model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/twitter-roberta-base-dec2021-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: 0.6346897022050466
+    - name: Precision
+      type: precision
+      value: 0.6240500670540903
+    - name: Recall
+      type: recall
+      value: 0.6456984273820536
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.586830362928695
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5777962671668668
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5983908809408913
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.77487922705314
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7618462226195798
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7883659072510697
+  - 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: 0.6225596529284164
+    - name: Precision
+      type: precision
+      value: 0.6519023282226007
+    - name: Recall
+      type: recall
+      value: 0.5957446808510638
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.578847416026638
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6085991227224318
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5537596756202443
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7413232104121477
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7762634866553095
+    - 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/twitter-roberta-base-dec2021-tweetner7-2021
+
+This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base-dec2021](https://huggingface.co/cardiffnlp/twitter-roberta-base-dec2021) 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): 0.6346897022050466
+- Precision (micro): 0.6240500670540903
+- Recall (micro): 0.6456984273820536
+- F1 (macro): 0.586830362928695
+- Precision (macro): 0.5777962671668668
+- Recall (macro): 0.5983908809408913
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.47679083094555874
+- creative_work: 0.4394942805538832
+- event: 0.4638082065467958
+- group: 0.5936801787424194
+- location: 0.646505376344086
+- person: 0.8201674554058972
+- product: 0.6673662119622246 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6258493958055198, 0.6436753593746133]
+    - 95%: [0.6239476803844971, 0.645859449522042] 
+- F1 (macro): 
+    - 90%: [0.6258493958055198, 0.6436753593746133]
+    - 95%: [0.6239476803844971, 0.645859449522042] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/twitter-roberta-base-dec2021-tweetner7-2021/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/twitter-roberta-base-dec2021-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/twitter-roberta-base-dec2021-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: cardiffnlp/twitter-roberta-base-dec2021
+ - 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/twitter-roberta-base-dec2021-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": 0.6218655967903711, "micro/f1_ci": {}, "micro/recall": 0.62, "micro/precision": 0.6237424547283702, "macro/f1": 0.5760577473801914, "macro/f1_ci": {}, "macro/recall": 0.5737958267673842, "macro/precision": 0.5822767568872816, "per_entity_metric": {"corporation": {"f1": 0.5670103092783505, "f1_ci": {}, "precision": 0.5978260869565217, "recall": 0.5392156862745098}, "creative_work": {"f1": 0.43529411764705883, "f1_ci": {}, "precision": 0.3854166666666667, "recall": 0.5}, "event": {"f1": 0.40944881889763773, "f1_ci": {}, "precision": 0.42276422764227645, "recall": 0.3969465648854962}, "group": {"f1": 0.6085011185682326, "f1_ci": {}, "precision": 0.6181818181818182, "recall": 0.5991189427312775}, "location": {"f1": 0.591549295774648, "f1_ci": {}, "precision": 0.6, "recall": 0.5833333333333334}, "person": {"f1": 0.8089500860585198, "f1_ci": {}, "precision": 0.7885906040268457, "recall": 0.8303886925795053}, "product": {"f1": 0.6116504854368933, "f1_ci": {}, "precision": 0.6631578947368421, "recall": 0.5675675675675675}}}, "2021.test": {"micro/f1": 0.6346897022050466, "micro/f1_ci": {"90": [0.6258493958055198, 0.6436753593746133], "95": [0.6239476803844971, 0.645859449522042]}, "micro/recall": 0.6456984273820536, "micro/precision": 0.6240500670540903, "macro/f1": 0.586830362928695, "macro/f1_ci": {"90": [0.5771640962569892, 0.5964231357348152], "95": [0.5752576281332251, 0.5978908050403762]}, "macro/recall": 0.5983908809408913, "macro/precision": 0.5777962671668668, "per_entity_metric": {"corporation": {"f1": 