model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bert-large-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: 0.6358014184397163
+    - name: Precision
+      type: precision
+      value: 0.6241506071070514
+    - name: Recall
+      type: recall
+      value: 0.647895467160037
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5900128438830615
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5789290375636192
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6041427086183797
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7721468702116793
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7580788945843548
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7867468486180178
+  - 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: 0.6248982912937348
+    - name: Precision
+      type: precision
+      value: 0.6545454545454545
+    - name: Recall
+      type: recall
+      value: 0.5978204462895693
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5862888361879851
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.616927771523058
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5608699203922957
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7357569180683668
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.770892552586697
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7036844836533471
+
+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-concat
+
+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_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): 0.6358014184397163
+- Precision (micro): 0.6241506071070514
+- Recall (micro): 0.647895467160037
+- F1 (macro): 0.5900128438830615
+- Precision (macro): 0.5789290375636192
+- Recall (macro): 0.6041427086183797
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.5013333333333333
+- creative_work: 0.4016441573693482
+- event: 0.47004180213655367
+- group: 0.5973851827019778
+- location: 0.6720321931589538
+- person: 0.8185623293903548
+- product: 0.6690909090909091 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.626687324917235, 0.6449412548744916]
+    - 95%: [0.6246460521646338, 0.6465123623688929] 
+- F1 (macro): 
+    - 90%: [0.626687324917235, 0.6449412548744916]
+    - 95%: [0.6246460521646338, 0.6465123623688929] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bert-large-tweetner7-2020-2021-concat/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bert-large-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/bert-large-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: bert-large-cased
+ - 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/bert-large-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": 0.6309464196294442, "micro/f1_ci": {}, "micro/recall": 0.63, "micro/precision": 0.6318956870611836, "macro/f1": 0.5854681564149163, "macro/f1_ci": {}, "macro/recall": 0.5839910318311838, "macro/precision": 0.5894227103402361, "per_entity_metric": {"corporation": {"f1": 0.5767441860465116, "f1_ci": {}, "precision": 0.5486725663716814, "recall": 0.6078431372549019}, "creative_work": {"f1": 0.44025157232704404, "f1_ci": {}, "precision": 0.4117647058823529, "recall": 0.47297297297297297}, "event": {"f1": 0.41666666666666663, "f1_ci": {}, "precision": 0.41353383458646614, "recall": 0.4198473282442748}, "group": {"f1": 0.634920634920635, "f1_ci": {}, "precision": 0.6542056074766355, "recall": 0.6167400881057269}, "location": {"f1": 0.582089552238806, "f1_ci": {}, "precision": 0.6290322580645161, "recall": 0.5416666666666666}, "person": {"f1": 0.809106830122592, "f1_ci": {}, "precision": 0.8020833333333334, "recall": 0.8162544169611308}, "product": {"f1": 0.6384976525821596, "f1_ci": {}, "precision": 0.6666666666666666, "recall": 0.6126126126126126}}}, "2021.test": {"micro/f1": 0.6358014184397163, "micro/f1_ci": {"90": [0.626687324917235, 0.6449412548744916], "95": [0.6246460521646338, 0.6465123623688929]}, "micro/recall": 0.647895467160037, "micro/precision": 0.6241506071070514, "macro/f1": 0.5900128438830615, "macro/f1_ci": {"90": [0.5803525272420662, 0.5998813669168028], "95": [0.5788003497848505, 0.601259279937617]}, "macro/recall": 0.6041427086183797, "macro/precision": 0.5789290375636192, "per_entity_metric": {"corporation": {"f1": 0.5013333333333333, "f1_ci": {"90": [0.4773041121935053, 