model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bertweet-base-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: 0.6555135815794207
+    - name: Precision
+      type: precision
+      value: 0.6807821646531323
+    - name: Recall
+      type: recall
+      value: 0.6320536540240518
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5958197063152341
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6249946723205074
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5736622995381765
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7780043175821539
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.8079461950429693
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7502023823291315
+  - 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: 0.6389047347404451
+    - name: Precision
+      type: precision
+      value: 0.7093096896770108
+    - name: Recall
+      type: recall
+      value: 0.5812143227815257
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.586121467459777
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6621669635440725
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5294767225396645
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7438676554478039
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.8258391386953768
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.6766995329527763
+
+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-base-tweetner7-random
+
+This model is a fine-tuned version of [vinai/bertweet-base](https://huggingface.co/vinai/bertweet-base) 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): 0.6555135815794207
+- Precision (micro): 0.6807821646531323
+- Recall (micro): 0.6320536540240518
+- F1 (macro): 0.5958197063152341
+- Precision (macro): 0.6249946723205074
+- Recall (macro): 0.5736622995381765
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.49599012954966076
+- creative_work: 0.40063091482649843
+- event: 0.47287615148413514
+- group: 0.6206664422753282
+- location: 0.6798096532970768
+- person: 0.8351528384279476
+- product: 0.6656118143459916 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6458952197843215, 0.6643997426393443]
+    - 95%: [0.6443089692503373, 0.6658257158915145] 
+- F1 (macro): 
+    - 90%: [0.6458952197843215, 0.6643997426393443]
+    - 95%: [0.6443089692503373, 0.6658257158915145] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bertweet-base-tweetner7-random/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bertweet-base-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-base-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-base
+ - 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/bertweet-base-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 → metric.test_2020.json}

@@ -1 +1 @@
-{"random.dev": {"micro/f1": 0.6552217453505008, "micro/f1_ci": {}, "micro/recall": 0.611318739989322, "micro/precision": 0.7059186189889026, "macro/f1": 0.5988762510196898, "macro/f1_ci": {}, "macro/recall": 0.5542334451605068, "macro/precision": 0.6593285098237874, "per_entity_metric": {"corporation": {"f1": 0.5209580838323353, "f1_ci": {}, "precision": 0.6170212765957447, "recall": 0.45077720207253885}, "creative_work": {"f1": 0.50187265917603, "f1_ci": {}, "precision": 0.638095238095238, "recall": 0.41358024691358025}, "event": {"f1": 0.38875878220140514, "f1_ci": {}, "precision": 0.45604395604395603, "recall": 0.33877551020408164}, "group": {"f1": 0.6428571428571428, "f1_ci": {}, "precision": 0.6625766871165644, "recall": 0.6242774566473989}, "location": {"f1": 0.637223974763407, "f1_ci": {}, "precision": 0.6558441558441559, "recall": 0.6196319018404908}, "person": {"f1": 0.8627819548872181, "f1_ci": {}, "precision": 0.864406779661017, "recall": 0.8611632270168855}, "product": {"f1": 0.6376811594202898, "f1_ci": {}, "precision": 0.7213114754098361, "recall": 0.5714285714285714}}}, "2021.test": {"micro/f1": 0.6555135815794207, "micro/f1_ci": {"90": [0.6458952197843215, 0.6643997426393443], "95": [0.6443089692503373, 0.6658257158915145]}, "micro/recall": 0.6320536540240518, "micro/precision": 0.6807821646531323, "macro/f1": 0.5958197063152341, "macro/f1_ci": {"90": [0.585785434411046, 0.6053748531850129], "95": [0.5838402026896898, 0.6072114575354797]}, "macro/recall": 0.5736622995381765, "macro/precision": 0.6249946723205074, "per_entity_metric": {"corporation": {"f1": 