--- datasets: - tner/tweetner7 metrics: - f1 - precision - recall pipeline_tag: token-classification widget: - text: 'Get the all-analog Classic Vinyl Edition of `Takin'' Off` Album from {@herbiehancock@} via {@bluenoterecords@} link below: {{URL}}' example_title: NER Example 1 base_model: vinai/bertweet-large model-index: - name: tner/bertweet-large-tweetner7-all results: - task: type: token-classification name: Token Classification dataset: name: tner/tweetner7 type: tner/tweetner7 args: tner/tweetner7 metrics: - type: f1 value: 0.6646206308610401 name: F1 (test_2021) - type: precision value: 0.653515144741254 name: Precision (test_2021) - type: recall value: 0.6761100832562442 name: Recall (test_2021) - type: f1_macro value: 0.6187282305429461 name: Macro F1 (test_2021) - type: precision_macro value: 0.6069581336386037 name: Macro Precision (test_2021) - type: recall_macro value: 0.6359356515638321 name: Macro Recall (test_2021) - type: f1_entity_span value: 0.7953163189905075 name: Entity Span F1 (test_2021) - type: precision_entity_span value: 0.7820254862508383 name: Entity Span Precision (test_2020) - type: recall_entity_span value: 0.8090667283450907 name: Entity Span Recall (test_2021) - type: f1 value: 0.6675704989154012 name: F1 (test_2020) - type: precision value: 0.6990346394094265 name: Precision (test_2020) - type: recall value: 0.6388168137000519 name: Recall (test_2020) - type: f1_macro value: 0.6307734401161805 name: Macro F1 (test_2020) - type: precision_macro value: 0.6616549497337102 name: Macro Precision (test_2020) - type: recall_macro value: 0.6085863177550797 name: Macro Recall (test_2020) - type: f1_entity_span value: 0.7759088442756376 name: Entity Span F1 (test_2020) - type: precision_entity_span value: 0.8129619101762365 name: Entity Span Precision (test_2020) - type: recall_entity_span value: 0.742086144265698 name: Entity Span Recall (test_2020) --- # tner/bertweet-large-tweetner7-all This model is a fine-tuned version of [vinai/bertweet-large](https://huggingface.co./vinai/bertweet-large) on the [tner/tweetner7](https://huggingface.co./datasets/tner/tweetner7) dataset (`train_all` split). Model fine-tuning is done via [T-NER](https://github.com/asahi417/tner)'s hyper-parameter search (see the repository for more detail). It achieves the following results on the test set of 2021: - F1 (micro): 0.6646206308610401 - Precision (micro): 0.653515144741254 - Recall (micro): 0.6761100832562442 - F1 (macro): 0.6187282305429461 - Precision (macro): 0.6069581336386037 - Recall (macro): 0.6359356515638321 The per-entity breakdown of the F1 score on the test set are below: - corporation: 0.545042492917847 - creative_work: 0.47362250879249707 - event: 0.4915336236090953 - group: 0.623768877216021 - location: 0.6754716981132076 - person: 0.8414922656960875 - product: 0.6801661474558671 For F1 scores, the confidence interval is obtained by bootstrap as below: - F1 (micro): - 90%: [0.6554125820888649, 0.6736489128168938] - 95%: [0.6533077908395879, 0.675252368755536] - F1 (macro): - 90%: [0.6554125820888649, 0.6736489128168938] - 95%: [0.6533077908395879, 0.675252368755536] Full evaluation can be found at [metric file of NER](https://huggingface.co./tner/bertweet-large-tweetner7-all/raw/main/eval/metric.json) and [metric file of entity span](https://huggingface.co./tner/bertweet-large-tweetner7-all/raw/main/eval/metric_span.json). ### Usage This model can be used through the [tner library](https://github.com/asahi417/tner). Install the library via pip. ```shell pip install tner ``` [TweetNER7](https://huggingface.co./datasets/tner/tweetner7) pre-processed tweets where the account name and URLs are converted into special formats (see the dataset page for more detail), so we process tweets accordingly and then run the model prediction as below. ```python import re from urlextract import URLExtract from tner import TransformersNER extractor = URLExtract() def format_tweet(tweet): # mask web urls urls = extractor.find_urls(tweet) for url in urls: tweet = tweet.replace(url, "{{URL}}") # format twitter account tweet = re.sub(r"\b(\s*)(@[\S]+)\b", r'\1{\2@}', tweet) return tweet text = "Get the all-analog Classic Vinyl Edition of `Takin' Off` Album from @herbiehancock via @bluenoterecords link below: http://bluenote.lnk.to/AlbumOfTheWeek" text_format = format_tweet(text) model = TransformersNER("tner/bertweet-large-tweetner7-all") model.predict([text_format]) ``` It can be used via transformers library but it is not recommended as CRF layer is not supported at the moment. ### Training hyperparameters The following hyperparameters were used during training: - dataset: ['tner/tweetner7'] - dataset_split: train_all - dataset_name: None - local_dataset: None - model: vinai/bertweet-large - crf: True - max_length: 128 - epoch: 30 - batch_size: 32 - lr: 1e-05 - random_seed: 0 - gradient_accumulation_steps: 1 - weight_decay: 1e-07 - lr_warmup_step_ratio: 0.15 - max_grad_norm: 1 The full configuration can be found at [fine-tuning parameter file](https://huggingface.co./tner/bertweet-large-tweetner7-all/raw/main/trainer_config.json). ### Reference If you use the model, please cite T-NER paper and TweetNER7 paper. - T-NER ``` @inproceedings{ushio-camacho-collados-2021-ner, title = "{T}-{NER}: An All-Round Python Library for Transformer-based Named Entity Recognition", author = "Ushio, Asahi and Camacho-Collados, Jose", booktitle = "Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations", month = apr, year = "2021", address = "Online", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2021.eacl-demos.7", doi = "10.18653/v1/2021.eacl-demos.7", pages = "53--62", abstract = "Language model (LM) pretraining has led to consistent improvements in many NLP downstream tasks, including named entity recognition (NER). In this paper, we present T-NER (Transformer-based Named Entity Recognition), a Python library for NER LM finetuning. In addition to its practical utility, T-NER facilitates the study and investigation of the cross-domain and cross-lingual generalization ability of LMs finetuned on NER. Our library also provides a web app where users can get model predictions interactively for arbitrary text, which facilitates qualitative model evaluation for non-expert programmers. We show the potential of the library by compiling nine public NER datasets into a unified format and evaluating the cross-domain and cross- lingual performance across the datasets. The results from our initial experiments show that in-domain performance is generally competitive across datasets. However, cross-domain generalization is challenging even with a large pretrained LM, which has nevertheless capacity to learn domain-specific features if fine- tuned on a combined dataset. To facilitate future research, we also release all our LM checkpoints via the Hugging Face model hub.", } ``` - TweetNER7 ``` @inproceedings{ushio-etal-2022-tweet, title = "{N}amed {E}ntity {R}ecognition in {T}witter: {A} {D}ataset and {A}nalysis on {S}hort-{T}erm {T}emporal {S}hifts", author = "Ushio, Asahi and Neves, Leonardo and Silva, Vitor and Barbieri, Francesco. and Camacho-Collados, Jose", booktitle = "The 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing", month = nov, year = "2022", address = "Online", publisher = "Association for Computational Linguistics", } ```