0
- keras_hub/src/models/phi3/__init__.py +20 -0
- keras_hub/src/models/phi3/phi3_attention.py +260 -0
- keras_hub/src/models/phi3/phi3_backbone.py +224 -0
- keras_hub/src/models/phi3/phi3_causal_lm.py +218 -0
- keras_hub/src/models/phi3/phi3_causal_lm_preprocessor.py +173 -0
- keras_hub/src/models/phi3/phi3_decoder.py +260 -0
- keras_hub/src/models/phi3/phi3_layernorm.py +48 -0
- keras_hub/src/models/phi3/phi3_preprocessor.py +190 -0
- keras_hub/src/models/phi3/phi3_presets.py +50 -0
- keras_hub/src/models/phi3/phi3_rotary_embedding.py +137 -0
- keras_hub/src/models/phi3/phi3_tokenizer.py +94 -0
- keras_hub/src/models/preprocessor.py +207 -0
- keras_hub/src/models/resnet/__init__.py +13 -0
- keras_hub/src/models/resnet/resnet_backbone.py +612 -0
- keras_hub/src/models/resnet/resnet_image_classifier.py +136 -0
- keras_hub/src/models/roberta/__init__.py +20 -0
- keras_hub/src/models/roberta/roberta_backbone.py +184 -0
- keras_hub/src/models/roberta/roberta_classifier.py +209 -0
- keras_hub/src/models/roberta/roberta_masked_lm.py +136 -0
- keras_hub/src/models/roberta/roberta_masked_lm_preprocessor.py +198 -0
- keras_hub/src/models/roberta/roberta_preprocessor.py +192 -0
- keras_hub/src/models/roberta/roberta_presets.py +43 -0
- keras_hub/src/models/roberta/roberta_tokenizer.py +132 -0
- keras_hub/src/models/seq_2_seq_lm.py +54 -0
- keras_hub/src/models/t5/__init__.py +20 -0
- keras_hub/src/models/t5/t5_backbone.py +261 -0
- keras_hub/src/models/t5/t5_layer_norm.py +35 -0
- keras_hub/src/models/t5/t5_multi_head_attention.py +324 -0
- keras_hub/src/models/t5/t5_presets.py +95 -0
- keras_hub/src/models/t5/t5_tokenizer.py +100 -0
- keras_hub/src/models/t5/t5_transformer_layer.py +178 -0
- keras_hub/src/models/task.py +419 -0
- keras_hub/src/models/vgg/__init__.py +13 -0
- keras_hub/src/models/vgg/vgg_backbone.py +158 -0
- keras_hub/src/models/vgg/vgg_image_classifier.py +124 -0
- keras_hub/src/models/vit_det/__init__.py +13 -0
- keras_hub/src/models/vit_det/vit_det_backbone.py +204 -0
- keras_hub/src/models/vit_det/vit_layers.py +565 -0
- keras_hub/src/models/whisper/__init__.py +20 -0
- keras_hub/src/models/whisper/whisper_audio_feature_extractor.py +260 -0
- keras_hub/src/models/whisper/whisper_backbone.py +305 -0
- keras_hub/src/models/whisper/whisper_cached_multi_head_attention.py +153 -0
- keras_hub/src/models/whisper/whisper_decoder.py +141 -0
- keras_hub/src/models/whisper/whisper_encoder.py +106 -0
- keras_hub/src/models/whisper/whisper_preprocessor.py +326 -0
- keras_hub/src/models/whisper/whisper_presets.py +148 -0
- keras_hub/src/models/whisper/whisper_tokenizer.py +163 -0
- keras_hub/src/models/xlm_roberta/__init__.py +26 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_backbone.py +81 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_classifier.py +225 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_masked_lm.py +141 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_masked_lm_preprocessor.py +195 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_preprocessor.py +205 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_presets.py +43 -0
- keras_hub/src/models/xlm_roberta/xlm_roberta_tokenizer.py +191 -0
- keras_hub/src/models/xlnet/__init__.py +13 -0
- keras_hub/src/models/xlnet/relative_attention.py +459 -0
- keras_hub/src/models/xlnet/xlnet_backbone.py +222 -0
- keras_hub/src/models/xlnet/xlnet_content_and_query_embedding.py +133 -0
- keras_hub/src/models/xlnet/xlnet_encoder.py +378 -0
- keras_hub/src/samplers/__init__.py +13 -0
- keras_hub/src/samplers/beam_sampler.py +207 -0
- keras_hub/src/samplers/contrastive_sampler.py +231 -0
- keras_hub/src/samplers/greedy_sampler.py +50 -0
- keras_hub/src/samplers/random_sampler.py +77 -0
