model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bert-base-tweetner7-2021
+  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: 0.6067163296677308
+    - name: Precision
+      type: precision
+      value: 0.6207355431889668
+    - name: Recall
+      type: recall
+      value: 0.5933163737280296
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5550005793330179
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5693376167794506
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.541740961323845
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7572011592831372
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7749394673123486
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7402567364403839
+  - 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: 0.5844643372798152
+    - name: Precision
+      type: precision
+      value: 0.6588541666666666
+    - name: Recall
+      type: recall
+      value: 0.5251686559418786
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5421676095032785
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6164636587810847
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.4850420088484678
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7086341322552699
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.798828125
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.6367410482615464
+
+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/bert-base-tweetner7-2021
+
+This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the 
+[tner/tweetner7](https://huggingface.co/datasets/tner/tweetner7) dataset (`train_2021` 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): 0.6067163296677308
+- Precision (micro): 0.6207355431889668
+- Recall (micro): 0.5933163737280296
+- F1 (macro): 0.5550005793330179
+- Precision (macro): 0.5693376167794506
+- Recall (macro): 0.541740961323845
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.46803779877709834
+- creative_work: 0.35353535353535354
+- event: 0.4128014842300557
+- group: 0.5622895622895623
+- location: 0.6477675407512402
+- person: 0.7988785046728971
+- product: 0.6416938110749185 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.5975843576319135, 0.6165630502101772]
+    - 95%: [0.5960112385452907, 0.617571609894858] 
+- F1 (macro): 
+    - 90%: [0.5975843576319135, 0.6165630502101772]
+    - 95%: [0.5960112385452907, 0.617571609894858] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bert-base-tweetner7-2021/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bert-base-tweetner7-2021/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/bert-base-tweetner7-2021")
+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_2021
+ - dataset_name: None
+ - local_dataset: None
+ - model: bert-base-cased
+ - crf: True
+ - max_length: 128
+ - epoch: 30
+ - batch_size: 32
+ - lr: 0.0001
+ - 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/bert-base-tweetner7-2021/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": 0.6017607457276023, "micro/f1_ci": {}, "micro/recall": 0.581, "micro/precision": 0.6240601503759399, "macro/f1": 0.5555350315685348, "macro/f1_ci": {}, "macro/recall": 0.533274675470052, "macro/precision": 0.5804101426861542, "per_entity_metric": {"corporation": {"f1": 0.5102040816326531, "f1_ci": {}, "precision": 0.5319148936170213, "recall": 0.49019607843137253}, "creative_work": {"f1": 0.39999999999999997, "f1_ci": {}, "precision": 0.4426229508196721, "recall": 0.36486486486486486}, "event": {"f1": 0.375, "f1_ci": {}, "precision": 0.384, "recall": 0.366412213740458}, "group": {"f1": 0.6009174311926606, "f1_ci": {}, "precision": 0.6267942583732058, "recall": 0.5770925110132159}, "location": {"f1": 0.6029411764705882, "f1_ci": {}, "precision": 0.640625, "recall": 0.5694444444444444}, "person": {"f1": 0.7799642218246868, "f1_ci": {}, "precision": 0.7898550724637681, "recall": 0.7703180212014135}, "product": {"f1": 0.619718309859155, "f1_ci": {}, "precision": 0.6470588235294118, "recall": 0.5945945945945946}}}, "2021.test": {"micro/f1": 0.6067163296677308, "micro/f1_ci": {"90": [0.5975843576319135, 0.6165630502101772], "95": [0.5960112385452907, 0.617571609894858]}, "micro/recall": 0.5933163737280296, "micro/precision": 0.6207355431889668, "macro/f1": 0.5550005793330179, "macro/f1_ci": {"90": [0.5454776939652932, 0.5646936919721942], "95": [0.5432999131038413, 0.5663329297982065]}, "macro/recall": 0.541740961323845, "macro/precision": 0.5693376167794506, "per_entity_metric": {"corporation": {"f1": 