0.5251458394589827], "95": [0.47212530903053357, 0.5299806250616865]}, "precision": 0.48205128205128206, "recall": 0.5222222222222223}, "creative_work": {"f1": 0.4016441573693482, "f1_ci": {"90": [0.3726155462184874, 0.4322116501076389], "95": [0.3661632723388598, 0.4400077984403119]}, "precision": 0.35185185185185186, "recall": 0.46785225718194257}, "event": {"f1": 0.47004180213655367, "f1_ci": {"90": [0.44757361821728175, 0.49203100743385103], "95": [0.4413532011324317, 0.49699281439717774]}, "precision": 0.48007590132827327, "recall": 0.4604185623293904}, "group": {"f1": 0.5973851827019778, "f1_ci": {"90": [0.5768783472459447, 0.6186116613189769], "95": [0.5739186663249559, 0.6234884659948738]}, "precision": 0.6081911262798635, "recall": 0.5869565217391305}, "location": {"f1": 0.6720321931589538, "f1_ci": {"90": [0.6440601569406398, 0.6977364758273161], "95": [0.6338632755879807, 0.7029718042123281]}, "precision": 0.6464516129032258, "recall": 0.6997206703910615}, "person": {"f1": 0.8185623293903548, "f1_ci": {"90": [0.8069829934867924, 0.8294376799389931], "95": [0.8044913837616584, 0.8315346050035313]}, "precision": 0.8081207330219188, "recall": 0.8292772861356932}, "product": {"f1": 0.6690909090909091, "f1_ci": {"90": [0.6468565841312396, 0.6917961840792326], "95": [0.6434676970626467, 0.6968318058220542]}, "precision": 0.6757607555089192, "recall": 0.6625514403292181}}}, "2020.test": {"micro/f1": 0.6248982912937348, "micro/f1_ci": {"90": [0.6036183066759389, 0.6448988322002785], "95": [0.5997215835288842, 0.6486262984430212]}, "micro/recall": 0.5978204462895693, "micro/precision": 0.6545454545454545, "macro/f1": 0.5862888361879851, "macro/f1_ci": {"90": [0.5625855374311386, 0.6074574296186365], "95": [0.5597026514157946, 0.6122702483863757]}, "macro/recall": 0.5608699203922957, "macro/precision": 0.616927771523058, "per_entity_metric": {"corporation": {"f1": 0.5555555555555557, "f1_ci": {"90": [0.4929666282756757, 0.6157425077749504], "95": [0.4815702026201889, 0.6239593828404353]}, "precision": 0.5614973262032086, "recall": 0.5497382198952879}, "creative_work": {"f1": 0.47645429362880887, "f1_ci": {"90": [0.41873557692307695, 0.531952314765316], "95": [0.4105496653883751, 0.5422103927516406]}, "precision": 0.4725274725274725, "recall": 0.48044692737430167}, "event": {"f1": 0.4307692307692308, "f1_ci": {"90": [0.3765041888804265, 0.48304065137052626], "95": [0.3662964004499437, 0.4933740655220503]}, "precision": 0.4392156862745098, "recall": 0.4226415094339623}, "group": {"f1": 0.5488029465930019, "f1_ci": {"90": [0.49312550362610796, 0.6021388689955173], "95": [0.4815344237631145, 0.6138935323507213]}, "precision": 0.6422413793103449, "recall": 0.4790996784565916}, "location": {"f1": 0.6389776357827476, "f1_ci": {"90": [0.5704350369233019, 0.6954414208174869], "95": [0.5592458090141159, 0.7055226928772518]}, "precision": 0.6756756756756757, "recall": 0.6060606060606061}, "person": {"f1": 0.8030954428202924, "f1_ci": {"90": [0.7764797488787779, 0.8275862068965517], "95": [0.7715901091859506, 0.8317556848228449]}, "precision": 0.8236331569664903, "recall": 0.7835570469798657}, "product": {"f1": 0.6503667481662592, "f1_ci": {"90": [0.5995089521794967, 0.6982585104933628], "95": [0.5893208083725082, 0.7077953136504405]}, "precision": 0.7037037037037037, "recall": 0.6045454545454545}}}, "2021.test (span detection)": {"micro/f1": 0.7721468702116793, "micro/f1_ci": {}, "micro/recall": 0.7867468486180178, "micro/precision": 0.7580788945843548, "macro/f1": 0.7721468702116793, "macro/f1_ci": {}, "macro/recall": 0.7867468486180178, "macro/precision": 0.7580788945843548}, "2020.test (span detection)": {"micro/f1": 0.7357569180683668, "micro/f1_ci": {}, "micro/recall": 0.7036844836533471, "micro/precision": 0.770892552586697, "macro/f1": 0.7357569180683668, "macro/f1_ci": {}, "macro/recall": 0.7036844836533471, "macro/precision": 0.770892552586697}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6248982912937348, "micro/f1_ci": {"90": [0.6036183066759389, 0.6448988322002785], "95": [0.5997215835288842, 0.6486262984430212]}, "micro/recall": 0.5978204462895693, "micro/precision": 0.6545454545454545, "macro/f1": 0.5862888361879851, "macro/f1_ci": {"90": [0.5625855374311386, 0.6074574296186365], "95": [0.5597026514157946, 0.6122702483863757]}, "macro/recall": 