- keras_hub/src/samplers/sampler.py +237 -0
- keras_hub/src/samplers/serialization.py +97 -0
- keras_hub/src/samplers/top_k_sampler.py +92 -0
- keras_hub/src/samplers/top_p_sampler.py +113 -0
- keras_hub/src/tests/__init__.py +13 -0
- keras_hub/src/tests/test_case.py +608 -0
- keras_hub/src/tokenizers/__init__.py +13 -0
- keras_hub/src/tokenizers/byte_pair_tokenizer.py +638 -0
- keras_hub/src/tokenizers/byte_tokenizer.py +299 -0
- keras_hub/src/tokenizers/sentence_piece_tokenizer.py +267 -0
- keras_hub/src/tokenizers/sentence_piece_tokenizer_trainer.py +150 -0
- keras_hub/src/tokenizers/tokenizer.py +235 -0
- keras_hub/src/tokenizers/unicode_codepoint_tokenizer.py +355 -0
- keras_hub/src/tokenizers/word_piece_tokenizer.py +544 -0
- keras_hub/src/tokenizers/word_piece_tokenizer_trainer.py +176 -0
- keras_hub/src/utils/__init__.py +13 -0
- keras_hub/src/utils/keras_utils.py +130 -0
- keras_hub/src/utils/pipeline_model.py +293 -0
- keras_hub/src/utils/preset_utils.py +621 -0
- keras_hub/src/utils/python_utils.py +21 -0
- keras_hub/src/utils/tensor_utils.py +206 -0
- keras_hub/src/utils/timm/__init__.py +13 -0
- keras_hub/src/utils/timm/convert.py +37 -0
- keras_hub/src/utils/timm/convert_resnet.py +171 -0
- keras_hub/src/utils/transformers/__init__.py +13 -0
- keras_hub/src/utils/transformers/convert.py +101 -0
- keras_hub/src/utils/transformers/convert_bert.py +173 -0
- keras_hub/src/utils/transformers/convert_distilbert.py +184 -0
- keras_hub/src/utils/transformers/convert_gemma.py +187 -0
- keras_hub/src/utils/transformers/convert_gpt2.py +186 -0
- keras_hub/src/utils/transformers/convert_llama3.py +136 -0
- keras_hub/src/utils/transformers/convert_pali_gemma.py +303 -0
- keras_hub/src/utils/transformers/safetensor_utils.py +97 -0
- keras_hub/src/version_utils.py +23 -0
- keras_hub_nightly-0.15.0.dev20240823171555.dist-info/METADATA +34 -0
- keras_hub_nightly-0.15.0.dev20240823171555.dist-info/RECORD +297 -0
- keras_hub_nightly-0.15.0.dev20240823171555.dist-info/WHEEL +5 -0
- keras_hub_nightly-0.15.0.dev20240823171555.dist-info/top_level.txt +1 -0
@@ -0,0 +1,198 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
|
15
|
+
import keras
|
16
|
+
from absl import logging
|
17
|
+
|
18
|
+
from keras_hub.src.api_export import keras_hub_export
|
19
|
+
from keras_hub.src.layers.preprocessing.masked_lm_mask_generator import (
|
20
|
+
MaskedLMMaskGenerator,
|
21
|
+
)
|
22
|
+
from keras_hub.src.models.roberta.roberta_preprocessor import (
|
23
|
+
RobertaPreprocessor,
|
24
|
+
)
|
25
|
+
|
26
|
+
|
27
|
+
@keras_hub_export("keras_hub.models.RobertaMaskedLMPreprocessor")
|
28
|
+
class RobertaMaskedLMPreprocessor(RobertaPreprocessor):
|
29
|
+
"""RoBERTa preprocessing for the masked language modeling task.
|
30
|
+
|
31
|
+
This preprocessing layer will prepare inputs for a masked language modeling
|
32
|
+
task. It is primarily intended for use with the
|
33
|
+
`keras_hub.models.RobertaMaskedLM` task model. Preprocessing will occur in
|
34
|
+
multiple steps.
|
35
|
+
|
36
|
+
1. Tokenize any number of input segments using the `tokenizer`.
|
37
|
+
2. Pack the inputs together with the appropriate `"<s>"`, `"</s>"` and
|
38
|
+
`"<pad>"` tokens, i.e., adding a single `"<s>"` at the start of the
|
39
|
+
entire sequence, `"</s></s>"` between each segment,
|
40
|
+
and a `"</s>"` at the end of the entire sequence.
|
41
|
+
3. Randomly select non-special tokens to mask, controlled by
|
42
|
+
`mask_selection_rate`.
|
43
|
+
4. Construct a `(x, y, sample_weight)` tuple suitable for training with a
|
44
|
+
`keras_hub.models.RobertaMaskedLM` task model.
|
45
|
+
|
46
|
+
Args:
|
47
|
+
tokenizer: A `keras_hub.models.RobertaTokenizer` instance.
|
48
|
+
sequence_length: int. The length of the packed inputs.
