model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bertweet-base-tweetner7-2020-2021-concat
+  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: 0.6536203522504892
+    - name: Precision
+      type: precision
+      value: 0.6327812060192703
+    - name: Recall
+      type: recall
+      value: 0.6758788159111934
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.6052211252463111
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5838227039402247
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6302754427289782
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7898680384701409
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7646421998484356
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.8168150803746964
+  - 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: 0.6574172892209178
+    - name: Precision
+      type: precision
+      value: 0.6765513454146074
+    - name: Recall
+      type: recall
+      value: 0.6393357550596782
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.6161494551388561
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6335227896210995
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6030680287240185
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7691486522551374
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7917582417582417
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7477944992215879
+
+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-base-tweetner7-2020-2021-concat
+
+This model is a fine-tuned version of [vinai/bertweet-base](https://huggingface.co/vinai/bertweet-base) 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): 0.6536203522504892
+- Precision (micro): 0.6327812060192703
+- Recall (micro): 0.6758788159111934
+- F1 (macro): 0.6052211252463111
+- Precision (macro): 0.5838227039402247
+- Recall (macro): 0.6302754427289782
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.5250836120401337
+- creative_work: 0.4653774173424829
+- event: 0.4805781391147245
+- group: 0.6033376123234916
+- location: 0.6567164179104478
+- person: 0.8408236347358997
+- product: 0.6646310432569975 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6447872756148977, 0.6633207283107695]
+    - 95%: [0.6425923702362265, 0.6650666703489687] 
+- F1 (macro): 
+    - 90%: [0.6447872756148977, 0.6633207283107695]
+    - 95%: [0.6425923702362265, 0.6650666703489687] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bertweet-base-tweetner7-2020-2021-concat/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bertweet-base-tweetner7-2020-2021-concat/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-base-tweetner7-2020-2021-concat")
+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_all
+ - dataset_name: None
+ - local_dataset: None
+ - model: vinai/bertweet-base
+ - 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-base-tweetner7-2020-2021-concat/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 @@
-{"2021.dev": {"micro/f1": 0.6479289940828402, "micro/f1_ci": {}, "micro/recall": 0.657, "micro/precision": 0.6391050583657587, "macro/f1": 0.6043271222909856, "macro/f1_ci": {}, "macro/recall": 0.6132417907641617, "macro/precision": 0.5968927910350338, "per_entity_metric": {"corporation": {"f1": 0.5603864734299516, "f1_ci": {}, "precision": 0.5523809523809524, "recall": 0.5686274509803921}, "creative_work": {"f1": 0.48407643312101906, "f1_ci": {}, "precision": 0.4578313253012048, "recall": 0.5135135135135135}, "event": {"f1": 0.4528301886792453, "f1_ci": {}, "precision": 0.44776119402985076, "recall": 0.4580152671755725}, "group": {"f1": 0.6269315673289183, "f1_ci": {}, "precision": 0.6283185840707964, "recall": 0.6255506607929515}, "location": {"f1": 0.6442953020134228, "f1_ci": {}, "precision": 0.6233766233766234, "recall": 0.6666666666666666}, "person": {"f1": 0.8361774744027305, "f1_ci": {}, "precision": 0.8085808580858086, "recall": 0.8657243816254417}, "product": {"f1": 0.6255924170616114, "f1_ci": {}, "precision": 0.66, "recall": 0.5945945945945946}}}, "2021.test": {"micro/f1": 0.6536203522504892, "micro/f1_ci": {"90": [0.6447872756148977, 0.6633207283107695], "95": [0.6425923702362265, 0.6650666703489687]}, "micro/recall": 0.6758788159111934, "micro/precision": 0.6327812060192703, "macro/f1": 0.6052211252463111, "macro/f1_ci": {"90": [0.5954577789975593, 0.6145053233650972], "95": [0.5935811679603806, 0.6159321952528645]}, "macro/recall": 0.6302754427289782, "macro/precision": 