model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bert-large-tweetner7-2020-2021-continuous
+  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: 0.6319818203564167
+    - name: Precision
+      type: precision
+      value: 0.6544463710676245
+    - name: Recall
+      type: recall
+      value: 0.6110083256244219
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5766988664971804
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.601237684920777
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5559244768648601
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7603780356501973
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7875108412836079
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7350526194055742
+  - 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: 0.6247533126585846
+    - name: Precision
+      type: precision
+      value: 0.6839506172839506
+    - name: Recall
+      type: recall
+      value: 0.5749870264660093
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.578717595313749
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6410778727928796
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5301549277792547
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7245559627854524
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7932098765432098
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.6668396471198754
+
+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-large-tweetner7-2020-2021-continuous
+
+This model is a fine-tuned version of [tner/bert-large-tweetner-2020](https://huggingface.co/tner/bert-large-tweetner-2020) on the 
+[tner/tweetner7](https://huggingface.co/datasets/tner/tweetner7) dataset (`train_2021` split). The model is first fine-tuned on `train_2020`, and then continuously fine-tuned on `train_2021`. 
+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): 0.6319818203564167
+- Precision (micro): 0.6544463710676245
+- Recall (micro): 0.6110083256244219
+- F1 (macro): 0.5766988664971804
+- Precision (macro): 0.601237684920777
+- Recall (macro): 0.5559244768648601
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.514024041213509
+- creative_work: 0.39736070381231675
+- event: 0.42546740778170794
+- group: 0.5859649122807017
+- location: 0.6335664335664336
+- person: 0.8127490039840638
+- product: 0.6677595628415302 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6231013705127983, 0.6413574593408826]
+    - 95%: [0.6217502353949177, 0.6428942705896876] 
+- F1 (macro): 
+    - 90%: [0.6231013705127983, 0.6413574593408826]
+    - 95%: [0.6217502353949177, 0.6428942705896876] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bert-large-tweetner7-2020-2021-continuous/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bert-large-tweetner7-2020-2021-continuous/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-large-tweetner7-2020-2021-continuous")
+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: tner/bert-large-tweetner-2020
+ - crf: True
+ - max_length: 128
+ - epoch: 30
+ - batch_size: 32
+ - lr: 1e-06
+ - 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/bert-large-tweetner7-2020-2021-continuous/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": 0.6307053941908715, "micro/f1_ci": {}, "micro/recall": 0.608, "micro/precision": 0.6551724137931034, "macro/f1": 0.5769830109938662, "macro/f1_ci": {}, "macro/recall": 0.5541914611675349, "macro/precision": 0.6061513565180056, "per_entity_metric": {"corporation": {"f1": 0.5797101449275363, "f1_ci": {}, "precision": 0.5714285714285714, "recall": 0.5882352941176471}, "creative_work": {"f1": 0.41428571428571426, "f1_ci": {}, "precision": 0.4393939393939394, "recall": 0.3918918918918919}, "event": {"f1": 0.34632034632034636, "f1_ci": {}, "precision": 0.4, "recall": 0.3053435114503817}, "group": {"f1": 0.6273584905660378, "f1_ci": {}, "precision": 0.6751269035532995, "recall": 0.5859030837004405}, "location": {"f1": 0.6176470588235293, "f1_ci": {}, "precision": 0.65625, "recall": 0.5833333333333334}, "person": {"f1": 0.8135593220338982, "f1_ci": {}, "precision": 0.7817589576547231, "recall": 0.8480565371024735}, "product": {"f1": 0.6400000000000001, "f1_ci": {}, "precision": 0.7191011235955056, "recall": 0.5765765765765766}}}, "2021.test": {"micro/f1": 0.6319818203564167, "micro/f1_ci": {"90": [0.6231013705127983, 0.6413574593408826], "95": [0.6217502353949177, 0.6428942705896876]}, "micro/recall": 0.6110083256244219, "micro/precision": 0.6544463710676245, "macro/f1": 0.5766988664971804, "macro/f1_ci": {"90": [0.5664509989359489, 