|
49
|
+
truncate: string. The algorithm to truncate a list of batched segments
|
50
|
+
to fit within `sequence_length`. The value can be either
|
51
|
+
`round_robin` or `waterfall`:
|
52
|
+
- `"round_robin"`: Available space is assigned one token at a
|
53
|
+
time in a round-robin fashion to the inputs that still need
|
54
|
+
some, until the limit is reached.
|
55
|
+
- `"waterfall"`: The allocation of the budget is done using a
|
56
|
+
"waterfall" algorithm that allocates quota in a
|
57
|
+
left-to-right manner and fills up the buckets until we run
|
58
|
+
out of budget. It supports an arbitrary number of segments.
|
59
|
+
mask_selection_rate: float. The probability an input token will be
|
60
|
+
dynamically masked.
|
61
|
+
mask_selection_length: int. The maximum number of masked tokens
|
62
|
+
in a given sample.
|
63
|
+
mask_token_rate: float. The probability the a selected token will be
|
64
|
+
replaced with the mask token.
|
65
|
+
random_token_rate: float. The probability the a selected token will be
|
66
|
+
replaced with a random token from the vocabulary. A selected token
|
67
|
+
will be left as is with probability
|
68
|
+
`1 - mask_token_rate - random_token_rate`.
|
69
|
+
|
70
|
+
Call arguments:
|
71
|
+
x: A tensor of single string sequences, or a tuple of multiple
|
72
|
+
tensor sequences to be packed together. Inputs may be batched or
|
73
|
+
unbatched. For single sequences, raw python inputs will be converted
|
74
|
+
to tensors. For multiple sequences, pass tensors directly.
|
75
|
+
y: Label data. Should always be `None` as the layer generates labels.
|
76
|
+
sample_weight: Label weights. Should always be `None` as the layer
|
77
|
+
generates label weights.
|
78
|
+
|
79
|
+
Examples:
|
80
|
+
|
81
|
+
Directly calling the layer on data.
|
82
|
+
```python
|
83
|
+
# Load the preprocessor from a preset.
|
84
|
+
preprocessor = keras_hub.models.RobertaMaskedLMPreprocessor.from_preset(
|
85
|
+
"roberta_base_en"
|
86
|
+
)
|
87
|
+
|
88
|
+
# Tokenize and mask a single sentence.
|
89
|
+
preprocessor("The quick brown fox jumped.")
|
90
|
+
|
91
|
+
# Tokenize and mask a batch of single sentences.
|
92
|
+
preprocessor(["The quick brown fox jumped.", "Call me Ishmael."])
|
93
|
+
|
94
|
+
# Tokenize and mask sentence pairs.
|
95
|
+
# In this case, always convert input to tensors before calling the layer.
|
96
|
+
first = tf.constant(["The quick brown fox jumped.", "Call me Ishmael."])
|
97
|
+
second = tf.constant(["The fox tripped.", "Oh look, a whale."])
|
98
|
+
preprocessor((first, second))
|
99
|
+
```
|
100
|
+
|
101
|
+
Mapping with `tf.data.Dataset`.
|
102
|
+
```python
|
103
|
+
preprocessor = keras_hub.models.RobertaMaskedLMPreprocessor.from_preset(
|
104
|
+
"roberta_base_en"
|
105
|
+
)
|
106
|
+
|
107
|
+
first = tf.constant(["The quick brown fox jumped.", "Call me Ishmael."])
|
108
|
+
second = tf.constant(["The fox tripped.", "Oh look, a whale."])
|
109
|
+
|
110
|
+
# Map single sentences.
|
111
|
+
ds = tf.data.Dataset.from_tensor_slices(first)
|
112
|
+
ds = ds.map(preprocessor, num_parallel_calls=tf.data.AUTOTUNE)
|
113
|
+
|
114
|
+
# Map sentence pairs.
|
115
|
+
ds = tf.data.Dataset.from_tensor_slices((first, second))
|
116
|
+
# Watch out for tf.data's default unpacking of tuples here!
|
117
|
+
# Best to invoke the `preprocessor` directly in this case.
|
118
|
+
ds = ds.map(
|
119
|
+
lambda first, second: preprocessor(x=(first, second)),
|
120
|
+
num_parallel_calls=tf.data.AUTOTUNE,
|
121
|
+
)
|
122
|
+
```
|
123
|
+
"""
|
124
|
+
|
125
|
+
def __init__(
|
126
|
+
self,
|
127
|
+
tokenizer,
|
128
|
+
sequence_length=512,
|
129
|
+
truncate="round_robin",
|
130
|
+
mask_selection_rate=0.15,
|
131
|
+
mask_selection_length=96,
|
132
|
+
mask_token_rate=0.8,
|
133
|
+
random_token_rate=0.1,
|
134
|
+
**kwargs,
|
135
|
+
):
|
136
|
+
super().__init__(
|
137
|
+
tokenizer,
|
138
|
+
sequence_length=sequence_length,
|
139
|
+
truncate=truncate,
|
140
|
+
**kwargs,
|
141
|
+
)
|
142
|
+
self.mask_selection_rate = mask_selection_rate
|
143
|
+
self.mask_selection_length = mask_selection_length
|
144
|
+
self.mask_token_rate = mask_token_rate
|
145
|
+
self.random_token_rate = random_token_rate
|
146
|
+
self.masker = None
|
147
|
+
|
148
|
+
def build(self, input_shape):
|
149
|
+
super().build(input_shape)