0.47679083094555874, "f1_ci": {"90": [0.44962110787681986, 0.5026906852126324], "95": [0.44365607999181766, 0.5066567600970541]}, "precision": 0.49230769230769234, "recall": 0.4622222222222222}, "creative_work": {"f1": 0.4394942805538832, "f1_ci": {"90": [0.4094896245645097, 0.47030672977531146], "95": [0.4045742789092769, 0.47468701533345065]}, "precision": 0.3924731182795699, "recall": 0.4993160054719562}, "event": {"f1": 0.4638082065467958, "f1_ci": {"90": [0.43910944336457225, 0.4870765791700006], "95": [0.4365052503299458, 0.49078351018856553]}, "precision": 0.4700934579439252, "recall": 0.45768880800727935}, "group": {"f1": 0.5936801787424194, "f1_ci": {"90": [0.572508763379824, 0.6152662972615333], "95": [0.5691102521455328, 0.6196567468373919]}, "precision": 0.5758513931888545, "recall": 0.6126482213438735}, "location": {"f1": 0.646505376344086, "f1_ci": {"90": [0.6196162284066153, 0.6739581938679877], "95": [0.6134858292007842, 0.6786011609002209]}, "precision": 0.6230569948186528, "recall": 0.6717877094972067}, "person": {"f1": 0.8201674554058972, "f1_ci": {"90": [0.8092938133961863, 0.831688063557384], "95": [0.8071047952694514, 0.8335989757976281]}, "precision": 0.8098490294751977, "recall": 0.8307522123893806}, "product": {"f1": 0.6673662119622246, "f1_ci": {"90": [0.645805115139754, 0.6879920293654955], "95": [0.6426681184443668, 0.6919908598661983]}, "precision": 0.6809421841541756, "recall": 0.654320987654321}}}, "2020.test": {"micro/f1": 0.6225596529284164, "micro/f1_ci": {"90": [0.6019905097377952, 0.6423539604890164], "95": [0.5983476899788864, 0.6462412396621156]}, "micro/recall": 0.5957446808510638, "micro/precision": 0.6519023282226007, "macro/f1": 0.578847416026638, "macro/f1_ci": {"90": [0.5562972923933772, 0.5997658810655637], "95": [0.5527899851014745, 0.6043389771515122]}, "macro/recall": 0.5537596756202443, "macro/precision": 0.6085991227224318, "per_entity_metric": {"corporation": {"f1": 0.534435261707989, "f1_ci": {"90": [0.47284258230309767, 0.5900909519728407], "95": [0.4602715480664323, 0.6000127226463106]}, "precision": 0.563953488372093, "recall": 0.5078534031413613}, "creative_work": {"f1": 0.4754098360655738, "f1_ci": {"90": [0.4173550724637681, 0.5336544456167273], "95": [0.40729210272464195, 0.5422017152091206]}, "precision": 0.46524064171123, "recall": 0.4860335195530726}, "event": {"f1": 0.4296875, "f1_ci": {"90": [0.37522969052224375, 0.4818120260021668], "95": [0.3632396449704142, 0.48974936278674613]}, "precision": 0.44534412955465585, "recall": 0.41509433962264153}, "group": {"f1": 0.5376344086021505, "f1_ci": {"90": [0.4799852398523985, 0.5894486870704341], "95": [0.4690353394343622, 0.5983626121961282]}, "precision": 0.6072874493927125, "recall": 0.48231511254019294}, "location": {"f1": 0.6149068322981366, "f1_ci": {"90": [0.5443919770773639, 0.6801455414358639], "95": [0.5262757654592755, 0.6942954434100932]}, "precision": 0.6305732484076433, "recall": 0.6}, "person": {"f1": 0.8164665523156088, "f1_ci": {"90": [0.7882536040376671, 0.8402708997285495], "95": [0.7829841079059829, 0.8457393226658165]}, "precision": 0.8350877192982457, "recall": 0.7986577181208053}, "product": {"f1": 0.6433915211970074, "f1_ci": {"90": [0.5870512960868032, 0.6978238860148408], "95": [0.5771771057252095, 0.7093856924920999]}, "precision": 0.712707182320442, "recall": 0.5863636363636363}}}, "2021.test (span detection)": {"micro/f1": 0.77487922705314, "micro/f1_ci": {}, "micro/recall": 0.7883659072510697, "micro/precision": 0.7618462226195798, "macro/f1": 0.77487922705314, "macro/f1_ci": {}, "macro/recall": 0.7883659072510697, "macro/precision": 0.7618462226195798}, "2020.test (span detection)": {"micro/f1": 0.7413232104121477, "micro/f1_ci": {}, "micro/recall": 0.7093928386092372, "micro/precision": 0.7762634866553095, "macro/f1": 0.7413232104121477, "macro/f1_ci": {}, "macro/recall": 0.7093928386092372, "macro/precision": 0.7762634866553095}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6225596529284164, "micro/f1_ci": {"90": [0.6019905097377952, 0.6423539604890164], "95": [0.5983476899788864, 0.6462412396621156]}, "micro/recall": 0.5957446808510638, "micro/precision": 0.6519023282226007, "macro/f1": 0.578847416026638, "macro/f1_ci": {"90": [0.5562972923933772, 0.5997658810655637], "95": [0.5527899851014745, 0.6043389771515122]}, "macro/recall": 0.5537596756202443, "macro/precision": 