0.5608699203922957, "macro/precision": 0.616927771523058, "per_entity_metric": {"corporation": {"f1": 0.5555555555555557, "f1_ci": {"90": [0.4929666282756757, 0.6157425077749504], "95": [0.4815702026201889, 0.6239593828404353]}, "precision": 0.5614973262032086, "recall": 0.5497382198952879}, "creative_work": {"f1": 0.47645429362880887, "f1_ci": {"90": [0.41873557692307695, 0.531952314765316], "95": [0.4105496653883751, 0.5422103927516406]}, "precision": 0.4725274725274725, "recall": 0.48044692737430167}, "event": {"f1": 0.4307692307692308, "f1_ci": {"90": [0.3765041888804265, 0.48304065137052626], "95": [0.3662964004499437, 0.4933740655220503]}, "precision": 0.4392156862745098, "recall": 0.4226415094339623}, "group": {"f1": 0.5488029465930019, "f1_ci": {"90": [0.49312550362610796, 0.6021388689955173], "95": [0.4815344237631145, 0.6138935323507213]}, "precision": 0.6422413793103449, "recall": 0.4790996784565916}, "location": {"f1": 0.6389776357827476, "f1_ci": {"90": [0.5704350369233019, 0.6954414208174869], "95": [0.5592458090141159, 0.7055226928772518]}, "precision": 0.6756756756756757, "recall": 0.6060606060606061}, "person": {"f1": 0.8030954428202924, "f1_ci": {"90": [0.7764797488787779, 0.8275862068965517], "95": [0.7715901091859506, 0.8317556848228449]}, "precision": 0.8236331569664903, "recall": 0.7835570469798657}, "product": {"f1": 0.6503667481662592, "f1_ci": {"90": [0.5995089521794967, 0.6982585104933628], "95": [0.5893208083725082, 0.7077953136504405]}, "precision": 0.7037037037037037, "recall": 0.6045454545454545}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6358014184397163, "micro/f1_ci": {"90": [0.626687324917235, 0.6449412548744916], "95": [0.6246460521646338, 0.6465123623688929]}, "micro/recall": 0.647895467160037, "micro/precision": 0.6241506071070514, "macro/f1": 0.5900128438830615, "macro/f1_ci": {"90": [0.5803525272420662, 0.5998813669168028], "95": [0.5788003497848505, 0.601259279937617]}, "macro/recall": 0.6041427086183797, "macro/precision": 0.5789290375636192, "per_entity_metric": {"corporation": {"f1": 0.5013333333333333, "f1_ci": {"90": [0.4773041121935053, 0.5251458394589827], "95": [0.47212530903053357, 0.5299806250616865]}, "precision": 0.48205128205128206, "recall": 0.5222222222222223}, "creative_work": {"f1": 0.4016441573693482, "f1_ci": {"90": [0.3726155462184874, 0.4322116501076389], "95": [0.3661632723388598, 0.4400077984403119]}, "precision": 0.35185185185185186, "recall": 0.46785225718194257}, "event": {"f1": 0.47004180213655367, "f1_ci": {"90": [0.44757361821728175, 0.49203100743385103], "95": [0.4413532011324317, 0.49699281439717774]}, "precision": 0.48007590132827327, "recall": 0.4604185623293904}, "group": {"f1": 0.5973851827019778, "f1_ci": {"90": [0.5768783472459447, 0.6186116613189769], "95": [0.5739186663249559, 0.6234884659948738]}, "precision": 0.6081911262798635, "recall": 0.5869565217391305}, "location": {"f1": 0.6720321931589538, "f1_ci": {"90": [0.6440601569406398, 0.6977364758273161], "95": [0.6338632755879807, 0.7029718042123281]}, "precision": 0.6464516129032258, "recall": 0.6997206703910615}, "person": {"f1": 0.8185623293903548, "f1_ci": {"90": [0.8069829934867924, 0.8294376799389931], "95": [0.8044913837616584, 0.8315346050035313]}, "precision": 0.8081207330219188, "recall": 0.8292772861356932}, "product": {"f1": 0.6690909090909091, "f1_ci": {"90": [0.6468565841312396, 0.6917961840792326], "95": [0.6434676970626467, 0.6968318058220542]}, "precision": 0.6757607555089192, "recall": 0.6625514403292181}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7357569180683668, "micro/f1_ci": {}, "micro/recall": 0.7036844836533471, "micro/precision": 0.770892552586697, "macro/f1": 0.7357569180683668, "macro/f1_ci": {}, "macro/recall": 0.7036844836533471, "macro/precision": 0.770892552586697}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7721468702116793, "micro/f1_ci": {}, "micro/recall": 0.7867468486180178, "micro/precision": 0.7580788945843548, "macro/f1": 0.7721468702116793, "macro/f1_ci": {}, "macro/recall": 0.7867468486180178, "macro/precision": 0.7580788945843548}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2020_2021.train", "model": "bert-large-cased", "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": "bert-large-cased", "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}