model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bertweet-large-tweetner7-2020
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: tner/tweetner7/test_2021
+      type: tner/tweetner7/test_2021
+      args: tner/tweetner7/test_2021
+    metrics:
+    - name: F1
+      type: f1
+      value: 0.6401254269555967
+    - name: Precision
+      type: precision
+      value: 0.6205623710780589
+    - name: Recall
+      type: recall
+      value: 0.6609620721554117
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5947383155381057
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5738855505495571
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6206178838164583
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7826184343151529
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7586581261535121
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.8081415519833468
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: tner/tweetner7/test_2020
+      type: tner/tweetner7/test_2020
+      args: tner/tweetner7/test_2020
+    metrics:
+    - name: F1
+      type: f1
+      value: 0.659346545259775
+    - name: Precision
+      type: precision
+      value: 0.6812396236856668
+    - name: Recall
+      type: recall
+      value: 0.6388168137000519
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.6261309560026784
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6527657911787169
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6111694484964181
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7738478027867096
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.8
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.749351323300467
+
+pipeline_tag: token-classification
+widget:
+- text: "Get the all-analog Classic Vinyl Edition of `Takin' Off` Album from {{@Herbie Hancock@}} via {{USERNAME}} link below: {{URL}}"
+  example_title: "NER Example 1"
+---
+# tner/bertweet-large-tweetner7-2020
+
+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_2020` 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): 0.6401254269555967
+- Precision (micro): 0.6205623710780589
+- Recall (micro): 0.6609620721554117
+- F1 (macro): 0.5947383155381057
+- Precision (macro): 0.5738855505495571
+- Recall (macro): 0.6206178838164583
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.5229357798165137
+- creative_work: 0.4629981024667932
+- event: 0.4499572284003422
+- group: 0.592749032030975
+- location: 0.6553030303030303
+- person: 0.8273135669362084
+- product: 0.6519114688128772 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6315544728348781, 0.6491758274095626]
+    - 95%: [0.6294268706225905, 0.6515448119225267] 
+- F1 (macro): 
+    - 90%: [0.6315544728348781, 0.6491758274095626]
+    - 95%: [0.6294268706225905, 0.6515448119225267] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bertweet-large-tweetner7-2020/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bertweet-large-tweetner7-2020/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
+```
+and activate model as below.
+```python
+from tner import TransformersNER
+model = TransformersNER("tner/bertweet-large-tweetner7-2020")
+model.predict(["Jacob Collier is a Grammy awarded English artist from London"])
+```
+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_2020
+ - 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: 0.3
+ - max_grad_norm: 1
+
+The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/tner/bertweet-large-tweetner7-2020/raw/main/trainer_config.json).
+
+### Reference
+If you use any resource from T-NER, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/).
+
+```
+
+@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.",
+}
+
+```

eval/metric.json

@@ -1 +0,0 @@
-{"2020.dev": {"micro/f1": 0.6437029063509151, "micro/f1_ci": {}, "micro/recall": 0.6248693834900731, "micro/precision": 0.6637069922308546, "macro/f1": 0.591642865243014, "macro/f1_ci": {}, "macro/recall": 0.5755700410698614, "macro/precision": 0.6133965500147562, "per_entity_metric": {"corporation": {"f1": 0.5051546391752577, "f1_ci": {}, "precision": 0.5297297297297298, "recall": 0.4827586206896552}, "creative_work": {"f1": 0.54, "f1_ci": {}, "precision": 0.5625, "recall": 0.5192307692307693}, "event": {"f1": 0.39312977099236646, "f1_ci": {}, "precision": 0.3843283582089552, "recall": 0.40234375}, "group": {"f1": 0.5495049504950495, "f1_ci": {}, "precision": 0.6271186440677966, "recall": 0.4889867841409692}, "location": {"f1": 0.6227848101265823, "f1_ci": {}, "precision": 0.5747663551401869, "recall": 0.6795580110497238}, "person": {"f1": 0.874675885911841, "f1_ci": {}, "precision": 