model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/roberta-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: 0.6404223573969929
+    - name: Precision
+      type: precision
+      value: 0.6356801093643198
+    - name: Recall
+      type: recall
+      value: 0.6452358926919519
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5923474605228576
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5882611136070073
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.6009576961702408
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7803730272596844
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7745500113921167
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7862842604371458
+  - 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: 0.6413755458515283
+    - name: Precision
+      type: precision
+      value: 0.6764536557282671
+    - name: Recall
+      type: recall
+      value: 0.6097560975609756
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5978359372811374
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.6331635922016912
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5696483348245823
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7425607425607427
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.783410138248848
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7057602490918526
+
+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/roberta-base-tweetner7-random
+
+This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-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): 0.6404223573969929
+- Precision (micro): 0.6356801093643198
+- Recall (micro): 0.6452358926919519
+- F1 (macro): 0.5923474605228576
+- Precision (macro): 0.5882611136070073
+- Recall (macro): 0.6009576961702408
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.5073313782991202
+- creative_work: 0.4235127478753541
+- event: 0.4598470363288719
+- group: 0.5972820682797481
+- location: 0.6794871794871795
+- person: 0.8231629100238226
+- product: 0.6558089033659067 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.6318442200239734, 0.6492710692925945]
+    - 95%: [0.6304569517689761, 0.6507742568199575] 
+- F1 (macro): 
+    - 90%: [0.6318442200239734, 0.6492710692925945]
+    - 95%: [0.6304569517689761, 0.6507742568199575] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/roberta-base-tweetner7-random/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/roberta-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/roberta-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: roberta-base
+ - 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/roberta-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

@@ -1 +0,0 @@
-{"random.dev": {"micro/f1": 0.6366382861851139, "micro/f1_ci": {}, "micro/recall": 0.6187933796049119, "micro/precision": 0.6555429864253394, "macro/f1": 0.5877085684942696, "macro/f1_ci": {}, "macro/recall": 0.5721804509246399, "macro/precision": 0.6073972739006512, "per_entity_metric": {"corporation": {"f1": 0.57356608478803, "f1_ci": {}, "precision": 0.5528846153846154, "recall": 0.5958549222797928}, "creative_work": {"f1": 0.46258503401360546, "f1_ci": {}, "precision": 0.5151515151515151, "recall": 0.41975308641975306}, "event": {"f1": 0.3905579399141631, "f1_ci": {}, "precision": 0.4117647058823529, "recall": 0.37142857142857144}, "group": {"f1": 0.61236802413273, "f1_ci": {}, "precision": 0.6403785488958991, "recall": 0.5867052023121387}, "location": {"f1": 0.6246246246246248, "f1_ci": {}, "precision": 0.611764705882353, "recall": 0.6380368098159509}, "person": {"f1": 0.8488262910798122, "f1_ci": {}, "precision": 0.849624060150376, "recall": 0.8480300187617261}, "product": {"f1": 0.6014319809069213, "f1_ci": {}, "precision": 0.6702127659574468, "recall": 0.5454545454545454}}}, "2021.test": {"micro/f1": 0.6404223573969929, "micro/f1_ci": {"90": [0.6318442200239734, 0.6492710692925945], "95": [0.6304569517689761, 0.6507742568199575]}, "micro/recall": 0.6452358926919519, "micro/precision": 0.6356801093643198, "macro/f1": 0.5923474605228576, "macro/f1_ci": {"90": [0.5834248476754859, 0.6017201600387958], "95": [0.5814450008984621, 0.6034547429897775]}, "macro/recall": 0.6009576961702408, "macro/precision": 0.5882611136070073, "per_entity_metric": {"corporation": {"f1": 