0.46803779877709834, "f1_ci": {"90": [0.4436611083218358, 0.49340886719357124], "95": [0.4403605081922379, 0.49785520431348507]}, "precision": 0.4682981090100111, "recall": 0.4677777777777778}, "creative_work": {"f1": 0.35353535353535354, "f1_ci": {"90": [0.32452379690481237, 0.38226572468955217], "95": [0.31762495778453226, 0.38916201657904076]}, "precision": 0.37404580152671757, "recall": 0.33515731874145005}, "event": {"f1": 0.4128014842300557, "f1_ci": {"90": [0.3895090774496654, 0.43329822290179854], "95": [0.3855879213054751, 0.43708841260659903]}, "precision": 0.4210028382213813, "recall": 0.4049135577797998}, "group": {"f1": 0.5622895622895623, "f1_ci": {"90": [0.5412921294490409, 0.58370523185127], "95": [0.5357699488765177, 0.5881970460741816]}, "precision": 0.5750688705234159, "recall": 0.5500658761528326}, "location": {"f1": 0.6477675407512402, "f1_ci": {"90": [0.6194087633767223, 0.6769953899395165], "95": [0.6122619065457905, 0.6810659039168759]}, "precision": 0.6575539568345323, "recall": 0.638268156424581}, "person": {"f1": 0.7988785046728971, "f1_ci": {"90": [0.7872775543579533, 0.8102531748771992], "95": [0.7851677872261382, 0.8117485932784152]}, "precision": 0.8100833965125095, "recall": 0.7879793510324484}, "product": {"f1": 0.6416938110749185, "f1_ci": {"90": [0.6186304045603652, 0.6652321201925173], "95": [0.6149650845890683, 0.6692824465619494]}, "precision": 0.6793103448275862, "recall": 0.6080246913580247}}}, "2020.test": {"micro/f1": 0.5844643372798152, "micro/f1_ci": {"90": [0.563635081420695, 0.6049707267950686], "95": [0.5599700870857128, 0.6090208008473901]}, "micro/recall": 0.5251686559418786, "micro/precision": 0.6588541666666666, "macro/f1": 0.5421676095032785, "macro/f1_ci": {"90": [0.5207339779479219, 0.5627309034099124], "95": [0.5164049065912122, 0.566586222357101]}, "macro/recall": 0.4850420088484678, "macro/precision": 0.6164636587810847, "per_entity_metric": {"corporation": {"f1": 0.48837209302325585, "f1_ci": {"90": [0.4331173898052879, 0.5406831776841848], "95": [0.4240425874275906, 0.5492604072290306]}, "precision": 0.5490196078431373, "recall": 0.4397905759162304}, "creative_work": {"f1": 0.4304635761589404, "f1_ci": {"90": [0.3727598566308244, 0.48666666666666664], "95": [0.3650740025740025, 0.4952410437859355]}, "precision": 0.5284552845528455, "recall": 0.36312849162011174}, "event": {"f1": 0.3227176220806794, "f1_ci": {"90": [0.26938415101139795, 0.3763103665250905], "95": [0.2586177574478067, 0.38613812128642844]}, "precision": 0.36893203883495146, "recall": 0.28679245283018867}, "group": {"f1": 0.5065176908752327, "f1_ci": {"90": [0.4571428571428572, 0.5595023477009111], "95": [0.4444005270092226, 0.5682343778632948]}, "precision": 0.6017699115044248, "recall": 0.43729903536977494}, "location": {"f1": 0.6153846153846153, "f1_ci": {"90": [0.5421016483516483, 0.6753693298105905], "95": [0.5293878527020563, 0.6838247153700189]}, "precision": 0.6865671641791045, "recall": 0.5575757575757576}, "person": {"f1": 0.7667560321715817, "f1_ci": {"90": [0.7357200782476069, 0.7935042398953037], "95": [0.7289370539370538, 0.7984247819667724]}, "precision": 0.8202676864244742, "recall": 0.7197986577181208}, "product": {"f1": 0.6649616368286445, "f1_ci": {"90": [0.6082217918778835, 0.7151641743808846], "95": [0.5989697210174176, 0.7246467391304348]}, "precision": 0.7602339181286549, "recall": 0.5909090909090909}}}, "2021.test (span detection)": {"micro/f1": 0.7572011592831372, "micro/f1_ci": {}, "micro/recall": 0.7402567364403839, "micro/precision": 0.7749394673123486, "macro/f1": 0.7572011592831372, "macro/f1_ci": {}, "macro/recall": 0.7402567364403839, "macro/precision": 0.7749394673123486}, "2020.test (span detection)": {"micro/f1": 0.7086341322552699, "micro/f1_ci": {}, "micro/recall": 0.6367410482615464, "micro/precision": 0.798828125, "macro/f1": 0.7086341322552699, "macro/f1_ci": {}, "macro/recall": 0.6367410482615464, "macro/precision": 0.798828125}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.5844643372798152, "micro/f1_ci": {"90": [0.563635081420695, 0.6049707267950686], "95": [0.5599700870857128, 0.6090208008473901]}, "micro/recall": 0.5251686559418786, "micro/precision": 0.6588541666666666, "macro/f1": 0.5421676095032785, "macro/f1_ci": {"90": [0.5207339779479219, 0.5627309034099124], "95": [0.5164049065912122, 0.566586222357101]}, "macro/recall": 0.4850420088484678, "macro/precision": 0.6164636587810847, "per_entity_metric": {"corporation": {"f1": 