|
150
|
+
# Defer packer creation to `build()` so that we can be sure tokenizer
|
151
|
+
# assets have loaded when restoring a saved model.
|
152
|
+
self.masker = MaskedLMMaskGenerator(
|
153
|
+
mask_selection_rate=self.mask_selection_rate,
|
154
|
+
mask_selection_length=self.mask_selection_length,
|
155
|
+
mask_token_rate=self.mask_token_rate,
|
156
|
+
random_token_rate=self.random_token_rate,
|
157
|
+
vocabulary_size=self.tokenizer.vocabulary_size(),
|
158
|
+
mask_token_id=self.tokenizer.mask_token_id,
|
159
|
+
unselectable_token_ids=[
|
160
|
+
self.tokenizer.start_token_id,
|
161
|
+
self.tokenizer.end_token_id,
|
162
|
+
self.tokenizer.pad_token_id,
|
163
|
+
],
|
164
|
+
)
|
165
|
+
self.built = True
|
166
|
+
|
167
|
+
def call(self, x, y=None, sample_weight=None):
|
168
|
+
if y is not None or sample_weight is not None:
|
169
|
+
logging.warning(
|
170
|
+
f"{self.__class__.__name__} generates `y` and `sample_weight` "
|
171
|
+
"based on your input data, but your data already contains `y` "
|
172
|
+
"or `sample_weight`. Your `y` and `sample_weight` will be "
|
173
|
+
"ignored."
|
174
|
+
)
|
175
|
+
|
176
|
+
x = super().call(x)
|
177
|
+
token_ids, padding_mask = x["token_ids"], x["padding_mask"]
|
178
|
+
masker_outputs = self.masker(token_ids)
|
179
|
+
x = {
|
180
|
+
"token_ids": masker_outputs["token_ids"],
|
181
|
+
"padding_mask": padding_mask,
|
182
|
+
"mask_positions": masker_outputs["mask_positions"],
|
183
|
+
}
|
184
|
+
y = masker_outputs["mask_ids"]
|
185
|
+
sample_weight = masker_outputs["mask_weights"]
|
186
|
+
return keras.utils.pack_x_y_sample_weight(x, y, sample_weight)
|
187
|
+
|
188
|
+
def get_config(self):
|
189
|
+
config = super().get_config()
|
190
|
+
config.update(
|
191
|
+
{
|
192
|
+
"mask_selection_rate": self.mask_selection_rate,
|
193
|
+
"mask_selection_length": self.mask_selection_length,
|
194
|
+
"mask_token_rate": self.mask_token_rate,
|
195
|
+
"random_token_rate": self.random_token_rate,
|
196
|
+
}
|
197
|
+
)
|
198
|
+
return config
|
@@ -0,0 +1,192 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
|
15
|
+
|
16
|
+
import keras
|
17
|
+
|
18
|
+
from keras_hub.src.api_export import keras_hub_export
|
19
|
+
from keras_hub.src.layers.preprocessing.multi_segment_packer import (
|
20
|
+
MultiSegmentPacker,
|
21
|
+
)
|
22
|
+
from keras_hub.src.models.preprocessor import Preprocessor
|
23
|
+
from keras_hub.src.models.roberta.roberta_tokenizer import RobertaTokenizer
|
24
|
+
from keras_hub.src.utils.keras_utils import (
|
25
|
+
convert_inputs_to_list_of_tensor_segments,
|
26
|
+
)
|
27
|
+
|
28
|
+
|
29
|
+
@keras_hub_export("keras_hub.models.RobertaPreprocessor")
|
30
|
+
class RobertaPreprocessor(Preprocessor):
|
31
|
+
"""A RoBERTa preprocessing layer which tokenizes and packs inputs.
|
32
|
+
|
33
|
+
This preprocessing layer will do three things:
|
34
|
+
|
35
|
+
1. Tokenize any number of input segments using the `tokenizer`.
|
36
|
+
2. Pack the inputs together with the appropriate `"<s>"`, `"</s>"` and
|
37
|
+
`"<pad>"` tokens, i.e., adding a single `"<s>"` at the start of the
|
38
|
+
entire sequence, `"</s></s>"` at the end of each segment, save the last
|
39
|
+
and a `"</s>"` at the end of the entire sequence.
|
40
|
+
3. Construct a dictionary with keys `"token_ids"`, `"padding_mask"` that
|
41
|
+
can be passed directly to a RoBERTa model.