0.5863507925127135], "95": [0.5648974825471297, 0.5879845963012866]}, "macro/recall": 0.5559244768648601, "macro/precision": 0.601237684920777, "per_entity_metric": {"corporation": {"f1": 0.514024041213509, "f1_ci": {"90": [0.48777635327635327, 0.5386428764634433], "95": [0.48256682507571314, 0.5427950687007129]}, "precision": 0.5301062573789846, "recall": 0.4988888888888889}, "creative_work": {"f1": 0.39736070381231675, "f1_ci": {"90": [0.3673069143565192, 0.4290928789350539], "95": [0.35964709226770636, 0.43496810207336517]}, "precision": 0.42812006319115326, "recall": 0.3707250341997264}, "event": {"f1": 0.42546740778170794, "f1_ci": {"90": [0.40080966868711565, 0.4504934716037546], "95": [0.3947971875592656, 0.45604206362953986]}, "precision": 0.4784090909090909, "recall": 0.3830755232029117}, "group": {"f1": 0.5859649122807017, "f1_ci": {"90": [0.5649910032383042, 0.608005869697348], "95": [0.5609718872903313, 0.6137378755305175]}, "precision": 0.6268768768768769, "recall": 0.5500658761528326}, "location": {"f1": 0.6335664335664336, "f1_ci": {"90": [0.6022469573815304, 0.662564426082152], "95": [0.5972619047619048, 0.6684591503346138]}, "precision": 0.634453781512605, "recall": 0.63268156424581}, "person": {"f1": 0.8127490039840638, "f1_ci": {"90": [0.8007926572866582, 0.8249959351621283], "95": [0.7984037478689154, 0.8268196857132452]}, "precision": 0.798576512455516, "recall": 0.827433628318584}, "product": {"f1": 0.6677595628415302, "f1_ci": {"90": [0.6464189850752727, 0.6878997822678681], "95": [0.6421587454017522, 0.6913517415533396]}, "precision": 0.7121212121212122, "recall": 0.6286008230452675}}}, "2020.test": {"micro/f1": 0.6247533126585846, "micro/f1_ci": {"90": [0.6028688973562135, 0.64588834464267], "95": [0.5994344280959615, 0.649232518476407]}, "micro/recall": 0.5749870264660093, "micro/precision": 0.6839506172839506, "macro/f1": 0.578717595313749, "macro/f1_ci": {"90": [0.5540098280773577, 0.6005537196979407], "95": [0.5512954123930293, 0.6060087044874898]}, "macro/recall": 0.5301549277792547, "macro/precision": 0.6410778727928796, "per_entity_metric": {"corporation": {"f1": 0.565597667638484, "f1_ci": {"90": [0.5050505050505051, 0.6224819178391817], "95": [0.4969315627024349, 0.6306885822510823]}, "precision": 0.6381578947368421, "recall": 0.5078534031413613}, "creative_work": {"f1": 0.4364820846905538, "f1_ci": {"90": [0.3723977546110665, 0.4983818770226537], "95": [0.36299049637405206, 0.515419097575336]}, "precision": 0.5234375, "recall": 0.3743016759776536}, "event": {"f1": 0.45508982035928147, "f1_ci": {"90": [0.4049049185589344, 0.5067778692762893], "95": [0.3923199130398428, 0.5195618366745284]}, "precision": 0.4830508474576271, "recall": 0.43018867924528303}, "group": {"f1": 0.5038167938931298, "f1_ci": {"90": [0.44796125194135145, 0.5607960998607093], "95": [0.4362921406545568, 0.5688220314081427]}, "precision": 0.6197183098591549, "recall": 0.42443729903536975}, "location": {"f1": 0.6026490066225166, "f1_ci": {"90": [0.5320404782483434, 0.6629593282072034], "95": [0.5245744513696118, 0.6756140811769343]}, "precision": 0.6642335766423357, "recall": 0.5515151515151515}, "person": {"f1": 0.8062283737024222, "f1_ci": {"90": [0.7760587632959272, 0.8315361561829818], "95": [0.7698106545550611, 0.8356172784193372]}, "precision": 0.8321428571428572, "recall": 0.7818791946308725}, "product": {"f1": 0.6811594202898551, "f1_ci": {"90": [0.6246480367439654, 0.730964467005076], "95": [0.6142940723633565, 0.7381622865912144]}, "precision": 0.7268041237113402, "recall": 0.6409090909090909}}}, "2021.test (span detection)": {"micro/f1": 0.7603780356501973, "micro/f1_ci": {}, "micro/recall": 0.7350526194055742, "micro/precision": 0.7875108412836079, "macro/f1": 0.7603780356501973, "macro/f1_ci": {}, "macro/recall": 0.7350526194055742, "macro/precision": 0.7875108412836079}, "2020.test (span detection)": {"micro/f1": 0.7245559627854524, "micro/f1_ci": {}, "micro/recall": 0.6668396471198754, "micro/precision": 0.7932098765432098, "macro/f1": 0.7245559627854524, "macro/f1_ci": {}, "macro/recall": 0.6668396471198754, "macro/precision": 0.7932098765432098}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6247533126585846, "micro/f1_ci": {"90": [0.6028688973562135, 0.64588834464267], "95": [0.5994344280959615, 0.649232518476407]}, "micro/recall": 0.5749870264660093, "micro/precision": 0.6839506172839506, "macro/f1": 0.578717595313749, "macro/f1_ci": {"90": [0.5540098280773577, 0.6005537196979407], "95": [0.5512954123930293, 0.6060087044874898]}, "macro/recall": 0.5301549277792547, "macro/precision": 