0.5838227039402247, "per_entity_metric": {"corporation": {"f1": 0.5250836120401337, "f1_ci": {"90": [0.499114468582439, 0.5502487923745039], "95": [0.49402204164639574, 0.5542265421428756]}, "precision": 0.5268456375838926, "recall": 0.5233333333333333}, "creative_work": {"f1": 0.4653774173424829, "f1_ci": {"90": [0.4350066554677983, 0.4944221755636148], "95": [0.42865964754879954, 0.5003082860885536]}, "precision": 0.42775229357798167, "recall": 0.5102599179206566}, "event": {"f1": 0.4805781391147245, "f1_ci": {"90": [0.45876392113230563, 0.5025502448910658], "95": [0.4547767349386974, 0.5085979655019902]}, "precision": 0.47713004484304933, "recall": 0.4840764331210191}, "group": {"f1": 0.6033376123234916, "f1_ci": {"90": [0.5833319019809917, 0.6255975435188365], "95": [0.5807051025464135, 0.6290415081557142]}, "precision": 0.5882352941176471, "recall": 0.619235836627141}, "location": {"f1": 0.6567164179104478, "f1_ci": {"90": [0.6289343271359739, 0.6827299264868976], "95": [0.6232678042111391, 0.6866568408235075]}, "precision": 0.5919282511210763, "recall": 0.7374301675977654}, "person": {"f1": 0.8408236347358997, "f1_ci": {"90": [0.8306288526403558, 0.8509253313521083], "95": [0.8287685929940346, 0.8526699869564047]}, "precision": 0.8172641837800209, "recall": 0.8657817109144543}, "product": {"f1": 0.6646310432569975, "f1_ci": {"90": [0.642891431340614, 0.6874707012510687], "95": [0.6397193439875759, 0.6905268361734545]}, "precision": 0.6576032225579054, "recall": 0.6718106995884774}}}, "2020.test": {"micro/f1": 0.6574172892209178, "micro/f1_ci": {"90": [0.6362605912738951, 0.6776829185560604], "95": [0.6329381337554338, 0.681496695257187]}, "micro/recall": 0.6393357550596782, "micro/precision": 0.6765513454146074, "macro/f1": 0.6161494551388561, "macro/f1_ci": {"90": [0.5922560215720907, 0.6371588233814224], "95": [0.5874972325138178, 0.6433530990576218]}, "macro/recall": 0.6030680287240185, "macro/precision": 0.6335227896210995, "per_entity_metric": {"corporation": {"f1": 0.5721925133689839, "f1_ci": {"90": [0.5118543960209868, 0.627970887027902], "95": [0.5013, 0.6377742851758119]}, "precision": 0.5846994535519126, "recall": 0.5602094240837696}, "creative_work": {"f1": 0.5409836065573771, "f1_ci": {"90": [0.4857961799177999, 0.593859943977591], "95": [0.4738351063829787, 0.6035510525280346]}, "precision": 0.5294117647058824, "recall": 0.553072625698324}, "event": {"f1": 0.48554913294797686, "f1_ci": {"90": [0.43302563132507327, 0.5353292120908344], "95": [0.42145597637198395, 0.5447158512340591]}, "precision": 0.49606299212598426, "recall": 0.47547169811320755}, "group": {"f1": 0.5734767025089605, "f1_ci": {"90": [0.519707990723171, 0.6271215968818284], "95": [0.5103473314135293, 0.6346256684491981]}, "precision": 0.6477732793522267, "recall": 0.5144694533762058}, "location": {"f1": 0.6457142857142857, "f1_ci": {"90": [0.5746137178098175, 0.7090515682906987], "95": [0.5604067645282423, 0.7207259585166563]}, "precision": 0.6108108108108108, "recall": 0.6848484848484848}, "person": {"f1": 0.8415672913117546, "f1_ci": {"90": [0.8160757025086428, 0.8660040531276134], "95": [0.8116914027938739, 0.8688960711564506]}, "precision": 0.8546712802768166, "recall": 0.8288590604026845}, "product": {"f1": 0.6535626535626535, "f1_ci": {"90": [0.6015352256997502, 0.7024343440641215], "95": [0.5921931733188277, 0.711786009526434]}, "precision": 0.7112299465240641, "recall": 0.6045454545454545}}}, "2021.test (span detection)": {"micro/f1": 0.7898680384701409, "micro/f1_ci": {}, "micro/recall": 0.8168150803746964, "micro/precision": 0.7646421998484356, "macro/f1": 0.7898680384701409, "macro/f1_ci": {}, "macro/recall": 0.8168150803746964, "macro/precision": 0.7646421998484356}, "2020.test (span detection)": {"micro/f1": 0.7691486522551374, "micro/f1_ci": {}, "micro/recall": 0.7477944992215879, "micro/precision": 0.7917582417582417, "macro/f1": 0.7691486522551374, "macro/f1_ci": {}, "macro/recall": 0.7477944992215879, "macro/precision": 0.7917582417582417}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6574172892209178, "micro/f1_ci": {"90": [0.6362605912738951, 0.6776829185560604], "95": [0.6329381337554338, 0.681496695257187]}, "micro/recall": 0.6393357550596782, "micro/precision": 0.6765513454146074, "macro/f1": 0.6161494551388561, "macro/f1_ci": {"90": [0.5922560215720907, 0.6371588233814224], "95": [0.5874972325138178, 0.6433530990576218]}, "macro/recall": 