|
42
|
+
|
43
|
+
This layer can be used directly with `tf.data.Dataset.map` to preprocess
|
44
|
+
string data in the `(x, y, sample_weight)` format used by
|
45
|
+
`keras.Model.fit`.
|
46
|
+
|
47
|
+
Args:
|
48
|
+
tokenizer: A `keras_hub.models.RobertaTokenizer` instance.
|
49
|
+
sequence_length: The length of the packed inputs.
|
50
|
+
truncate: string. The algorithm to truncate a list of batched segments
|
51
|
+
to fit within `sequence_length`. The value can be either
|
52
|
+
`round_robin` or `waterfall`:
|
53
|
+
- `"round_robin"`: Available space is assigned one token at a
|
54
|
+
time in a round-robin fashion to the inputs that still need
|
55
|
+
some, until the limit is reached.
|
56
|
+
- `"waterfall"`: The allocation of the budget is done using a
|
57
|
+
"waterfall" algorithm that allocates quota in a
|
58
|
+
left-to-right manner and fills up the buckets until we run
|
59
|
+
out of budget. It supports an arbitrary number of segments.
|
60
|
+
|
61
|
+
Call arguments:
|
62
|
+
x: A tensor of single string sequences, or a tuple of multiple
|
63
|
+
tensor sequences to be packed together. Inputs may be batched or
|
64
|
+
unbatched. For single sequences, raw python inputs will be converted
|
65
|
+
to tensors. For multiple sequences, pass tensors directly.
|
66
|
+
y: Any label data. Will be passed through unaltered.
|
67
|
+
sample_weight: Any label weight data. Will be passed through unaltered.
|
68
|
+
|
69
|
+
|
70
|
+
Examples:
|
71
|
+
|
72
|
+
Directly calling the layer on data.
|
73
|
+
```python
|
74
|
+
preprocessor = keras_hub.models.RobertaPreprocessor.from_preset(
|
75
|
+
"roberta_base_en"
|
76
|
+
)
|
77
|
+
|
78
|
+
# Tokenize and pack a single sentence.
|
79
|
+
preprocessor("The quick brown fox jumped.")
|
80
|
+
|
81
|
+
# Tokenize a batch of single sentences.
|
82
|
+
preprocessor(["The quick brown fox jumped.", "Call me Ishmael."])
|
83
|
+
|
84
|
+
# Preprocess a batch of sentence pairs.
|
85
|
+
# When handling multiple sequences, always convert to tensors first!
|
86
|
+
first = tf.constant(["The quick brown fox jumped.", "Call me Ishmael."])
|
87
|
+
second = tf.constant(["The fox tripped.", "Oh look, a whale."])
|
88
|
+
preprocessor((first, second))
|
89
|
+
|
90
|
+
# Custom vocabulary.
|
91
|
+
vocab = {"<s>": 0, "<pad>": 1, "</s>": 2, "<mask>": 3}
|
92
|
+
vocab = {**vocab, "a": 4, "Ġquick": 5, "Ġfox": 6}
|
93
|
+
merges = ["Ġ q", "u i", "c k", "ui ck", "Ġq uick", "Ġ f", "o x", "Ġf ox"]
|
94
|
+
tokenizer = keras_hub.models.RobertaTokenizer(
|
95
|
+
vocabulary=vocab,
|
96
|
+
merges=merges
|
97
|
+
)
|
98
|
+
preprocessor = keras_hub.models.RobertaPreprocessor(tokenizer)
|
99
|
+
preprocessor("a quick fox")
|
100
|
+
```
|
101
|
+
Mapping with `tf.data.Dataset`.
|
102
|
+
```python
|
103
|
+
preprocessor = keras_hub.models.RobertaPreprocessor.from_preset(
|
104
|
+
"roberta_base_en"
|
105
|
+
)
|
106
|
+
|
107
|
+
first = tf.constant(["The quick brown fox jumped.", "Call me Ishmael."])
|
108
|
+
second = tf.constant(["The fox tripped.", "Oh look, a whale."])
|
109
|
+
label = tf.constant([1, 1])
|
110
|
+
|
111
|
+
# Map labeled single sentences.
|
112
|
+
ds = tf.data.Dataset.from_tensor_slices((first, label))
|
113
|
+
ds = ds.map(preprocessor, num_parallel_calls=tf.data.AUTOTUNE)
|
114
|
+
|
115
|
+
# Map unlabeled single sentences.
|
116
|
+
ds = tf.data.Dataset.from_tensor_slices(first)
|
117
|
+
ds = ds.map(preprocessor, num_parallel_calls=tf.data.AUTOTUNE)