0.9051878354203936, "recall": 0.8461538461538461}, "product": {"f1": 0.6562500000000001, "f1_ci": {}, "precision": 0.7101449275362319, "recall": 0.6099585062240664}}}, "2021.test": {"micro/f1": 0.6401254269555967, "micro/f1_ci": {"90": [0.6315544728348781, 0.6491758274095626], "95": [0.6294268706225905, 0.6515448119225267]}, "micro/recall": 0.6609620721554117, "micro/precision": 0.6205623710780589, "macro/f1": 0.5947383155381057, "macro/f1_ci": {"90": [0.5852724504065837, 0.6043623226898465], "95": [0.5836410112124547, 0.6063768583335745]}, "macro/recall": 0.6206178838164583, "macro/precision": 0.5738855505495571, "per_entity_metric": {"corporation": {"f1": 0.5229357798165137, "f1_ci": {"90": [0.49706039551042225, 0.5489060654201801], "95": [0.49299356150347506, 0.5543603424996233]}, "precision": 0.4830508474576271, "recall": 0.57}, "creative_work": {"f1": 0.4629981024667932, "f1_ci": {"90": [0.43277573053403184, 0.4925610964986595], "95": [0.4272513399167463, 0.49761378932009187]}, "precision": 0.43058823529411766, "recall": 0.5006839945280438}, "event": {"f1": 0.4499572284003422, "f1_ci": {"90": [0.4274287289095753, 0.47154022208242086], "95": [0.4228081863105448, 0.47623473640595676]}, "precision": 0.4245359160613398, "recall": 0.4786169244767971}, "group": {"f1": 0.592749032030975, "f1_ci": {"90": [0.5722540581576642, 0.6147130977130977], "95": [0.5676533946830992, 0.6194597542880659]}, "precision": 0.636432350718065, "recall": 0.5546772068511199}, "location": {"f1": 0.6553030303030303, "f1_ci": {"90": [0.6304163397345088, 0.6817280494561573], "95": [0.6255962515548817, 0.6867043413492904]}, "precision": 0.597926267281106, "recall": 0.7248603351955307}, "person": {"f1": 0.8273135669362084, "f1_ci": {"90": [0.8160811561878993, 0.8380472862595445], "95": [0.814691063129817, 0.8397603550670991]}, "precision": 0.806869961444094, "recall": 0.8488200589970502}, "product": {"f1": 0.6519114688128772, "f1_ci": {"90": [0.6299024970087858, 0.6725193990424303], "95": [0.6268028691225062, 0.6764233561966475]}, "precision": 0.6377952755905512, "recall": 0.6666666666666666}}}, "2020.test": {"micro/f1": 0.659346545259775, "micro/f1_ci": {"90": [0.638530476656731, 0.6770002736609395], "95": [0.6355787571672109, 0.6814383304317009]}, "micro/recall": 0.6388168137000519, "micro/precision": 0.6812396236856668, "macro/f1": 0.6261309560026784, "macro/f1_ci": {"90": [0.6035614787399021, 0.6466837937863731], "95": [0.6008212526260872, 0.6498051149764593]}, "macro/recall": 0.6111694484964181, "macro/precision": 0.6527657911787169, "per_entity_metric": {"corporation": {"f1": 0.5966587112171838, "f1_ci": {"90": [0.5434307846076962, 0.6461646550171141], "95": [0.5326863869413766, 0.6552623249971564]}, "precision": 0.5482456140350878, "recall": 0.6544502617801047}, "creative_work": {"f1": 0.5892351274787537, "f1_ci": {"90": [0.5297221164826799, 0.6412325295653509], "95": [0.5167939021229471, 0.6498711140898077]}, "precision": 0.5977011494252874, "recall": 0.5810055865921788}, "event": {"f1": 0.4500907441016334, "f1_ci": {"90": [0.4, 0.5], "95": [0.3933869964988056, 0.5071014607812004]}, "precision": 0.43356643356643354, "recall": 0.4679245283018868}, "group": {"f1": 0.5454545454545454, "f1_ci": {"90": [0.4908707249478041, 0.5995649027168184], "95": [0.47806491874824353, 0.6098095911901321]}, "precision": 0.7076923076923077, "recall": 0.4437299035369775}, "location": {"f1": 0.6808510638297872, "f1_ci": {"90": [0.6158357771260997, 0.7378894784375705], "95": [0.6056659605996418, 0.7468373964681981]}, "precision": 0.6829268292682927, "recall": 0.6787878787878788}, "person": {"f1": 0.8358974358974359, "f1_ci": {"90": [0.8090079967608057, 0.8589565954118873], "95": [0.802661820512157, 0.864705203631851]}, "precision": 0.8519163763066202, "recall": 0.8204697986577181}, "product": {"f1": 0.6847290640394088, "f1_ci": {"90": [0.6328984758460346, 0.731190820093083], "95": [0.6213509150900847, 0.7403650506863269]}, "precision": 0.7473118279569892, "recall": 0.6318181818181818}}}, "2021.test (span detection)": {"micro/f1": 0.7826184343151529, "micro/f1_ci": {}, "micro/recall": 0.8081415519833468, "micro/precision": 0.7586581261535121, "macro/f1": 0.7826184343151529, "macro/f1_ci": {}, "macro/recall": 0.8081415519833468, "macro/precision": 0.7586581261535121}, "2020.test (span detection)": {"micro/f1": 0.7738478027867096, "micro/f1_ci": {}, "micro/recall": 0.749351323300467, "micro/precision": 0.8, "macro/f1": 0.7738478027867096, "macro/f1_ci": {}, "macro/recall": 0.749351323300467, "macro/precision": 0.8}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.659346545259775, "micro/f1_ci": {"90": [0.638530476656731, 