model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bert-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: 0.6142662426169924
+    - name: Precision
+      type: precision
+      value: 0.6035714285714285
+    - name: Recall
+      type: recall
+      value: 0.6253469010175763
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5614355349295936
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5513691216732639
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5731091951352001
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7585501647540052
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7455053042992742
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7720596738753325
+  - 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: 0.6218623481781376
+    - name: Precision
+      type: precision
+      value: 0.6479190101237345
+    - name: Recall
+      type: recall
+      value: 0.5978204462895693
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5814516218649598
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6074235531058303
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.559517342837518
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7379217273954116
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7688413948256468
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7093928386092372
+
+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
+
+This model is a fine-tuned version of [bert-large-cased](https://huggingface.co/bert-large-cased) 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): 0.6142662426169924
+- Precision (micro): 0.6035714285714285
+- Recall (micro): 0.6253469010175763
+- F1 (macro): 0.5614355349295936
+- Precision (macro): 0.5513691216732639
+- Recall (macro): 0.5731091951352001
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.501082251082251
+- creative_work: 0.39033693579148127
+- event: 0.4180478821362799
+- group: 0.573095401509952
+- location: 0.6112600536193029
+- person: 0.8060337178349601
+- product: 0.6301925025329281 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6054860911410611, 0.6239132125979686]
+    - 95%: [0.6039488039051357, 0.6252644472451034] 
+- F1 (macro): 
+    - 90%: [0.6054860911410611, 0.6239132125979686]
+    - 95%: [0.6039488039051357, 0.6252644472451034] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bert-large-tweetner7-2020/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bert-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/bert-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: bert-large-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: 0.3
+ - max_grad_norm: 1
+
+The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/tner/bert-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": 0.6221135560988862, "micro/f1_ci": {}, "micro/recall": 0.5982236154649948, "micro/precision": 0.6479909451046972, "macro/f1": 0.5688471785141028, "macro/f1_ci": {}, "macro/recall": 0.5483976347665008, "macro/precision": 0.5918745986778033, "per_entity_metric": {"corporation": {"f1": 0.4664879356568364, "f1_ci": {}, "precision": 0.5117647058823529, "recall": 0.42857142857142855}, "creative_work": {"f1": 0.47619047619047616, "f1_ci": {}, "precision": 0.4973821989528796, "recall": 0.4567307692307692}, "event": {"f1": 0.37681159420289856, "f1_ci": {}, "precision": 0.4008810572687225, "recall": 0.35546875}, "group": {"f1": 0.5614849187935035, "f1_ci": {}, "precision": 0.5931372549019608, "recall": 0.5330396475770925}, "location": {"f1": 0.6049046321525885, "f1_ci": {}, "precision": 0.5967741935483871, "recall": 0.6132596685082873}, "person": {"f1": 0.8379310344827586, "f1_ci": {}, "precision": 0.8647686832740213, "recall": 0.8127090301003345}, "product": {"f1": 0.6581196581196581, "f1_ci": {}, "precision": 0.6784140969162996, "recall": 0.6390041493775933}}}, "2021.test": {"micro/f1": 0.6142662426169924, "micro/f1_ci": {"90": [0.6054860911410611, 0.6239132125979686], "95": [0.6039488039051357, 0.6252644472451034]}, "micro/recall": 0.6253469010175763, "micro/precision": 0.6035714285714285, "macro/f1": 0.5614355349295936, "macro/f1_ci": {"90": [0.5520803007571674, 0.5710648719603469], "95": [0.5504902392526134, 0.5725661254009528]}, "macro/recall": 0.5731091951352001, "macro/precision": 