0.5073313782991202, "f1_ci": {"90": [0.48285841008729546, 0.5319891471347782], "95": [0.4769294748422465, 0.5359498674646497]}, "precision": 0.45287958115183247, "recall": 0.5766666666666667}, "creative_work": {"f1": 0.4235127478753541, "f1_ci": {"90": [0.3920038797569273, 0.4535109162157119], "95": [0.3865055698995115, 0.45722587346643756]}, "precision": 0.4390602055800294, "recall": 0.40902872777017785}, "event": {"f1": 0.4598470363288719, "f1_ci": {"90": [0.4361915405444048, 0.4815770919408292], "95": [0.43228713377486544, 0.48659304158476635]}, "precision": 0.48439073514602216, "recall": 0.43767060964513194}, "group": {"f1": 0.5972820682797481, "f1_ci": {"90": [0.5769763743967469, 0.6182142646082546], "95": [0.5723013739350298, 0.6231996920200187]}, "precision": 0.6010673782521682, "recall": 0.5935441370223979}, "location": {"f1": 0.6794871794871795, "f1_ci": {"90": [0.6532722132111245, 0.7069142033886049], "95": [0.6483586531023602, 0.711041212742627]}, "precision": 0.6279620853080569, "recall": 0.7402234636871509}, "person": {"f1": 0.8231629100238226, "f1_ci": {"90": [0.8123723616450855, 0.8337047799239938], "95": [0.8103268247356568, 0.8359212540153715]}, "precision": 0.8182149362477231, "recall": 0.8281710914454278}, "product": {"f1": 0.6558089033659067, "f1_ci": {"90": [0.6343529502269406, 0.6783496906386242], "95": [0.6296462048936172, 0.6817721129053114]}, "precision": 0.6942528735632184, "recall": 0.6213991769547325}}}, "2020.test": {"micro/f1": 0.6413755458515283, "micro/f1_ci": {"90": [0.6222555476147691, 0.6597865946700048], "95": [0.6193131081459617, 0.6639230650053427]}, "micro/recall": 0.6097560975609756, "micro/precision": 0.6764536557282671, "macro/f1": 0.5978359372811374, "macro/f1_ci": {"90": [0.5764363670094901, 0.618103079896346], "95": [0.5718154248562978, 0.6229228623936891]}, "macro/recall": 0.5696483348245823, "macro/precision": 0.6331635922016912, "per_entity_metric": {"corporation": {"f1": 0.5757575757575757, "f1_ci": {"90": [0.5194705037892698, 0.6281803562144057], "95": [0.5085437100213219, 0.6400349854227406]}, "precision": 0.5560975609756098, "recall": 0.5968586387434555}, "creative_work": {"f1": 0.4761904761904763, "f1_ci": {"90": [0.4155083554583789, 0.5296136875104415], "95": [0.404413869605668, 0.5397183276493623]}, "precision": 0.5514705882352942, "recall": 0.41899441340782123}, "event": {"f1": 0.42190669371196754, "f1_ci": {"90": [0.3673381231542293, 0.4758234343588129], "95": [0.3601694915254237, 0.48509915610074]}, "precision": 0.45614035087719296, "recall": 0.39245283018867927}, "group": {"f1": 0.5648312611012433, "f1_ci": {"90": [0.5137745892154593, 0.6134932533733134], "95": [0.5036973180076628, 0.6260172918988073]}, "precision": 0.6309523809523809, "recall": 0.5112540192926045}, "location": {"f1": 0.6707317073170732, "f1_ci": {"90": [0.6065793065793067, 0.7238615901724621], "95": [0.5954137077134027, 0.7374486720829003]}, "precision": 0.6748466257668712, "recall": 0.6666666666666666}, "person": {"f1": 0.8270547945205479, "f1_ci": {"90": [0.7992982351897211, 0.8511725937669974], "95": [0.7930445979361251, 0.8559374070770146]}, "precision": 0.8444055944055944, "recall": 0.8104026845637584}, "product": {"f1": 0.6483790523690773, "f1_ci": {"90": [0.5959166297912384, 0.7033119235335302], "95": [0.5850000000000001, 0.7156025968817367]}, "precision": 0.7182320441988951, "recall": 0.5909090909090909}}}, "2021.test (span detection)": {"micro/f1": 0.7803730272596844, "micro/f1_ci": {}, "micro/recall": 0.7862842604371458, "micro/precision": 0.7745500113921167, "macro/f1": 0.7803730272596844, "macro/f1_ci": {}, "macro/recall": 0.7862842604371458, "macro/precision": 0.7745500113921167}, "2020.test (span detection)": {"micro/f1": 0.7425607425607427, "micro/f1_ci": {}, "micro/recall": 0.7057602490918526, "micro/precision": 0.783410138248848, "macro/f1": 0.7425607425607427, "macro/f1_ci": {}, "macro/recall": 0.7057602490918526, "macro/precision": 0.783410138248848}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6413755458515283, "micro/f1_ci": {"90": [0.6222555476147691, 0.6597865946700048], "95": [0.6193131081459617, 0.6639230650053427]}, "micro/recall": 0.6097560975609756, "micro/precision": 0.6764536557282671, "macro/f1": 0.5978359372811374, "macro/f1_ci": {"90": [0.5764363670094901, 