|
118
|
+
|
119
|
+
# Map labeled sentence pairs.
|
120
|
+
ds = tf.data.Dataset.from_tensor_slices(((first, second), label))
|
121
|
+
ds = ds.map(preprocessor, num_parallel_calls=tf.data.AUTOTUNE)
|
122
|
+
|
123
|
+
# Map unlabeled sentence pairs.
|
124
|
+
ds = tf.data.Dataset.from_tensor_slices((first, second))
|
125
|
+
# Watch out for tf.data's default unpacking of tuples here!
|
126
|
+
# Best to invoke the `preprocessor` directly in this case.
|
127
|
+
ds = ds.map(
|
128
|
+
lambda first, second: preprocessor(x=(first, second)),
|
129
|
+
num_parallel_calls=tf.data.AUTOTUNE,
|
130
|
+
)
|
131
|
+
```
|
132
|
+
"""
|
133
|
+
|
134
|
+
tokenizer_cls = RobertaTokenizer
|
135
|
+
|
136
|
+
def __init__(
|
137
|
+
self,
|
138
|
+
tokenizer,
|
139
|
+
sequence_length=512,
|
140
|
+
truncate="round_robin",
|
141
|
+
**kwargs,
|
142
|
+
):
|
143
|
+
super().__init__(**kwargs)
|
144
|
+
|
145
|
+
self.tokenizer = tokenizer
|
146
|
+
self.packer = None
|
147
|
+
self.truncate = truncate
|
148
|
+
self.sequence_length = sequence_length
|
149
|
+
|
150
|
+
def build(self, input_shape):
|
151
|
+
# Defer packer creation to `build()` so that we can be sure tokenizer
|
152
|
+
# assets have loaded when restoring a saved model.
|
153
|
+
self.packer = MultiSegmentPacker(
|
154
|
+
start_value=self.tokenizer.start_token_id,
|
155
|
+
end_value=self.tokenizer.end_token_id,
|
156
|
+
sep_value=[self.tokenizer.end_token_id] * 2,
|
157
|
+
pad_value=self.tokenizer.pad_token_id,
|
158
|
+
truncate=self.truncate,
|
159
|
+
sequence_length=self.sequence_length,
|
160
|
+
)
|
161
|
+
self.built = True
|
162
|
+
|
163
|
+
def call(self, x, y=None, sample_weight=None):
|
164
|
+
x = convert_inputs_to_list_of_tensor_segments(x)
|
165
|
+
x = [self.tokenizer(segment) for segment in x]
|
166
|
+
token_ids, _ = self.packer(x)
|
167
|
+
x = {
|
168
|
+
"token_ids": token_ids,
|
169
|
+
"padding_mask": token_ids != self.tokenizer.pad_token_id,
|
170
|
+
}
|
171
|
+
return keras.utils.pack_x_y_sample_weight(x, y, sample_weight)
|
172
|
+
|
173
|
+
def get_config(self):
|
174
|
+
config = super().get_config()
|
175
|
+
config.update(
|
176
|
+
{
|
177
|
+
"sequence_length": self.sequence_length,
|
178
|
+
"truncate": self.truncate,
|
179
|
+
}
|
180
|
+
)
|
181
|
+
return config
|
182
|
+
|
183
|
+
@property
|
184
|
+
def sequence_length(self):
|
185
|
+
"""The padded length of model input sequences."""
|
186
|
+
return self._sequence_length
|
187
|
+
|
188
|
+
@sequence_length.setter
|
189
|
+
def sequence_length(self, value):
|
190
|
+
self._sequence_length = value
|
191
|
+
if self.packer is not None:
|
192
|
+
self.packer.sequence_length = value
|
@@ -0,0 +1,43 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
"""RoBERTa model preset configurations."""
|
15
|
+
|
16
|
+
backbone_presets = {
|
17
|
+
"roberta_base_en": {
|
18
|
+
"metadata": {
|
19
|
+
"description": (
|
20
|
+
"12-layer RoBERTa model where case is maintained."
|
21
|
+
"Trained on English Wikipedia, BooksCorpus, CommonCraw, and OpenWebText."
|
22
|
+
),
|
23
|
+
"params": 124052736,
|
24
|
+
"official_name": "RoBERTa",
|
25
|
+
"path": "roberta",
|
26
|
+
"model_card": "https://github.com/facebookresearch/fairseq/blob/main/examples/roberta/README.md",
|
27
|
+
},
|
28
|
+
"kaggle_handle": "kaggle://keras/roberta/keras/roberta_base_en/2",
|
29
|
+
},
|
30
|
+
"roberta_large_en": {
|
31
|
+
"metadata": {
|
32
|
+
"description": (
|
33
|
+
"24-layer RoBERTa model where case is maintained."
|
34
|
+
"Trained on English Wikipedia, BooksCorpus, CommonCraw, and OpenWebText."
