├── .gitignore ├── README.md ├── bash ├── download_weights_1.sh ├── download_weights_2.sh ├── env_setup.sh ├── install_deps.sh ├── jarvislabs_env_setup.sh └── sync_code_kaggle.sh ├── config ├── config.yaml ├── paths │ ├── colab.yaml │ ├── jarvislabs.yaml │ └── kaggle.yaml └── trainer │ ├── alberta_xxlarge_v2.yaml │ ├── bart_base.yaml │ ├── bert_for_patents.yaml │ ├── bert_large_uncased.yaml │ ├── cocolm_base.yaml │ ├── cocolm_large.yaml │ ├── deberta_base.yaml │ ├── deberta_large.yaml │ ├── deberta_v2x_large.yaml │ ├── deberta_v2xx_large.yaml │ ├── deberta_v3_base.yaml │ ├── deberta_v3_large.yaml │ ├── deberta_v3_small.yaml │ ├── deberta_xlarge.yaml │ ├── default_trainer.yaml │ ├── electra_large.yaml │ ├── funnel_large.yaml │ ├── funnel_xlarge.yaml │ ├── mpnet_patentsberta.yaml │ ├── nystromformer_512.yaml │ ├── patentsberta_v3.yaml │ ├── roberta_base.yaml │ └── roberta_large.yaml ├── run.sh ├── run_scripts ├── 100_105_msd_attn_pool.sh ├── 112transformer_head.sh ├── 113_120_anferico_bert.sh ├── 127_130_mean_conv_pool_dv3large.sh ├── 131_134_mean_conv1d_bforpatent.sh ├── 135_136_mean_pool_layernorm.sh ├── 137_140_max_meanmax_pool_dv3large.sh ├── 141_144_max_meanmax_bforpatent.sh ├── 145_148_cocolm.sh ├── 149_150_deberta_v2_xlarge.sh ├── 151_152_deberta_v2_xlarge.sh ├── 153_154_deberta_xlarge.sh ├── 155_157_mpnet_patentsberta.sh ├── 158_163_roberta_base_large.sh ├── 164_166_albert_xxlargev2.sh ├── 167_169_electra_large.sh ├── 170_175_funnel_large_xlarge.sh ├── 176_178_albert_xxlargev2_lowlr.sh ├── 179_181_deberta_v3_small.sh ├── 182_184_deberta_large.sh ├── 185_187_patentsbertav3.sh ├── 188_190_bert_large_uncased.sh ├── 191_193_nystromformer.sh ├── 194_196_bart_base.sh ├── 196_199_deberta_v2_xlarge.sh ├── 200_203_deberta_v2_xlarge.sh ├── 204_205_cocolm_msd.sh ├── 206_207_cocolm_attn_pool.sh ├── 208_209_cocolm_conv1d_pool.sh ├── 210_211_cocolm_mean_max_pool.sh ├── 212_217_electra_msd_attn_pool.sh ├── 218_223_electra_conv_mean_max_pool.sh ├── 95_97_baseline_fin.sh └── preliminary_test_script.sh ├── secrets └── .env └── src ├── cocolm ├── configuration_cocolm.py ├── convert_state_dict.py ├── modeling_cocolm.py ├── tokenization_cocolm.py └── tokenization_utils.py ├── data ├── cpc_texts.py └── dataset.py ├── extensions ├── callbacks.py ├── logger.py ├── scoring.py └── trainer.py ├── modeling ├── __init__.py ├── mixins.py └── models │ ├── bert │ ├── bert.py │ ├── bert_cnn1d_pool.py │ ├── bert_max_pool.py │ ├── bert_mean_max_pool.py │ ├── bert_mean_pool.py │ └── bert_mean_pool_layernorm.py │ ├── cocolm │ ├── cocolm.py │ ├── cocolm_conv1d_pool.py │ ├── cocolm_general.py │ └── cocolm_mean_max_concatenate_pool.py │ ├── deberta_v2 │ ├── deberta_v2_attn_head.py │ ├── deberta_v2_bce.py │ ├── deberta_v2_ce.py │ ├── deberta_v2_cnn1d_pool.py │ ├── deberta_v2_concatenate_pooling.py │ ├── deberta_v2_corrloss.py │ ├── deberta_v2_general.py │ ├── deberta_v2_max_pool.py │ ├── deberta_v2_mean_max_pool.py │ ├── deberta_v2_mean_pool.py │ ├── deberta_v2_mean_pool_layernorm.py │ ├── deberta_v2_mse.py │ └── deberta_v2_transformer_head.py │ └── electra │ ├── electra_conv1d_pool.py │ ├── electra_general.py │ └── electra_mean_max_pool.py ├── train.py └── utils.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/.gitignore -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/README.md -------------------------------------------------------------------------------- /bash/download_weights_1.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/download_weights_1.sh -------------------------------------------------------------------------------- /bash/download_weights_2.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/download_weights_2.sh -------------------------------------------------------------------------------- /bash/env_setup.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/env_setup.sh -------------------------------------------------------------------------------- /bash/install_deps.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/install_deps.sh -------------------------------------------------------------------------------- /bash/jarvislabs_env_setup.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/jarvislabs_env_setup.sh -------------------------------------------------------------------------------- /bash/sync_code_kaggle.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/bash/sync_code_kaggle.sh -------------------------------------------------------------------------------- /config/config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/config.yaml -------------------------------------------------------------------------------- /config/paths/colab.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/paths/colab.yaml -------------------------------------------------------------------------------- /config/paths/jarvislabs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/paths/jarvislabs.yaml -------------------------------------------------------------------------------- /config/paths/kaggle.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/paths/kaggle.yaml -------------------------------------------------------------------------------- /config/trainer/alberta_xxlarge_v2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/alberta_xxlarge_v2.yaml -------------------------------------------------------------------------------- /config/trainer/bart_base.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/bart_base.yaml -------------------------------------------------------------------------------- /config/trainer/bert_for_patents.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/bert_for_patents.yaml -------------------------------------------------------------------------------- /config/trainer/bert_large_uncased.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/bert_large_uncased.yaml -------------------------------------------------------------------------------- /config/trainer/cocolm_base.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/cocolm_base.yaml -------------------------------------------------------------------------------- /config/trainer/cocolm_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/cocolm_large.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_base.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_base.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_large.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_v2x_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_v2x_large.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_v2xx_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_v2xx_large.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_v3_base.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_v3_base.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_v3_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_v3_large.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_v3_small.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_v3_small.yaml -------------------------------------------------------------------------------- /config/trainer/deberta_xlarge.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/deberta_xlarge.yaml -------------------------------------------------------------------------------- /config/trainer/default_trainer.