0.6085991227224318, "per_entity_metric": {"corporation": {"f1": 0.534435261707989, "f1_ci": {"90": [0.47284258230309767, 0.5900909519728407], "95": [0.4602715480664323, 0.6000127226463106]}, "precision": 0.563953488372093, "recall": 0.5078534031413613}, "creative_work": {"f1": 0.4754098360655738, "f1_ci": {"90": [0.4173550724637681, 0.5336544456167273], "95": [0.40729210272464195, 0.5422017152091206]}, "precision": 0.46524064171123, "recall": 0.4860335195530726}, "event": {"f1": 0.4296875, "f1_ci": {"90": [0.37522969052224375, 0.4818120260021668], "95": [0.3632396449704142, 0.48974936278674613]}, "precision": 0.44534412955465585, "recall": 0.41509433962264153}, "group": {"f1": 0.5376344086021505, "f1_ci": {"90": [0.4799852398523985, 0.5894486870704341], "95": [0.4690353394343622, 0.5983626121961282]}, "precision": 0.6072874493927125, "recall": 0.48231511254019294}, "location": {"f1": 0.6149068322981366, "f1_ci": {"90": [0.5443919770773639, 0.6801455414358639], "95": [0.5262757654592755, 0.6942954434100932]}, "precision": 0.6305732484076433, "recall": 0.6}, "person": {"f1": 0.8164665523156088, "f1_ci": {"90": [0.7882536040376671, 0.8402708997285495], "95": [0.7829841079059829, 0.8457393226658165]}, "precision": 0.8350877192982457, "recall": 0.7986577181208053}, "product": {"f1": 0.6433915211970074, "f1_ci": {"90": [0.5870512960868032, 0.6978238860148408], "95": [0.5771771057252095, 0.7093856924920999]}, "precision": 0.712707182320442, "recall": 0.5863636363636363}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6346897022050466, "micro/f1_ci": {"90": [0.6258493958055198, 0.6436753593746133], "95": [0.6239476803844971, 0.645859449522042]}, "micro/recall": 0.6456984273820536, "micro/precision": 0.6240500670540903, "macro/f1": 0.586830362928695, "macro/f1_ci": {"90": [0.5771640962569892, 0.5964231357348152], "95": [0.5752576281332251, 0.5978908050403762]}, "macro/recall": 0.5983908809408913, "macro/precision": 0.5777962671668668, "per_entity_metric": {"corporation": {"f1": 0.47679083094555874, "f1_ci": {"90": [0.44962110787681986, 0.5026906852126324], "95": [0.44365607999181766, 0.5066567600970541]}, "precision": 0.49230769230769234, "recall": 0.4622222222222222}, "creative_work": {"f1": 0.4394942805538832, "f1_ci": {"90": [0.4094896245645097, 0.47030672977531146], "95": [0.4045742789092769, 0.47468701533345065]}, "precision": 0.3924731182795699, "recall": 0.4993160054719562}, "event": {"f1": 0.4638082065467958, "f1_ci": {"90": [0.43910944336457225, 0.4870765791700006], "95": [0.4365052503299458, 0.49078351018856553]}, "precision": 0.4700934579439252, "recall": 0.45768880800727935}, "group": {"f1": 0.5936801787424194, "f1_ci": {"90": [0.572508763379824, 0.6152662972615333], "95": [0.5691102521455328, 0.6196567468373919]}, "precision": 0.5758513931888545, "recall": 0.6126482213438735}, "location": {"f1": 0.646505376344086, "f1_ci": {"90": [0.6196162284066153, 0.6739581938679877], "95": [0.6134858292007842, 0.6786011609002209]}, "precision": 0.6230569948186528, "recall": 0.6717877094972067}, "person": {"f1": 0.8201674554058972, "f1_ci": {"90": [0.8092938133961863, 0.831688063557384], "95": [0.8071047952694514, 0.8335989757976281]}, "precision": 0.8098490294751977, "recall": 0.8307522123893806}, "product": {"f1": 0.6673662119622246, "f1_ci": {"90": [0.645805115139754, 0.6879920293654955], "95": [0.6426681184443668, 0.6919908598661983]}, "precision": 0.6809421841541756, "recall": 0.654320987654321}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7413232104121477, "micro/f1_ci": {}, "micro/recall": 0.7093928386092372, "micro/precision": 0.7762634866553095, "macro/f1": 0.7413232104121477, "macro/f1_ci": {}, "macro/recall": 0.7093928386092372, "macro/precision": 0.7762634866553095}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.77487922705314, "micro/f1_ci": {}, "micro/recall": 0.7883659072510697, "micro/precision": 0.7618462226195798, "macro/f1": 0.77487922705314, "macro/f1_ci": {}, "macro/recall": 0.7883659072510697, "macro/precision": 0.7618462226195798}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2021.train", "model": "cardiffnlp/twitter-roberta-base-dec2021", "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_2021", "dataset_name": null, "local_dataset": null, "model": "cardiffnlp/twitter-roberta-base-dec2021", "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}