0.6410778727928796, "per_entity_metric": {"corporation": {"f1": 0.565597667638484, "f1_ci": {"90": [0.5050505050505051, 0.6224819178391817], "95": [0.4969315627024349, 0.6306885822510823]}, "precision": 0.6381578947368421, "recall": 0.5078534031413613}, "creative_work": {"f1": 0.4364820846905538, "f1_ci": {"90": [0.3723977546110665, 0.4983818770226537], "95": [0.36299049637405206, 0.515419097575336]}, "precision": 0.5234375, "recall": 0.3743016759776536}, "event": {"f1": 0.45508982035928147, "f1_ci": {"90": [0.4049049185589344, 0.5067778692762893], "95": [0.3923199130398428, 0.5195618366745284]}, "precision": 0.4830508474576271, "recall": 0.43018867924528303}, "group": {"f1": 0.5038167938931298, "f1_ci": {"90": [0.44796125194135145, 0.5607960998607093], "95": [0.4362921406545568, 0.5688220314081427]}, "precision": 0.6197183098591549, "recall": 0.42443729903536975}, "location": {"f1": 0.6026490066225166, "f1_ci": {"90": [0.5320404782483434, 0.6629593282072034], "95": [0.5245744513696118, 0.6756140811769343]}, "precision": 0.6642335766423357, "recall": 0.5515151515151515}, "person": {"f1": 0.8062283737024222, "f1_ci": {"90": [0.7760587632959272, 0.8315361561829818], "95": [0.7698106545550611, 0.8356172784193372]}, "precision": 0.8321428571428572, "recall": 0.7818791946308725}, "product": {"f1": 0.6811594202898551, "f1_ci": {"90": [0.6246480367439654, 0.730964467005076], "95": [0.6142940723633565, 0.7381622865912144]}, "precision": 0.7268041237113402, "recall": 0.6409090909090909}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6319818203564167, "micro/f1_ci": {"90": [0.6231013705127983, 0.6413574593408826], "95": [0.6217502353949177, 0.6428942705896876]}, "micro/recall": 0.6110083256244219, "micro/precision": 0.6544463710676245, "macro/f1": 0.5766988664971804, "macro/f1_ci": {"90": [0.5664509989359489, 0.5863507925127135], "95": [0.5648974825471297, 0.5879845963012866]}, "macro/recall": 0.5559244768648601, "macro/precision": 0.601237684920777, "per_entity_metric": {"corporation": {"f1": 0.514024041213509, "f1_ci": {"90": [0.48777635327635327, 0.5386428764634433], "95": [0.48256682507571314, 0.5427950687007129]}, "precision": 0.5301062573789846, "recall": 0.4988888888888889}, "creative_work": {"f1": 0.39736070381231675, "f1_ci": {"90": [0.3673069143565192, 0.4290928789350539], "95": [0.35964709226770636, 0.43496810207336517]}, "precision": 0.42812006319115326, "recall": 0.3707250341997264}, "event": {"f1": 0.42546740778170794, "f1_ci": {"90": [0.40080966868711565, 0.4504934716037546], "95": [0.3947971875592656, 0.45604206362953986]}, "precision": 0.4784090909090909, "recall": 0.3830755232029117}, "group": {"f1": 0.5859649122807017, "f1_ci": {"90": [0.5649910032383042, 0.608005869697348], "95": [0.5609718872903313, 0.6137378755305175]}, "precision": 0.6268768768768769, "recall": 0.5500658761528326}, "location": {"f1": 0.6335664335664336, "f1_ci": {"90": [0.6022469573815304, 0.662564426082152], "95": [0.5972619047619048, 0.6684591503346138]}, "precision": 0.634453781512605, "recall": 0.63268156424581}, "person": {"f1": 0.8127490039840638, "f1_ci": {"90": [0.8007926572866582, 0.8249959351621283], "95": [0.7984037478689154, 0.8268196857132452]}, "precision": 0.798576512455516, "recall": 0.827433628318584}, "product": {"f1": 0.6677595628415302, "f1_ci": {"90": [0.6464189850752727, 0.6878997822678681], "95": [0.6421587454017522, 0.6913517415533396]}, "precision": 0.7121212121212122, "recall": 0.6286008230452675}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7245559627854524, "micro/f1_ci": {}, "micro/recall": 0.6668396471198754, "micro/precision": 0.7932098765432098, "macro/f1": 0.7245559627854524, "macro/f1_ci": {}, "macro/recall": 0.6668396471198754, "macro/precision": 0.7932098765432098}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7603780356501973, "micro/f1_ci": {}, "micro/recall": 0.7350526194055742, "micro/precision": 0.7875108412836079, "macro/f1": 0.7603780356501973, "macro/f1_ci": {}, "macro/recall": 0.7350526194055742, "macro/precision": 0.7875108412836079}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2021.train", "model": "tner/bert-large-tweetner-2020", "crf": true, "max_length": 128, "epoch": 30, "batch_size": 32, "lr": 1e-06, "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_2021", "dataset_name": null, "local_dataset": null, "model": "tner/bert-large-tweetner-2020", "crf": true, "max_length": 128, "epoch": 30, "batch_size": 32, "lr": 1e-06, "random_seed": 0, "gradient_accumulation_steps": 1, "weight_decay": 1e-07, "lr_warmup_step_ratio": 0.3, "max_grad_norm": 1}