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/default_trainer.yaml -------------------------------------------------------------------------------- /config/trainer/electra_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/electra_large.yaml -------------------------------------------------------------------------------- /config/trainer/funnel_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/funnel_large.yaml -------------------------------------------------------------------------------- /config/trainer/funnel_xlarge.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/funnel_xlarge.yaml -------------------------------------------------------------------------------- /config/trainer/mpnet_patentsberta.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/mpnet_patentsberta.yaml -------------------------------------------------------------------------------- /config/trainer/nystromformer_512.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/nystromformer_512.yaml -------------------------------------------------------------------------------- /config/trainer/patentsberta_v3.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/patentsberta_v3.yaml -------------------------------------------------------------------------------- /config/trainer/roberta_base.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/roberta_base.yaml -------------------------------------------------------------------------------- /config/trainer/roberta_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/config/trainer/roberta_large.yaml -------------------------------------------------------------------------------- /run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run.sh -------------------------------------------------------------------------------- /run_scripts/100_105_msd_attn_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/100_105_msd_attn_pool.sh -------------------------------------------------------------------------------- /run_scripts/112transformer_head.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/112transformer_head.sh -------------------------------------------------------------------------------- /run_scripts/113_120_anferico_bert.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/113_120_anferico_bert.sh -------------------------------------------------------------------------------- /run_scripts/127_130_mean_conv_pool_dv3large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/127_130_mean_conv_pool_dv3large.sh -------------------------------------------------------------------------------- /run_scripts/131_134_mean_conv1d_bforpatent.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/131_134_mean_conv1d_bforpatent.sh -------------------------------------------------------------------------------- /run_scripts/135_136_mean_pool_layernorm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/135_136_mean_pool_layernorm.sh -------------------------------------------------------------------------------- /run_scripts/137_140_max_meanmax_pool_dv3large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/137_140_max_meanmax_pool_dv3large.sh -------------------------------------------------------------------------------- /run_scripts/141_144_max_meanmax_bforpatent.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/141_144_max_meanmax_bforpatent.sh -------------------------------------------------------------------------------- /run_scripts/145_148_cocolm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/145_148_cocolm.sh -------------------------------------------------------------------------------- /run_scripts/149_150_deberta_v2_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/149_150_deberta_v2_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/151_152_deberta_v2_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/151_152_deberta_v2_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/153_154_deberta_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/153_154_deberta_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/155_157_mpnet_patentsberta.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/155_157_mpnet_patentsberta.sh -------------------------------------------------------------------------------- /run_scripts/158_163_roberta_base_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/158_163_roberta_base_large.sh -------------------------------------------------------------------------------- /run_scripts/164_166_albert_xxlargev2.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/164_166_albert_xxlargev2.sh -------------------------------------------------------------------------------- /run_scripts/167_169_electra_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/167_169_electra_large.sh -------------------------------------------------------------------------------- /run_scripts/170_175_funnel_large_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/170_175_funnel_large_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/176_178_albert_xxlargev2_lowlr.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/176_178_albert_xxlargev2_lowlr.sh -------------------------------------------------------------------------------- /run_scripts/179_181_deberta_v3_small.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/179_181_deberta_v3_small.sh -------------------------------------------------------------------------------- /run_scripts/182_184_deberta_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/182_184_deberta_large.sh -------------------------------------------------------------------------------- /run_scripts/185_187_patentsbertav3.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/185_187_patentsbertav3.sh -------------------------------------------------------------------------------- /run_scripts/188_190_bert_large_uncased.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/188_190_bert_large_uncased.sh -------------------------------------------------------------------------------- /run_scripts/191_193_nystromformer.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/191_193_nystromformer.sh -------------------------------------------------------------------------------- /run_scripts/194_196_bart_base.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/194_196_bart_base.sh -------------------------------------------------------------------------------- /run_scripts/196_199_deberta_v2_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/196_199_deberta_v2_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/200_203_deberta_v2_xlarge.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/200_203_deberta_v2_xlarge.sh -------------------------------------------------------------------------------- /run_scripts/204_205_cocolm_msd.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/204_205_cocolm_msd.sh -------------------------------------------------------------------------------- /run_scripts/206_207_cocolm_attn_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/206_207_cocolm_attn_pool.sh -------------------------------------------------------------------------------- /run_scripts/208_209_cocolm_conv1d_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/208_209_cocolm_conv1d_pool.sh -------------------------------------------------------------------------------- /run_scripts/210_211_cocolm_mean_max_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/210_211_cocolm_mean_max_pool.sh -------------------------------------------------------------------------------- /run_scripts/212_217_electra_msd_attn_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/212_217_electra_msd_attn_pool.sh -------------------------------------------------------------------------------- /run_scripts/218_223_electra_conv_mean_max_pool.