|
35
|
+
),
|
36
|
+
"params": 354307072,
|
37
|
+
"official_name": "RoBERTa",
|
38
|
+
"path": "roberta",
|
39
|
+
"model_card": "https://github.com/facebookresearch/fairseq/blob/main/examples/roberta/README.md",
|
40
|
+
},
|
41
|
+
"kaggle_handle": "kaggle://keras/roberta/keras/roberta_large_en/2",
|
42
|
+
},
|
43
|
+
}
|
@@ -0,0 +1,132 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
|
15
|
+
|
16
|
+
from keras_hub.src.api_export import keras_hub_export
|
17
|
+
from keras_hub.src.tokenizers.byte_pair_tokenizer import BytePairTokenizer
|
18
|
+
|
19
|
+
|
20
|
+
@keras_hub_export("keras_hub.models.RobertaTokenizer")
|
21
|
+
class RobertaTokenizer(BytePairTokenizer):
|
22
|
+
"""A RoBERTa tokenizer using Byte-Pair Encoding subword segmentation.
|
23
|
+
|
24
|
+
This tokenizer class will tokenize raw strings into integer sequences and
|
25
|
+
is based on `keras_hub.tokenizers.BytePairTokenizer`. Unlike the
|
26
|
+
underlying tokenizer, it will check for all special tokens needed by RoBERTa
|
27
|
+
models and provides a `from_preset()` method to automatically download
|
28
|
+
a matching vocabulary for a RoBERTa preset.
|
29
|
+
|
30
|
+
This tokenizer does not provide truncation or padding of inputs. It can be
|
31
|
+
combined with a `keras_hub.models.RobertaPreprocessor` layer for input
|
32
|
+
packing.
|
33
|
+
|
34
|
+
If input is a batch of strings (rank > 0), the layer will output a
|
35
|
+
`tf.RaggedTensor` where the last dimension of the output is ragged.
|
36
|
+
|
37
|
+
If input is a scalar string (rank == 0), the layer will output a dense
|
38
|
+
`tf.Tensor` with static shape `[None]`.
|
39
|
+
|
40
|
+
Args:
|
41
|
+
vocabulary: A dictionary mapping tokens to integer ids, or file path
|
42
|
+
to a json file containing the token to id mapping.
|
43
|
+
merges: A list of merge rules or a string file path, If passing a file
|
44
|
+
path. the file should have one merge rule per line. Every merge
|
45
|
+
rule contains merge entities separated by a space.
|
46
|
+
|
47
|
+
Examples:
|
48
|
+
```python
|
49
|
+
# Unbatched input.
|
50
|
+
tokenizer = keras_hub.models.RobertaTokenizer.from_preset(
|
51
|
+
"roberta_base_en",
|
52
|
+
)
|
53
|
+
tokenizer("The quick brown fox jumped.")
|
54
|
+
|
55
|
+
# Batched input.
|
56
|
+
tokenizer(["The quick brown fox jumped.", "The fox slept."])
|
57
|
+
|
58
|
+
# Detokenization.
|
59
|
+
tokenizer.detokenize(tokenizer("The quick brown fox jumped."))
|
60
|
+
|
61
|
+
# Custom vocabulary.
|
62
|
+
# Note: 'Ġ' is space
|
63
|
+
vocab = {"<s>": 0, "<pad>": 1, "</s>": 2, "<mask>": 3}
|
64
|
+
vocab = {**vocab, "a": 4, "Ġquick": 5, "Ġfox": 6}
|
65
|
+
merges = ["Ġ q", "u i", "c k", "ui ck", "Ġq uick"]
|
66
|
+
merges += ["Ġ f", "o x", "Ġf ox"]
|
67
|
+
tokenizer = keras_hub.models.RobertaTokenizer(
|
68
|
+
vocabulary=vocab,
|
69
|
+
merges=merges
|
70
|
+
)
|
71
|
+
tokenizer(["a quick fox", "a fox quick"])
|
72
|
+
```
|
73
|
+
"""
|
74
|
+
|
75
|
+
def __init__(
|
76
|
+
self,
|
77
|
+
vocabulary=None,
|
78
|
+
merges=None,
|
79
|
+
**kwargs,
|
80
|
+
):
|
81
|
+
self.start_token = "<s>"
|
82
|
+
self.pad_token = "<pad>"
|
83
|
+
self.end_token = "</s>"
|
84
|
+
self.mask_token = "<mask>"
|
85
|
+
|
86
|
+
super().__init__(
|
87
|
+
vocabulary=vocabulary,
|
88
|
+
merges=merges,
|
89
|
+
unsplittable_tokens=[
|
90
|
+
self.start_token,
|
91
|
+
self.pad_token,
|
92
|
+
self.end_token,
|
93
|
+
self.mask_token,
|
94
|
+
],
|
95
|
+
**kwargs,
|
96
|
+
)
|
97
|
+
|
98
|
+
def set_vocabulary_and_merges(self, vocabulary, merges):
|
99
|
+
super().set_vocabulary_and_merges(vocabulary, merges)
|
100
|
+
|
101
|
+
if vocabulary is not None:
|
102
|
+
# Check for necessary special tokens.
|
103
|
+
for token in [
|
104
|
+
self.start_token,
|
105
|
+
self.pad_token,
|
106
|
+
self.end_token,
|
107
|
+
self.mask_token,
|
108
|
+
]:
|
109
|
+
if token not in self.vocabulary:
|
110
|
+
raise ValueError(
|
111
|
+
f"Cannot find token `'{token}'` in the provided "
|
112
|
+
f"`vocabulary`. Please provide `'{token}'` in your "
|
113
|
+
"`vocabulary` or use a pretrained `vocabulary` name."
