From f5a236c3ca45a4f83e6cfead489b2f5d2db754a8 Mon Sep 17 00:00:00 2001 From: Wietse de Vries Date: Thu, 19 Dec 2019 13:28:53 +0100 Subject: [PATCH] Add Dutch pre-trained BERT model --- docs/source/pretrained_models.rst | 4 ++++ src/transformers/configuration_bert.py | 1 + src/transformers/modeling_bert.py | 1 + src/transformers/modeling_tf_bert.py | 1 + src/transformers/tokenization_bert.py | 3 +++ 5 files changed, 10 insertions(+) diff --git a/docs/source/pretrained_models.rst b/docs/source/pretrained_models.rst index eb7b41ffc9..39e942e3e4 100644 --- a/docs/source/pretrained_models.rst +++ b/docs/source/pretrained_models.rst @@ -88,6 +88,10 @@ For a list that includes community-uploaded models, refer to `https://huggingfac | | ``bert-base-finnish-uncased-v1`` | | 12-layer, 768-hidden, 12-heads, 110M parameters. | | | | | Trained on uncased Finnish text. | | | | (see `details on turkunlp.org `__). | +| +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+ +| | ``bert-base-dutch-cased`` | | 12-layer, 768-hidden, 12-heads, 110M parameters. | +| | | | Trained on cased Dutch text. | +| | | (see `details on wietsedv repository `__). | +-------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+ | GPT | ``openai-gpt`` | | 12-layer, 768-hidden, 12-heads, 110M parameters. | | | | | OpenAI GPT English model | diff --git a/src/transformers/configuration_bert.py b/src/transformers/configuration_bert.py index 71d190af8a..d668d04cb8 100644 --- a/src/transformers/configuration_bert.py +++ b/src/transformers/configuration_bert.py @@ -45,6 +45,7 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = { "bert-base-japanese-char-whole-word-masking": "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json", "bert-base-finnish-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/config.json", "bert-base-finnish-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/config.json", + "bert-base-dutch-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/wietsedv/bert-base-dutch-cased/config.json", } diff --git a/src/transformers/modeling_bert.py b/src/transformers/modeling_bert.py index 7a881a0805..62e752b89c 100644 --- a/src/transformers/modeling_bert.py +++ b/src/transformers/modeling_bert.py @@ -53,6 +53,7 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = { "bert-base-japanese-char-whole-word-masking": "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin", "bert-base-finnish-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/pytorch_model.bin", "bert-base-finnish-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/pytorch_model.bin", + "bert-base-dutch-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/wietsedv/bert-base-dutch-cased/pytorch_model.bin", } diff --git a/src/transformers/modeling_tf_bert.py b/src/transformers/modeling_tf_bert.py index 08b7b7e466..9c0ca51da7 100644 --- a/src/transformers/modeling_tf_bert.py +++ b/src/transformers/modeling_tf_bert.py @@ -49,6 +49,7 @@ TF_BERT_PRETRAINED_MODEL_ARCHIVE_MAP = { "bert-base-japanese-char-whole-word-masking": "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-tf_model.h5", "bert-base-finnish-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/tf_model.h5", "bert-base-finnish-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/tf_model.h5", + "bert-base-dutch-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/wietsedv/bert-base-dutch-cased/tf_model.h5" } diff --git a/src/transformers/tokenization_bert.py b/src/transformers/tokenization_bert.py index cefb917798..98aa87766f 100644 --- a/src/transformers/tokenization_bert.py +++ b/src/transformers/tokenization_bert.py @@ -48,6 +48,7 @@ PRETRAINED_VOCAB_FILES_MAP = { "bert-base-german-dbmdz-uncased": "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-vocab.txt", "bert-base-finnish-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-cased-v1/vocab.txt", "bert-base-finnish-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/TurkuNLP/bert-base-finnish-uncased-v1/vocab.txt", + "bert-base-dutch-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/wietsedv/bert-base-dutch-cased/vocab.txt", } } @@ -69,6 +70,7 @@ PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = { "bert-base-german-dbmdz-uncased": 512, "bert-base-finnish-cased-v1": 512, "bert-base-finnish-uncased-v1": 512, + "bert-base-dutch-cased": 512, } PRETRAINED_INIT_CONFIGURATION = { @@ -89,6 +91,7 @@ PRETRAINED_INIT_CONFIGURATION = { "bert-base-german-dbmdz-uncased": {"do_lower_case": True}, "bert-base-finnish-cased-v1": {"do_lower_case": False}, "bert-base-finnish-uncased-v1": {"do_lower_case": True}, + "bert-base-dutch-cased": {'do_lower_case': False}, }