model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bertweet-large-tweetner7-random
+  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: 0.6486182247987844
+    - name: Precision
+      type: precision
+      value: 0.6318675293343569
+    - name: Recall
+      type: recall
+      value: 0.6662812210915818
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.604868641257225
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.589818811310092
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6244372176840122
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7843071034560397
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7640092115363527
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.8057129640337689
+  - 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: 0.6602409638554216
+    - name: Precision
+      type: precision
+      value: 0.6819690265486725
+    - name: Recall
+      type: recall
+      value: 0.6398546964193046
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.6271520713994495
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.652241965435762
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6095524645700295
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7720332172515403
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7978959025470653
+    - 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-large-tweetner7-random
+
+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_random` 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): 0.6486182247987844
+- Precision (micro): 0.6318675293343569
+- Recall (micro): 0.6662812210915818
+- F1 (macro): 0.604868641257225
+- Precision (macro): 0.589818811310092
+- Recall (macro): 0.6244372176840122
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.5359342915811088
+- creative_work: 0.454661558109834
+- event: 0.46186621218576907
+- group: 0.6163606010016696
+- location: 0.6615873015873016
+- person: 0.8278614184654453
+- product: 0.675809105869446 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6400171924527076, 0.6574431063551344]
+    - 95%: [0.6382155801831687, 0.6592086893227054] 
+- F1 (macro): 
+    - 90%: [0.6400171924527076, 0.6574431063551344]
+    - 95%: [0.6382155801831687, 0.6592086893227054] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bertweet-large-tweetner7-random/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bertweet-large-tweetner7-random/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-random")
+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_random
+ - 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-random/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 @@
-{"random.dev": {"micro/f1": 0.6406879871002419, "micro/f1_ci": {}, "micro/recall": 0.6364121729845168, "micro/precision": 0.645021645021645, "macro/f1": 0.5987771758365046, "macro/f1_ci": {}, "macro/recall": 0.5936204855874079, "macro/precision": 0.6087739435250593, "per_entity_metric": {"corporation": {"f1": 0.5626598465473146, "f1_ci": {}, "precision": 0.5555555555555556, "recall": 0.5699481865284974}, "creative_work": {"f1": 0.5214723926380369, "f1_ci": {}, "precision": 0.5182926829268293, "recall": 0.5246913580246914}, "event": {"f1": 0.3688212927756654, "f1_ci": {}, "precision": 0.34519572953736655, "recall": 0.39591836734693875}, "group": {"f1": 0.6137724550898204, "f1_ci": {}, "precision": 0.6366459627329193, "recall": 0.5924855491329479}, "location": {"f1": 0.6449704142011835, "f1_ci": {}, "precision": 0.6228571428571429, "recall": 0.6687116564417178}, "person": {"f1": 0.8643592142188962, "f1_ci": {}, "precision": 0.8619402985074627, "recall": 0.8667917448405253}, "product": {"f1": 0.6153846153846153, "f1_ci": {}, "precision": 0.7209302325581395, "recall": 0.5367965367965368}}}, "2021.test": {"micro/f1": 0.6486182247987844, "micro/f1_ci": {"90": [0.6400171924527076, 0.6574431063551344], "95": [0.6382155801831687, 0.6592086893227054]}, "micro/recall": 0.6662812210915818, "micro/precision": 0.6318675293343569, "macro/f1": 0.604868641257225, "macro/f1_ci": {"90": [0.595296924609441, 0.614397231507266], "95": [0.5933778702735921, 0.6163061595932199]}, "macro/recall": 0.6244372176840122, "macro/precision": 0.589818811310092, "per_entity_metric": {"corporation": {"f1": 