0.6770002736609395], "95": [0.6355787571672109, 0.6814383304317009]}, "micro/recall": 0.6388168137000519, "micro/precision": 0.6812396236856668, "macro/f1": 0.6261309560026784, "macro/f1_ci": {"90": [0.6035614787399021, 0.6466837937863731], "95": [0.6008212526260872, 0.6498051149764593]}, "macro/recall": 0.6111694484964181, "macro/precision": 0.6527657911787169, "per_entity_metric": {"corporation": {"f1": 0.5966587112171838, "f1_ci": {"90": [0.5434307846076962, 0.6461646550171141], "95": [0.5326863869413766, 0.6552623249971564]}, "precision": 0.5482456140350878, "recall": 0.6544502617801047}, "creative_work": {"f1": 0.5892351274787537, "f1_ci": {"90": [0.5297221164826799, 0.6412325295653509], "95": [0.5167939021229471, 0.6498711140898077]}, "precision": 0.5977011494252874, "recall": 0.5810055865921788}, "event": {"f1": 0.4500907441016334, "f1_ci": {"90": [0.4, 0.5], "95": [0.3933869964988056, 0.5071014607812004]}, "precision": 0.43356643356643354, "recall": 0.4679245283018868}, "group": {"f1": 0.5454545454545454, "f1_ci": {"90": [0.4908707249478041, 0.5995649027168184], "95": [0.47806491874824353, 0.6098095911901321]}, "precision": 0.7076923076923077, "recall": 0.4437299035369775}, "location": {"f1": 0.6808510638297872, "f1_ci": {"90": [0.6158357771260997, 0.7378894784375705], "95": [0.6056659605996418, 0.7468373964681981]}, "precision": 0.6829268292682927, "recall": 0.6787878787878788}, "person": {"f1": 0.8358974358974359, "f1_ci": {"90": [0.8090079967608057, 0.8589565954118873], "95": [0.802661820512157, 0.864705203631851]}, "precision": 0.8519163763066202, "recall": 0.8204697986577181}, "product": {"f1": 0.6847290640394088, "f1_ci": {"90": [0.6328984758460346, 0.731190820093083], "95": [0.6213509150900847, 0.7403650506863269]}, "precision": 0.7473118279569892, "recall": 0.6318181818181818}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6401254269555967, "micro/f1_ci": {"90": [0.6315544728348781, 0.6491758274095626], "95": [0.6294268706225905, 0.6515448119225267]}, "micro/recall": 0.6609620721554117, "micro/precision": 0.6205623710780589, "macro/f1": 0.5947383155381057, "macro/f1_ci": {"90": [0.5852724504065837, 0.6043623226898465], "95": [0.5836410112124547, 0.6063768583335745]}, "macro/recall": 0.6206178838164583, "macro/precision": 0.5738855505495571, "per_entity_metric": {"corporation": {"f1": 0.5229357798165137, "f1_ci": {"90": [0.49706039551042225, 0.5489060654201801], "95": [0.49299356150347506, 0.5543603424996233]}, "precision": 0.4830508474576271, "recall": 0.57}, "creative_work": {"f1": 0.4629981024667932, "f1_ci": {"90": [0.43277573053403184, 0.4925610964986595], "95": [0.4272513399167463, 0.49761378932009187]}, "precision": 0.43058823529411766, "recall": 0.5006839945280438}, "event": {"f1": 0.4499572284003422, "f1_ci": {"90": [0.4274287289095753, 0.47154022208242086], "95": [0.4228081863105448, 0.47623473640595676]}, "precision": 0.4245359160613398, "recall": 0.4786169244767971}, "group": {"f1": 0.592749032030975, "f1_ci": {"90": [0.5722540581576642, 0.6147130977130977], "95": [0.5676533946830992, 0.6194597542880659]}, "precision": 0.636432350718065, "recall": 0.5546772068511199}, "location": {"f1": 0.6553030303030303, "f1_ci": {"90": [0.6304163397345088, 0.6817280494561573], "95": [0.6255962515548817, 0.6867043413492904]}, "precision": 0.597926267281106, "recall": 0.7248603351955307}, "person": {"f1": 0.8273135669362084, "f1_ci": {"90": [0.8160811561878993, 0.8380472862595445], "95": [0.814691063129817, 0.8397603550670991]}, "precision": 0.806869961444094, "recall": 0.8488200589970502}, "product": {"f1": 0.6519114688128772, "f1_ci": {"90": [0.6299024970087858, 0.6725193990424303], "95": [0.6268028691225062, 0.6764233561966475]}, "precision": 0.6377952755905512, "recall": 0.6666666666666666}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7738478027867096, "micro/f1_ci": {}, "micro/recall": 0.749351323300467, "micro/precision": 0.8, "macro/f1": 0.7738478027867096, "macro/f1_ci": {}, "macro/recall": 0.749351323300467, "macro/precision": 0.8}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7826184343151529, "micro/f1_ci": {}, "micro/recall": 0.8081415519833468, "micro/precision": 0.7586581261535121, "macro/f1": 0.7826184343151529, "macro/f1_ci": {}, "macro/recall": 0.8081415519833468, "macro/precision": 0.7586581261535121}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2020.train", "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": 0.3, "max_grad_norm": 1}
+{"dataset": ["tner/tweetner7"], "dataset_split": "train_2020", "dataset_name": null, "local_dataset": null, "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": 0.3, "max_grad_norm": 1}