0.49599012954966076, "f1_ci": {"90": [0.46883518739617464, 0.522878197204526], "95": [0.4619099785663253, 0.5290631359930873]}, "precision": 0.5575589459084604, "recall": 0.44666666666666666}, "creative_work": {"f1": 0.40063091482649843, "f1_ci": {"90": [0.36874728260869566, 0.43303929430633514], "95": [0.3629171852523146, 0.43848724048960525]}, "precision": 0.4729981378026071, "recall": 0.34746922024623805}, "event": {"f1": 0.47287615148413514, "f1_ci": {"90": [0.44740757125719577, 0.49634094981119414], "95": [0.4430748367654445, 0.5009967828889779]}, "precision": 0.5403508771929825, "recall": 0.42038216560509556}, "group": {"f1": 0.6206664422753282, "f1_ci": {"90": [0.5997751796772071, 0.6415149815388977], "95": [0.5964844757099186, 0.6457991975341445]}, "precision": 0.6345492085340675, "recall": 0.6073781291172595}, "location": {"f1": 0.6798096532970768, "f1_ci": {"90": [0.6518312969865437, 0.7069473970234486], "95": [0.6458321283960411, 0.7113758984903604]}, "precision": 0.6622516556291391, "recall": 0.6983240223463687}, "person": {"f1": 0.8351528384279476, "f1_ci": {"90": [0.8242736044555532, 0.8456305452479886], "95": [0.8225215199841636, 0.8476744109543876]}, "precision": 0.8243534482758621, "recall": 0.8462389380530974}, "product": {"f1": 0.6656118143459916, "f1_ci": {"90": [0.6426097657451765, 0.6862539349422875], "95": [0.6373492325760903, 0.6910826330952685]}, "precision": 0.6829004329004329, "recall": 0.6491769547325102}}}, "2020.test": {"micro/f1": 0.6389047347404451, "micro/f1_ci": {"90": [0.6187593935345156, 0.6584639871206162], "95": [0.6145490760162841, 0.6627502292906674]}, "micro/recall": 0.5812143227815257, "micro/precision": 0.7093096896770108, "macro/f1": 0.586121467459777, "macro/f1_ci": {"90": [0.5621903624073958, 0.6070822851718417], "95": [0.5579615035485642, 0.6117235352428714]}, "macro/recall": 0.5294767225396645, "macro/precision": 0.6621669635440725, "per_entity_metric": {"corporation": {"f1": 0.5438066465256798, "f1_ci": {"90": [0.4757904290634208, 0.6006616735627367], "95": [0.4604531711048777, 0.6162243446453973]}, "precision": 0.6428571428571429, "recall": 0.4712041884816754}, "creative_work": {"f1": 0.45051194539249145, "f1_ci": {"90": [0.3833661885663789, 0.5145250175345909], "95": [0.3745328676905309, 0.5244931327665141]}, "precision": 0.5789473684210527, "recall": 0.3687150837988827}, "event": {"f1": 0.4197002141327623, "f1_ci": {"90": [0.3664302564646572, 0.46942169328928907], "95": [0.35712632275132267, 0.47826311071268485]}, "precision": 0.48514851485148514, "recall": 0.36981132075471695}, "group": {"f1": 0.5588235294117647, "f1_ci": {"90": [0.5081523214770568, 0.6080405121136144], "95": [0.49904922344448926, 0.6185954944178629]}, "precision": 0.6523605150214592, "recall": 0.4887459807073955}, "location": {"f1": 0.6602564102564102, "f1_ci": {"90": [0.5993669608811564, 0.7197252248114239], "95": [0.5878272065772066, 0.7295617970622579]}, "precision": 0.7006802721088435, "recall": 0.6242424242424243}, "person": {"f1": 0.8336192109777015, "f1_ci": {"90": [0.8076141024001363, 0.8562484124105687], "95": [0.8038348561764808, 0.8596338619451357]}, "precision": 0.8526315789473684, "recall": 0.8154362416107382}, "product": {"f1": 0.6361323155216285, "f1_ci": {"90": [0.5794345441276866, 0.6856165902940188], "95": [0.5674891662127922, 0.6944029017164559]}, "precision": 0.7225433526011561, "recall": 0.5681818181818182}}}, "2021.test (span detection)": {"micro/f1": 0.7780043175821539, "micro/f1_ci": {}, "micro/recall": 0.7502023823291315, "micro/precision": 0.8079461950429693, "macro/f1": 0.7780043175821539, "macro/f1_ci": {}, "macro/recall": 0.7502023823291315, "macro/precision": 0.8079461950429693}, "2020.test (span detection)": {"micro/f1": 0.7438676554478039, "micro/f1_ci": {}, "micro/recall": 0.6766995329527763, "micro/precision": 0.8258391386953768, "macro/f1": 0.7438676554478039, "macro/f1_ci": {}, "macro/recall": 0.6766995329527763, "macro/precision": 0.8258391386953768}}