0.48837209302325585, "f1_ci": {"90": [0.4331173898052879, 0.5406831776841848], "95": [0.4240425874275906, 0.5492604072290306]}, "precision": 0.5490196078431373, "recall": 0.4397905759162304}, "creative_work": {"f1": 0.4304635761589404, "f1_ci": {"90": [0.3727598566308244, 0.48666666666666664], "95": [0.3650740025740025, 0.4952410437859355]}, "precision": 0.5284552845528455, "recall": 0.36312849162011174}, "event": {"f1": 0.3227176220806794, "f1_ci": {"90": [0.26938415101139795, 0.3763103665250905], "95": [0.2586177574478067, 0.38613812128642844]}, "precision": 0.36893203883495146, "recall": 0.28679245283018867}, "group": {"f1": 0.5065176908752327, "f1_ci": {"90": [0.4571428571428572, 0.5595023477009111], "95": [0.4444005270092226, 0.5682343778632948]}, "precision": 0.6017699115044248, "recall": 0.43729903536977494}, "location": {"f1": 0.6153846153846153, "f1_ci": {"90": [0.5421016483516483, 0.6753693298105905], "95": [0.5293878527020563, 0.6838247153700189]}, "precision": 0.6865671641791045, "recall": 0.5575757575757576}, "person": {"f1": 0.7667560321715817, "f1_ci": {"90": [0.7357200782476069, 0.7935042398953037], "95": [0.7289370539370538, 0.7984247819667724]}, "precision": 0.8202676864244742, "recall": 0.7197986577181208}, "product": {"f1": 0.6649616368286445, "f1_ci": {"90": [0.6082217918778835, 0.7151641743808846], "95": [0.5989697210174176, 0.7246467391304348]}, "precision": 0.7602339181286549, "recall": 0.5909090909090909}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6067163296677308, "micro/f1_ci": {"90": [0.5975843576319135, 0.6165630502101772], "95": [0.5960112385452907, 0.617571609894858]}, "micro/recall": 0.5933163737280296, "micro/precision": 0.6207355431889668, "macro/f1": 0.5550005793330179, "macro/f1_ci": {"90": [0.5454776939652932, 0.5646936919721942], "95": [0.5432999131038413, 0.5663329297982065]}, "macro/recall": 0.541740961323845, "macro/precision": 0.5693376167794506, "per_entity_metric": {"corporation": {"f1": 0.46803779877709834, "f1_ci": {"90": [0.4436611083218358, 0.49340886719357124], "95": [0.4403605081922379, 0.49785520431348507]}, "precision": 0.4682981090100111, "recall": 0.4677777777777778}, "creative_work": {"f1": 0.35353535353535354, "f1_ci": {"90": [0.32452379690481237, 0.38226572468955217], "95": [0.31762495778453226, 0.38916201657904076]}, "precision": 0.37404580152671757, "recall": 0.33515731874145005}, "event": {"f1": 0.4128014842300557, "f1_ci": {"90": [0.3895090774496654, 0.43329822290179854], "95": [0.3855879213054751, 0.43708841260659903]}, "precision": 0.4210028382213813, "recall": 0.4049135577797998}, "group": {"f1": 0.5622895622895623, "f1_ci": {"90": [0.5412921294490409, 0.58370523185127], "95": [0.5357699488765177, 0.5881970460741816]}, "precision": 0.5750688705234159, "recall": 0.5500658761528326}, "location": {"f1": 0.6477675407512402, "f1_ci": {"90": [0.6194087633767223, 0.6769953899395165], "95": [0.6122619065457905, 0.6810659039168759]}, "precision": 0.6575539568345323, "recall": 0.638268156424581}, "person": {"f1": 0.7988785046728971, "f1_ci": {"90": [0.7872775543579533, 0.8102531748771992], "95": [0.7851677872261382, 0.8117485932784152]}, "precision": 0.8100833965125095, "recall": 0.7879793510324484}, "product": {"f1": 0.6416938110749185, "f1_ci": {"90": [0.6186304045603652, 0.6652321201925173], "95": [0.6149650845890683, 0.6692824465619494]}, "precision": 0.6793103448275862, "recall": 0.6080246913580247}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7086341322552699, "micro/f1_ci": {}, "micro/recall": 0.6367410482615464, "micro/precision": 0.798828125, "macro/f1": 0.7086341322552699, "macro/f1_ci": {}, "macro/recall": 0.6367410482615464, "macro/precision": 0.798828125}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7572011592831372, "micro/f1_ci": {}, "micro/recall": 0.7402567364403839, "micro/precision": 0.7749394673123486, "macro/f1": 0.7572011592831372, "macro/f1_ci": {}, "macro/recall": 0.7402567364403839, "macro/precision": 0.7749394673123486}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2021.train", "model": "bert-base-cased", "crf": true, "max_length": 128, "epoch": 30, "batch_size": 32, "lr": 0.0001, "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_2021", "dataset_name": null, "local_dataset": null, "model": "bert-base-cased", "crf": true, "max_length": 128, "epoch": 30, "batch_size": 32, "lr": 0.0001, "random_seed": 0, "gradient_accumulation_steps": 1, "weight_decay": 1e-07, "lr_warmup_step_ratio": 0.15, "max_grad_norm": 1}