0.5359342915811088, "f1_ci": {"90": [0.5124250180674441, 0.562196264340956], "95": [0.5066242713301538, 0.5660601358658982]}, "precision": 0.49809160305343514, "recall": 0.58}, "creative_work": {"f1": 0.454661558109834, "f1_ci": {"90": [0.42294214039784905, 0.4866248288332962], "95": [0.4174022952917621, 0.49281072809973214]}, "precision": 0.42634730538922155, "recall": 0.48700410396716826}, "event": {"f1": 0.46186621218576907, "f1_ci": {"90": [0.4407249060721636, 0.48402759846047894], "95": [0.43625117639460426, 0.4896707137226466]}, "precision": 0.4342948717948718, "recall": 0.4931756141947225}, "group": {"f1": 0.6163606010016696, "f1_ci": {"90": [0.5951285544493952, 0.6383988899613899], "95": [0.5921567684006442, 0.6431355549407772]}, "precision": 0.6249153689911984, "recall": 0.6080368906455863}, "location": {"f1": 0.6615873015873016, "f1_ci": {"90": [0.6349800972344961, 0.6883097309494146], "95": [0.6280602263375301, 0.6938339172840142]}, "precision": 0.6065192083818394, "recall": 0.7276536312849162}, "person": {"f1": 0.8278614184654453, "f1_ci": {"90": [0.8174552782574815, 0.8383023554114327], "95": [0.8153046175852993, 0.840402111046631]}, "precision": 0.814709032488397, "recall": 0.8414454277286135}, "product": {"f1": 0.675809105869446, "f1_ci": {"90": [0.6533841788755885, 0.6975096627158484], "95": [0.648618174354878, 0.7003805018510901]}, "precision": 0.7238542890716804, "recall": 0.6337448559670782}}}, "2020.test": {"micro/f1": 0.6602409638554216, "micro/f1_ci": {"90": [0.6396879277900763, 0.6792509981861777], "95": [0.6349100561142579, 0.6825887859642396]}, "micro/recall": 0.6398546964193046, "micro/precision": 0.6819690265486725, "macro/f1": 0.6271520713994495, "macro/f1_ci": {"90": [0.6033849011192637, 0.6476622153533104], "95": [0.6003911437652572, 0.6517141318434708]}, "macro/recall": 0.6095524645700295, "macro/precision": 0.652241965435762, "per_entity_metric": {"corporation": {"f1": 0.578125, "f1_ci": {"90": [0.516110546261207, 0.6295629539951574], "95": [0.5061691209281571, 0.6428661162957645]}, "precision": 0.5751295336787565, "recall": 0.581151832460733}, "creative_work": {"f1": 0.5819209039548022, "f1_ci": {"90": [0.5173447188077885, 0.6378554936846392], "95": [0.5104798432384638, 0.6490518162393163]}, "precision": 0.5885714285714285, "recall": 0.5754189944134078}, "event": {"f1": 0.4763572679509632, "f1_ci": {"90": [0.4245514079895219, 0.523213358070501], "95": [0.4141386430678466, 0.532264690402697]}, "precision": 0.4444444444444444, "recall": 0.5132075471698113}, "group": {"f1": 0.5878003696857671, "f1_ci": {"90": [0.5342837363973966, 0.6421063394683026], "95": [0.5218799565453557, 0.6518635025754232]}, "precision": 0.691304347826087, "recall": 0.5112540192926045}, "location": {"f1": 0.6824925816023738, "f1_ci": {"90": [0.6153846153846154, 0.740751896474788], "95": [0.6038780663780664, 0.7509359340481707]}, "precision": 0.6686046511627907, "recall": 0.696969696969697}, "person": {"f1": 0.8336221837088388, "f1_ci": {"90": [0.8053213941676154, 0.8559769434579175], "95": [0.7985865724381626, 0.861693433325605]}, "precision": 0.8620071684587813, "recall": 0.8070469798657718}, "product": {"f1": 0.649746192893401, "f1_ci": {"90": [0.5962006778842162, 0.7002640572369876], "95": [0.581727559347181, 0.7086697496146315]}, "precision": 0.735632183908046, "recall": 0.5818181818181818}}}, "2021.test (span detection)": {"micro/f1": 0.7843071034560397, "micro/f1_ci": {}, "micro/recall": 0.8057129640337689, "micro/precision": 0.7640092115363527, "macro/f1": 0.7843071034560397, "macro/f1_ci": {}, "macro/recall": 0.8057129640337689, "macro/precision": 0.7640092115363527}, "2020.test (span detection)": {"micro/f1": 0.7720332172515403, "micro/f1_ci": {}, "micro/recall": 0.7477944992215879, "micro/precision": 0.7978959025470653, "macro/f1": 0.7720332172515403, "macro/f1_ci": {}, "macro/recall": 0.7477944992215879, "macro/precision": 0.7978959025470653}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6602409638554216, "micro/f1_ci": {"90": [0.6396879277900763, 0.6792509981861777], "95": [0.6349100561142579, 0.6825887859642396]}, "micro/recall": 0.6398546964193046, "micro/precision": 0.6819690265486725, "macro/f1": 0.6271520713994495, "macro/f1_ci": {"90": [0.6033849011192637, 0.6476622153533104], "95": [0.6003911437652572, 