0.6030680287240185, "macro/precision": 0.6335227896210995, "per_entity_metric": {"corporation": {"f1": 0.5721925133689839, "f1_ci": {"90": [0.5118543960209868, 0.627970887027902], "95": [0.5013, 0.6377742851758119]}, "precision": 0.5846994535519126, "recall": 0.5602094240837696}, "creative_work": {"f1": 0.5409836065573771, "f1_ci": {"90": [0.4857961799177999, 0.593859943977591], "95": [0.4738351063829787, 0.6035510525280346]}, "precision": 0.5294117647058824, "recall": 0.553072625698324}, "event": {"f1": 0.48554913294797686, "f1_ci": {"90": [0.43302563132507327, 0.5353292120908344], "95": [0.42145597637198395, 0.5447158512340591]}, "precision": 0.49606299212598426, "recall": 0.47547169811320755}, "group": {"f1": 0.5734767025089605, "f1_ci": {"90": [0.519707990723171, 0.6271215968818284], "95": [0.5103473314135293, 0.6346256684491981]}, "precision": 0.6477732793522267, "recall": 0.5144694533762058}, "location": {"f1": 0.6457142857142857, "f1_ci": {"90": [0.5746137178098175, 0.7090515682906987], "95": [0.5604067645282423, 0.7207259585166563]}, "precision": 0.6108108108108108, "recall": 0.6848484848484848}, "person": {"f1": 0.8415672913117546, "f1_ci": {"90": [0.8160757025086428, 0.8660040531276134], "95": [0.8116914027938739, 0.8688960711564506]}, "precision": 0.8546712802768166, "recall": 0.8288590604026845}, "product": {"f1": 0.6535626535626535, "f1_ci": {"90": [0.6015352256997502, 0.7024343440641215], "95": [0.5921931733188277, 0.711786009526434]}, "precision": 0.7112299465240641, "recall": 0.6045454545454545}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6536203522504892, "micro/f1_ci": {"90": [0.6447872756148977, 0.6633207283107695], "95": [0.6425923702362265, 0.6650666703489687]}, "micro/recall": 0.6758788159111934, "micro/precision": 0.6327812060192703, "macro/f1": 0.6052211252463111, "macro/f1_ci": {"90": [0.5954577789975593, 0.6145053233650972], "95": [0.5935811679603806, 0.6159321952528645]}, "macro/recall": 0.6302754427289782, "macro/precision": 0.5838227039402247, "per_entity_metric": {"corporation": {"f1": 0.5250836120401337, "f1_ci": {"90": [0.499114468582439, 0.5502487923745039], "95": [0.49402204164639574, 0.5542265421428756]}, "precision": 0.5268456375838926, "recall": 0.5233333333333333}, "creative_work": {"f1": 0.4653774173424829, "f1_ci": {"90": [0.4350066554677983, 0.4944221755636148], "95": [0.42865964754879954, 0.5003082860885536]}, "precision": 0.42775229357798167, "recall": 0.5102599179206566}, "event": {"f1": 0.4805781391147245, "f1_ci": {"90": [0.45876392113230563, 0.5025502448910658], "95": [0.4547767349386974, 0.5085979655019902]}, "precision": 0.47713004484304933, "recall": 0.4840764331210191}, "group": {"f1": 0.6033376123234916, "f1_ci": {"90": [0.5833319019809917, 0.6255975435188365], "95": [0.5807051025464135, 0.6290415081557142]}, "precision": 0.5882352941176471, "recall": 0.619235836627141}, "location": {"f1": 0.6567164179104478, "f1_ci": {"90": [0.6289343271359739, 0.6827299264868976], "95": [0.6232678042111391, 0.6866568408235075]}, "precision": 0.5919282511210763, "recall": 0.7374301675977654}, "person": {"f1": 0.8408236347358997, "f1_ci": {"90": [0.8306288526403558, 0.8509253313521083], "95": [0.8287685929940346, 0.8526699869564047]}, "precision": 0.8172641837800209, "recall": 0.8657817109144543}, "product": {"f1": 0.6646310432569975, "f1_ci": {"90": [0.642891431340614, 0.6874707012510687], "95": [0.6397193439875759, 0.6905268361734545]}, "precision": 0.6576032225579054, "recall": 0.6718106995884774}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7691486522551374, "micro/f1_ci": {}, "micro/recall": 0.7477944992215879, "micro/precision": 0.7917582417582417, "macro/f1": 0.7691486522551374, "macro/f1_ci": {}, "macro/recall": 0.7477944992215879, "macro/precision": 0.7917582417582417}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7898680384701409, "micro/f1_ci": {}, "micro/recall": 0.8168150803746964, "micro/precision": 0.7646421998484356, "macro/f1": 0.7898680384701409, "macro/f1_ci": {}, "macro/recall": 0.8168150803746964, "macro/precision": 0.7646421998484356}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2020_2021.train", "model": "vinai/bertweet-base", "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}
+{"dataset": ["tner/tweetner7"], "dataset_split": "train_all", "dataset_name": null, "local_dataset": null, "model": "vinai/bertweet-base", "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}