|
114
|
+
)
|
115
|
+
|
116
|
+
self.start_token_id = self.token_to_id(self.start_token)
|
117
|
+
self.pad_token_id = self.token_to_id(self.pad_token)
|
118
|
+
self.end_token_id = self.token_to_id(self.end_token)
|
119
|
+
self.mask_token_id = self.token_to_id(self.mask_token)
|
120
|
+
else:
|
121
|
+
self.start_token_id = None
|
122
|
+
self.pad_token_id = None
|
123
|
+
self.end_token_id = None
|
124
|
+
self.mask_token_id = None
|
125
|
+
|
126
|
+
def get_config(self):
|
127
|
+
config = super().get_config()
|
128
|
+
# In the constructor, we pass the list of special tokens to the
|
129
|
+
# `unsplittable_tokens` arg of the superclass' constructor. Hence, we
|
130
|
+
# delete it from the config here.
|
131
|
+
del config["unsplittable_tokens"]
|
132
|
+
return config
|
@@ -0,0 +1,54 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
from keras_hub.src.api_export import keras_hub_export
|
15
|
+
from keras_hub.src.models.causal_lm import CausalLM
|
16
|
+
|
17
|
+
|
18
|
+
@keras_hub_export("keras_hub.models.Seq2SeqLM")
|
19
|
+
class Seq2SeqLM(CausalLM):
|
20
|
+
"""Base class for sequence to sequence language modeling tasks.
|
21
|
+
|
22
|
+
`Seq2SeqLM` tasks wrap a `keras_hub.models.Backbone` and
|
23
|
+
a `keras_hub.models.Preprocessor` to create a model that can be used for
|
24
|
+
generation and generative fine-tuning, when generation is conditioned on
|
25
|
+
additional input sequence in a sequence-to-sequence setting.
|
26
|
+
|
27
|
+
`Seq2SeqLM` tasks provide an additional, high-level `generate()` function
|
28
|
+
which can be used to auto-regressively sample an output sequence token by
|
29
|
+
token. The `compile()` method of `Seq2SeqLM` classes contains an additional
|
30
|
+
`sampler` argument, which can be used to pass a `keras_hub.samplers.Sampler`
|
31
|
+
to control how the predicted distribution will be sampled.
|
32
|
+
|
33
|
+
When calling `fit()`, each input should contain an input and output
|
34
|
+
sequence. The model will be trained to predict the output sequence
|
35
|
+
token-by-token using a causal mask, similar to a `keras_hub.models.CausalLM`
|
36
|
+
task. Unlike the `CausalLM` task, an input sequence must be passed, and
|
37
|
+
can be attended to in full by all tokens in the output sequence.
|
38
|
+
|
39
|
+
All `Seq2SeqLM` tasks include a `from_preset()` constructor which can be
|
40
|
+
used to load a pre-trained config and weights.
|
41
|
+
|
42
|
+
Example:
|
43
|
+
```python
|
44
|
+
# Load a Bart backbone with pre-trained weights.
|
45
|
+
seq_2_seq_lm = keras_hub.models.Seq2SeqLM.from_preset(
|
46
|
+
"bart_base_en",
|
47
|
+
)
|
48
|
+
seq_2_seq_lm.compile(sampler="top_k")
|
49
|
+
# Generate conditioned on the `"The quick brown fox."` as an input sequence.
|
50
|
+
seq_2_seq_lm.generate("The quick brown fox.", max_length=30)
|
51
|
+
```
|
52
|
+
"""
|
53
|
+
|
54
|
+
# TODO: fill in during https://github.com/keras-team/keras-hub/pull/1425
|
@@ -0,0 +1,20 @@
|
|
1
|
+
# Copyright 2024 The KerasHub Authors
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing, software
|
10
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
|
15
|
+
from keras_hub.src.models.t5.t5_backbone import T5Backbone
|
16
|
+
from keras_hub.src.models.t5.t5_presets import backbone_presets
|
17
|
+
from keras_hub.src.models.t5.t5_tokenizer import T5Tokenizer
|
18
|
+
from keras_hub.src.utils.preset_utils import register_presets
|
19
|
+
|
20
|
+
register_presets(backbone_presets, (T5Backbone, T5Tokenizer))
|