0.6517141318434708]}, "macro/recall": 0.6095524645700295, "macro/precision": 0.652241965435762, "per_entity_metric": {"corporation": {"f1": 0.578125, "f1_ci": {"90": [0.516110546261207, 0.6295629539951574], "95": [0.5061691209281571, 0.6428661162957645]}, "precision": 0.5751295336787565, "recall": 0.581151832460733}, "creative_work": {"f1": 0.5819209039548022, "f1_ci": {"90": [0.5173447188077885, 0.6378554936846392], "95": [0.5104798432384638, 0.6490518162393163]}, "precision": 0.5885714285714285, "recall": 0.5754189944134078}, "event": {"f1": 0.4763572679509632, "f1_ci": {"90": [0.4245514079895219, 0.523213358070501], "95": [0.4141386430678466, 0.532264690402697]}, "precision": 0.4444444444444444, "recall": 0.5132075471698113}, "group": {"f1": 0.5878003696857671, "f1_ci": {"90": [0.5342837363973966, 0.6421063394683026], "95": [0.5218799565453557, 0.6518635025754232]}, "precision": 0.691304347826087, "recall": 0.5112540192926045}, "location": {"f1": 0.6824925816023738, "f1_ci": {"90": [0.6153846153846154, 0.740751896474788], "95": [0.6038780663780664, 0.7509359340481707]}, "precision": 0.6686046511627907, "recall": 0.696969696969697}, "person": {"f1": 0.8336221837088388, "f1_ci": {"90": [0.8053213941676154, 0.8559769434579175], "95": [0.7985865724381626, 0.861693433325605]}, "precision": 0.8620071684587813, "recall": 0.8070469798657718}, "product": {"f1": 0.649746192893401, "f1_ci": {"90": [0.5962006778842162, 0.7002640572369876], "95": [0.581727559347181, 0.7086697496146315]}, "precision": 0.735632183908046, "recall": 0.5818181818181818}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6486182247987844, "micro/f1_ci": {"90": [0.6400171924527076, 0.6574431063551344], "95": [0.6382155801831687, 0.6592086893227054]}, "micro/recall": 0.6662812210915818, "micro/precision": 0.6318675293343569, "macro/f1": 0.604868641257225, "macro/f1_ci": {"90": [0.595296924609441, 0.614397231507266], "95": [0.5933778702735921, 0.6163061595932199]}, "macro/recall": 0.6244372176840122, "macro/precision": 0.589818811310092, "per_entity_metric": {"corporation": {"f1": 0.5359342915811088, "f1_ci": {"90": [0.5124250180674441, 0.562196264340956], "95": [0.5066242713301538, 0.5660601358658982]}, "precision": 0.49809160305343514, "recall": 0.58}, "creative_work": {"f1": 0.454661558109834, "f1_ci": {"90": [0.42294214039784905, 0.4866248288332962], "95": [0.4174022952917621, 0.49281072809973214]}, "precision": 0.42634730538922155, "recall": 0.48700410396716826}, "event": {"f1": 0.46186621218576907, "f1_ci": {"90": [0.4407249060721636, 0.48402759846047894], "95": [0.43625117639460426, 0.4896707137226466]}, "precision": 0.4342948717948718, "recall": 0.4931756141947225}, "group": {"f1": 0.6163606010016696, "f1_ci": {"90": [0.5951285544493952, 0.6383988899613899], "95": [0.5921567684006442, 0.6431355549407772]}, "precision": 0.6249153689911984, "recall": 0.6080368906455863}, "location": {"f1": 0.6615873015873016, "f1_ci": {"90": [0.6349800972344961, 0.6883097309494146], "95": [0.6280602263375301, 0.6938339172840142]}, "precision": 0.6065192083818394, "recall": 0.7276536312849162}, "person": {"f1": 0.8278614184654453, "f1_ci": {"90": [0.8174552782574815, 0.8383023554114327], "95": [0.8153046175852993, 0.840402111046631]}, "precision": 0.814709032488397, "recall": 0.8414454277286135}, "product": {"f1": 0.675809105869446, "f1_ci": {"90": [0.6533841788755885, 0.6975096627158484], "95": [0.648618174354878, 0.7003805018510901]}, "precision": 0.7238542890716804, "recall": 0.6337448559670782}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7720332172515403, "micro/f1_ci": {}, "micro/recall": 0.7477944992215879, "micro/precision": 0.7978959025470653, "macro/f1": 0.7720332172515403, "macro/f1_ci": {}, "macro/recall": 0.7477944992215879, "macro/precision": 0.7978959025470653}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7843071034560397, "micro/f1_ci": {}, "micro/recall": 0.8057129640337689, "micro/precision": 0.7640092115363527, "macro/f1": 0.7843071034560397, "macro/f1_ci": {}, "macro/recall": 0.8057129640337689, "macro/precision": 0.7640092115363527}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "random.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_random", "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}