0.5513691216732639, "per_entity_metric": {"corporation": {"f1": 0.501082251082251, "f1_ci": {"90": [0.4770074846982415, 0.5255478220892417], "95": [0.4709604335226225, 0.528632810053306]}, "precision": 0.4883966244725738, "recall": 0.5144444444444445}, "creative_work": {"f1": 0.39033693579148127, "f1_ci": {"90": [0.36235902604021875, 0.41935695260643474], "95": [0.3568715932217256, 0.42526268539851686]}, "precision": 0.3646080760095012, "recall": 0.41997264021887826}, "event": {"f1": 0.4180478821362799, "f1_ci": {"90": [0.39369736887085927, 0.4410493598015965], "95": [0.3889883888059071, 0.445061691971253]}, "precision": 0.42311276794035413, "recall": 0.41310282074613286}, "group": {"f1": 0.573095401509952, "f1_ci": {"90": [0.5520272454176741, 0.5957167433722008], "95": [0.5485938611177392, 0.601030294198623]}, "precision": 0.5981375358166189, "recall": 0.5500658761528326}, "location": {"f1": 0.6112600536193029, "f1_ci": {"90": [0.5835555295954594, 0.6392736996727908], "95": [0.5768684161500784, 0.6444776182556419]}, "precision": 0.5876288659793815, "recall": 0.6368715083798883}, "person": {"f1": 0.8060337178349601, "f1_ci": {"90": [0.7950308287592746, 0.8177910434069819], "95": [0.7929682405502338, 0.8191804141278767]}, "precision": 0.7769414984604858, "recall": 0.8373893805309734}, "product": {"f1": 0.6301925025329281, "f1_ci": {"90": [0.6085438799207493, 0.6510378873841347], "95": [0.6044293701255484, 0.6561154547162428]}, "precision": 0.6207584830339321, "recall": 0.6399176954732511}}}, "2020.test": {"micro/f1": 0.6218623481781376, "micro/f1_ci": {"90": [0.600598047054793, 0.6418760390271273], "95": [0.5984731519563263, 0.6462189897033449]}, "micro/recall": 0.5978204462895693, "micro/precision": 0.6479190101237345, "macro/f1": 0.5814516218649598, "macro/f1_ci": {"90": [0.5589758960859509, 0.6016250996825379], "95": [0.5546179229876598, 0.6059803861473116]}, "macro/recall": 0.559517342837518, "macro/precision": 0.6074235531058303, "per_entity_metric": {"corporation": {"f1": 0.5667574931880109, "f1_ci": {"90": [0.5073580123810711, 0.620876014005785], "95": [0.4931487876143134, 0.6306901737967915]}, "precision": 0.5909090909090909, "recall": 0.5445026178010471}, "creative_work": {"f1": 0.4375, "f1_ci": {"90": [0.3757942482146305, 0.4938698450223272], "95": [0.36569966814159277, 0.5046488095238096]}, "precision": 0.44508670520231214, "recall": 0.4301675977653631}, "event": {"f1": 0.47238095238095246, "f1_ci": {"90": [0.42434964414851734, 0.5212421212421213], "95": [0.4131135696459861, 0.5296192449244684]}, "precision": 0.47692307692307695, "recall": 0.4679245283018868}, "group": {"f1": 0.49723756906077354, "f1_ci": {"90": [0.4436419938068124, 0.5520573771362415], "95": [0.435036183523845, 0.5635895944482331]}, "precision": 0.5818965517241379, "recall": 0.4340836012861736}, "location": {"f1": 0.6261980830670928, "f1_ci": {"90": [0.5613978494623657, 0.6819569631302302], "95": [0.5519931034482759, 0.6885320192164827]}, "precision": 0.6621621621621622, "recall": 0.593939393939394}, "person": {"f1": 0.8003412969283276, "f1_ci": {"90": [0.7704149384371606, 0.8257669055082848], "95": [0.7641216708991424, 0.8306610084916783]}, "precision": 0.8142361111111112, "recall": 0.7869127516778524}, "product": {"f1": 0.6697459584295613, "f1_ci": {"90": [0.6156779024106119, 0.7180777656078862], "95": [0.6014288611228104, 0.72647030651341]}, "precision": 0.6807511737089202, "recall": 0.6590909090909091}}}, "2021.test (span detection)": {"micro/f1": 0.7585501647540052, "micro/f1_ci": {}, "micro/recall": 0.7720596738753325, "micro/precision": 0.7455053042992742, "macro/f1": 0.7585501647540052, "macro/f1_ci": {}, "macro/recall": 0.7720596738753325, "macro/precision": 0.7455053042992742}, "2020.test (span detection)": {"micro/f1": 0.7379217273954116, "micro/f1_ci": {}, "micro/recall": 0.7093928386092372, "micro/precision": 0.7688413948256468, "macro/f1": 0.7379217273954116, "macro/f1_ci": {}, "macro/recall": 0.7093928386092372, "macro/precision": 0.7688413948256468}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6218623481781376, "micro/f1_ci": {"90": [0.600598047054793, 0.6418760390271273], "95": [0.5984731519563263, 0.6462189897033449]}, "micro/recall": 0.5978204462895693, "micro/precision": 0.6479190101237345, "macro/f1": 