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/218_223_electra_conv_mean_max_pool.sh -------------------------------------------------------------------------------- /run_scripts/95_97_baseline_fin.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/95_97_baseline_fin.sh -------------------------------------------------------------------------------- /run_scripts/preliminary_test_script.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/run_scripts/preliminary_test_script.sh -------------------------------------------------------------------------------- /secrets/.env: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/secrets/.env -------------------------------------------------------------------------------- /src/cocolm/configuration_cocolm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/cocolm/configuration_cocolm.py -------------------------------------------------------------------------------- /src/cocolm/convert_state_dict.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/cocolm/convert_state_dict.py -------------------------------------------------------------------------------- /src/cocolm/modeling_cocolm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/cocolm/modeling_cocolm.py -------------------------------------------------------------------------------- /src/cocolm/tokenization_cocolm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/cocolm/tokenization_cocolm.py -------------------------------------------------------------------------------- /src/cocolm/tokenization_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/cocolm/tokenization_utils.py -------------------------------------------------------------------------------- /src/data/cpc_texts.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/data/cpc_texts.py -------------------------------------------------------------------------------- /src/data/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/data/dataset.py -------------------------------------------------------------------------------- /src/extensions/callbacks.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/extensions/callbacks.py -------------------------------------------------------------------------------- /src/extensions/logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/extensions/logger.py -------------------------------------------------------------------------------- /src/extensions/scoring.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/extensions/scoring.py -------------------------------------------------------------------------------- /src/extensions/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/extensions/trainer.py -------------------------------------------------------------------------------- /src/modeling/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/__init__.py -------------------------------------------------------------------------------- /src/modeling/mixins.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/mixins.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert_cnn1d_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert_cnn1d_pool.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert_max_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert_max_pool.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert_mean_max_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert_mean_max_pool.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert_mean_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert_mean_pool.py -------------------------------------------------------------------------------- /src/modeling/models/bert/bert_mean_pool_layernorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/bert/bert_mean_pool_layernorm.py -------------------------------------------------------------------------------- /src/modeling/models/cocolm/cocolm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/cocolm/cocolm.py -------------------------------------------------------------------------------- /src/modeling/models/cocolm/cocolm_conv1d_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/cocolm/cocolm_conv1d_pool.py -------------------------------------------------------------------------------- /src/modeling/models/cocolm/cocolm_general.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/cocolm/cocolm_general.py -------------------------------------------------------------------------------- /src/modeling/models/cocolm/cocolm_mean_max_concatenate_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/cocolm/cocolm_mean_max_concatenate_pool.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_attn_head.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_attn_head.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_bce.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_bce.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_ce.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_ce.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_cnn1d_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_cnn1d_pool.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_concatenate_pooling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_concatenate_pooling.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_corrloss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_corrloss.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_general.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_general.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_max_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_max_pool.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_mean_max_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_mean_max_pool.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_mean_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_mean_pool.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_mean_pool_layernorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_mean_pool_layernorm.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_mse.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_mse.py -------------------------------------------------------------------------------- /src/modeling/models/deberta_v2/deberta_v2_transformer_head.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/deberta_v2/deberta_v2_transformer_head.py -------------------------------------------------------------------------------- /src/modeling/models/electra/electra_conv1d_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/electra/electra_conv1d_pool.py -------------------------------------------------------------------------------- /src/modeling/models/electra/electra_general.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/electra/electra_general.py -------------------------------------------------------------------------------- /src/modeling/models/electra/electra_mean_max_pool.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/modeling/models/electra/electra_mean_max_pool.py -------------------------------------------------------------------------------- /src/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/train.py -------------------------------------------------------------------------------- /src/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Gladiator07/U.S.-Patent-Phrase-to-Phrase-Matching-Kaggle/HEAD/src/utils.py --------------------------------------------------------------------------------