0.618103079896346], "95": [0.5718154248562978, 0.6229228623936891]}, "macro/recall": 0.5696483348245823, "macro/precision": 0.6331635922016912, "per_entity_metric": {"corporation": {"f1": 0.5757575757575757, "f1_ci": {"90": [0.5194705037892698, 0.6281803562144057], "95": [0.5085437100213219, 0.6400349854227406]}, "precision": 0.5560975609756098, "recall": 0.5968586387434555}, "creative_work": {"f1": 0.4761904761904763, "f1_ci": {"90": [0.4155083554583789, 0.5296136875104415], "95": [0.404413869605668, 0.5397183276493623]}, "precision": 0.5514705882352942, "recall": 0.41899441340782123}, "event": {"f1": 0.42190669371196754, "f1_ci": {"90": [0.3673381231542293, 0.4758234343588129], "95": [0.3601694915254237, 0.48509915610074]}, "precision": 0.45614035087719296, "recall": 0.39245283018867927}, "group": {"f1": 0.5648312611012433, "f1_ci": {"90": [0.5137745892154593, 0.6134932533733134], "95": [0.5036973180076628, 0.6260172918988073]}, "precision": 0.6309523809523809, "recall": 0.5112540192926045}, "location": {"f1": 0.6707317073170732, "f1_ci": {"90": [0.6065793065793067, 0.7238615901724621], "95": [0.5954137077134027, 0.7374486720829003]}, "precision": 0.6748466257668712, "recall": 0.6666666666666666}, "person": {"f1": 0.8270547945205479, "f1_ci": {"90": [0.7992982351897211, 0.8511725937669974], "95": [0.7930445979361251, 0.8559374070770146]}, "precision": 0.8444055944055944, "recall": 0.8104026845637584}, "product": {"f1": 0.6483790523690773, "f1_ci": {"90": [0.5959166297912384, 0.7033119235335302], "95": [0.5850000000000001, 0.7156025968817367]}, "precision": 0.7182320441988951, "recall": 0.5909090909090909}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6404223573969929, "micro/f1_ci": {"90": [0.6318442200239734, 0.6492710692925945], "95": [0.6304569517689761, 0.6507742568199575]}, "micro/recall": 0.6452358926919519, "micro/precision": 0.6356801093643198, "macro/f1": 0.5923474605228576, "macro/f1_ci": {"90": [0.5834248476754859, 0.6017201600387958], "95": [0.5814450008984621, 0.6034547429897775]}, "macro/recall": 0.6009576961702408, "macro/precision": 0.5882611136070073, "per_entity_metric": {"corporation": {"f1": 0.5073313782991202, "f1_ci": {"90": [0.48285841008729546, 0.5319891471347782], "95": [0.4769294748422465, 0.5359498674646497]}, "precision": 0.45287958115183247, "recall": 0.5766666666666667}, "creative_work": {"f1": 0.4235127478753541, "f1_ci": {"90": [0.3920038797569273, 0.4535109162157119], "95": [0.3865055698995115, 0.45722587346643756]}, "precision": 0.4390602055800294, "recall": 0.40902872777017785}, "event": {"f1": 0.4598470363288719, "f1_ci": {"90": [0.4361915405444048, 0.4815770919408292], "95": [0.43228713377486544, 0.48659304158476635]}, "precision": 0.48439073514602216, "recall": 0.43767060964513194}, "group": {"f1": 0.5972820682797481, "f1_ci": {"90": [0.5769763743967469, 0.6182142646082546], "95": [0.5723013739350298, 0.6231996920200187]}, "precision": 0.6010673782521682, "recall": 0.5935441370223979}, "location": {"f1": 0.6794871794871795, "f1_ci": {"90": [0.6532722132111245, 0.7069142033886049], "95": [0.6483586531023602, 0.711041212742627]}, "precision": 0.6279620853080569, "recall": 0.7402234636871509}, "person": {"f1": 0.8231629100238226, "f1_ci": {"90": [0.8123723616450855, 0.8337047799239938], "95": [0.8103268247356568, 0.8359212540153715]}, "precision": 0.8182149362477231, "recall": 0.8281710914454278}, "product": {"f1": 0.6558089033659067, "f1_ci": {"90": [0.6343529502269406, 0.6783496906386242], "95": [0.6296462048936172, 0.6817721129053114]}, "precision": 0.6942528735632184, "recall": 0.6213991769547325}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7425607425607427, "micro/f1_ci": {}, "micro/recall": 0.7057602490918526, "micro/precision": 0.783410138248848, "macro/f1": 0.7425607425607427, "macro/f1_ci": {}, "macro/recall": 0.7057602490918526, "macro/precision": 0.783410138248848}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7803730272596844, "micro/f1_ci": {}, "micro/recall": 0.7862842604371458, "micro/precision": 0.7745500113921167, "macro/f1": 0.7803730272596844, "macro/f1_ci": {}, "macro/recall": 0.7862842604371458, "macro/precision": 0.7745500113921167}

trainer_config.json

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