+{"micro/f1": 0.6389047347404451, "micro/f1_ci": {"90": [0.6187593935345156, 0.6584639871206162], "95": [0.6145490760162841, 0.6627502292906674]}, "micro/recall": 0.5812143227815257, "micro/precision": 0.7093096896770108, "macro/f1": 0.586121467459777, "macro/f1_ci": {"90": [0.5621903624073958, 0.6070822851718417], "95": [0.5579615035485642, 0.6117235352428714]}, "macro/recall": 0.5294767225396645, "macro/precision": 0.6621669635440725, "per_entity_metric": {"corporation": {"f1": 0.5438066465256798, "f1_ci": {"90": [0.4757904290634208, 0.6006616735627367], "95": [0.4604531711048777, 0.6162243446453973]}, "precision": 0.6428571428571429, "recall": 0.4712041884816754}, "creative_work": {"f1": 0.45051194539249145, "f1_ci": {"90": [0.3833661885663789, 0.5145250175345909], "95": [0.3745328676905309, 0.5244931327665141]}, "precision": 0.5789473684210527, "recall": 0.3687150837988827}, "event": {"f1": 0.4197002141327623, "f1_ci": {"90": [0.3664302564646572, 0.46942169328928907], "95": [0.35712632275132267, 0.47826311071268485]}, "precision": 0.48514851485148514, "recall": 0.36981132075471695}, "group": {"f1": 0.5588235294117647, "f1_ci": {"90": [0.5081523214770568, 0.6080405121136144], "95": [0.49904922344448926, 0.6185954944178629]}, "precision": 0.6523605150214592, "recall": 0.4887459807073955}, "location": {"f1": 0.6602564102564102, "f1_ci": {"90": [0.5993669608811564, 0.7197252248114239], "95": [0.5878272065772066, 0.7295617970622579]}, "precision": 0.7006802721088435, "recall": 0.6242424242424243}, "person": {"f1": 0.8336192109777015, "f1_ci": {"90": [0.8076141024001363, 0.8562484124105687], "95": [0.8038348561764808, 0.8596338619451357]}, "precision": 0.8526315789473684, "recall": 0.8154362416107382}, "product": {"f1": 0.6361323155216285, "f1_ci": {"90": [0.5794345441276866, 0.6856165902940188], "95": [0.5674891662127922, 0.6944029017164559]}, "precision": 0.7225433526011561, "recall": 0.5681818181818182}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6555135815794207, "micro/f1_ci": {"90": [0.6458952197843215, 0.6643997426393443], "95": [0.6443089692503373, 0.6658257158915145]}, "micro/recall": 0.6320536540240518, "micro/precision": 0.6807821646531323, "macro/f1": 0.5958197063152341, "macro/f1_ci": {"90": [0.585785434411046, 0.6053748531850129], "95": [0.5838402026896898, 0.6072114575354797]}, "macro/recall": 0.5736622995381765, "macro/precision": 0.6249946723205074, "per_entity_metric": {"corporation": {"f1": 0.49599012954966076, "f1_ci": {"90": [0.46883518739617464, 0.522878197204526], "95": [0.4619099785663253, 0.5290631359930873]}, "precision": 0.5575589459084604, "recall": 0.44666666666666666}, "creative_work": {"f1": 0.40063091482649843, "f1_ci": {"90": [0.36874728260869566, 0.43303929430633514], "95": [0.3629171852523146, 0.43848724048960525]}, "precision": 0.4729981378026071, "recall": 0.34746922024623805}, "event": {"f1": 0.47287615148413514, "f1_ci": {"90": [0.44740757125719577, 0.49634094981119414], "95": [0.4430748367654445, 0.5009967828889779]}, "precision": 0.5403508771929825, "recall": 0.42038216560509556}, "group": {"f1": 0.6206664422753282, "f1_ci": {"90": [0.5997751796772071, 0.6415149815388977], "95": [0.5964844757099186, 0.6457991975341445]}, "precision": 0.6345492085340675, "recall": 0.6073781291172595}, "location": {"f1": 0.6798096532970768, "f1_ci": {"90": [0.6518312969865437, 0.7069473970234486], "95": [0.6458321283960411, 0.7113758984903604]}, "precision": 0.6622516556291391, "recall": 0.6983240223463687}, "person": {"f1": 0.8351528384279476, "f1_ci": {"90": [0.8242736044555532, 0.8456305452479886], "95": [0.8225215199841636, 0.8476744109543876]}, "precision": 0.8243534482758621, "recall": 0.8462389380530974}, "product": {"f1": 0.6656118143459916, "f1_ci": {"90": [0.6426097657451765, 0.6862539349422875], "95": [0.6373492325760903, 0.6910826330952685]}, "precision": 0.6829004329004329, "recall": 0.6491769547325102}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7438676554478039, "micro/f1_ci": {}, "micro/recall": 0.6766995329527763, "micro/precision": 0.8258391386953768, "macro/f1": 0.7438676554478039, "macro/f1_ci": {}, "macro/recall": 0.6766995329527763, "macro/precision": 0.8258391386953768}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7780043175821539, "micro/f1_ci": {}, "micro/recall": 0.7502023823291315, "micro/precision": 0.8079461950429693, "macro/f1": 0.7780043175821539, "macro/f1_ci": {}, "macro/recall": 0.7502023823291315, "macro/precision": 0.8079461950429693}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "random.train", "model": "vinai/bertweet-base", "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_random", "dataset_name": null, "local_dataset": null, "model": "vinai/bertweet-base", "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}