0.5814516218649598, "macro/f1_ci": {"90": [0.5589758960859509, 0.6016250996825379], "95": [0.5546179229876598, 0.6059803861473116]}, "macro/recall": 0.559517342837518, "macro/precision": 0.6074235531058303, "per_entity_metric": {"corporation": {"f1": 0.5667574931880109, "f1_ci": {"90": [0.5073580123810711, 0.620876014005785], "95": [0.4931487876143134, 0.6306901737967915]}, "precision": 0.5909090909090909, "recall": 0.5445026178010471}, "creative_work": {"f1": 0.4375, "f1_ci": {"90": [0.3757942482146305, 0.4938698450223272], "95": [0.36569966814159277, 0.5046488095238096]}, "precision": 0.44508670520231214, "recall": 0.4301675977653631}, "event": {"f1": 0.47238095238095246, "f1_ci": {"90": [0.42434964414851734, 0.5212421212421213], "95": [0.4131135696459861, 0.5296192449244684]}, "precision": 0.47692307692307695, "recall": 0.4679245283018868}, "group": {"f1": 0.49723756906077354, "f1_ci": {"90": [0.4436419938068124, 0.5520573771362415], "95": [0.435036183523845, 0.5635895944482331]}, "precision": 0.5818965517241379, "recall": 0.4340836012861736}, "location": {"f1": 0.6261980830670928, "f1_ci": {"90": [0.5613978494623657, 0.6819569631302302], "95": [0.5519931034482759, 0.6885320192164827]}, "precision": 0.6621621621621622, "recall": 0.593939393939394}, "person": {"f1": 0.8003412969283276, "f1_ci": {"90": [0.7704149384371606, 0.8257669055082848], "95": [0.7641216708991424, 0.8306610084916783]}, "precision": 0.8142361111111112, "recall": 0.7869127516778524}, "product": {"f1": 0.6697459584295613, "f1_ci": {"90": [0.6156779024106119, 0.7180777656078862], "95": [0.6014288611228104, 0.72647030651341]}, "precision": 0.6807511737089202, "recall": 0.6590909090909091}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6142662426169924, "micro/f1_ci": {"90": [0.6054860911410611, 0.6239132125979686], "95": [0.6039488039051357, 0.6252644472451034]}, "micro/recall": 0.6253469010175763, "micro/precision": 0.6035714285714285, "macro/f1": 0.5614355349295936, "macro/f1_ci": {"90": [0.5520803007571674, 0.5710648719603469], "95": [0.5504902392526134, 0.5725661254009528]}, "macro/recall": 0.5731091951352001, "macro/precision": 0.5513691216732639, "per_entity_metric": {"corporation": {"f1": 0.501082251082251, "f1_ci": {"90": [0.4770074846982415, 0.5255478220892417], "95": [0.4709604335226225, 0.528632810053306]}, "precision": 0.4883966244725738, "recall": 0.5144444444444445}, "creative_work": {"f1": 0.39033693579148127, "f1_ci": {"90": [0.36235902604021875, 0.41935695260643474], "95": [0.3568715932217256, 0.42526268539851686]}, "precision": 0.3646080760095012, "recall": 0.41997264021887826}, "event": {"f1": 0.4180478821362799, "f1_ci": {"90": [0.39369736887085927, 0.4410493598015965], "95": [0.3889883888059071, 0.445061691971253]}, "precision": 0.42311276794035413, "recall": 0.41310282074613286}, "group": {"f1": 0.573095401509952, "f1_ci": {"90": [0.5520272454176741, 0.5957167433722008], "95": [0.5485938611177392, 0.601030294198623]}, "precision": 0.5981375358166189, "recall": 0.5500658761528326}, "location": {"f1": 0.6112600536193029, "f1_ci": {"90": [0.5835555295954594, 0.6392736996727908], "95": [0.5768684161500784, 0.6444776182556419]}, "precision": 0.5876288659793815, "recall": 0.6368715083798883}, "person": {"f1": 0.8060337178349601, "f1_ci": {"90": [0.7950308287592746, 0.8177910434069819], "95": [0.7929682405502338, 0.8191804141278767]}, "precision": 0.7769414984604858, "recall": 0.8373893805309734}, "product": {"f1": 0.6301925025329281, "f1_ci": {"90": [0.6085438799207493, 0.6510378873841347], "95": [0.6044293701255484, 0.6561154547162428]}, "precision": 0.6207584830339321, "recall": 0.6399176954732511}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7379217273954116, "micro/f1_ci": {}, "micro/recall": 0.7093928386092372, "micro/precision": 0.7688413948256468, "macro/f1": 0.7379217273954116, "macro/f1_ci": {}, "macro/recall": 0.7093928386092372, "macro/precision": 0.7688413948256468}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7585501647540052, "micro/f1_ci": {}, "micro/recall": 0.7720596738753325, "micro/precision": 0.7455053042992742, "macro/f1": 0.7585501647540052, "macro/f1_ci": {}, "macro/recall": 0.7720596738753325, "macro/precision": 0.7455053042992742}

trainer_config.json

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