model update

README.md

@@ -0,0 +1,176 @@
+---
+datasets:
+- tner/tweetner7
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/bert-base-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: 0.6008989019741707
+    - name: Precision
+      type: precision
+      value: 0.591443610706686
+    - name: Recall
+      type: recall
+      value: 0.6106614246068455
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5467450408285621
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.537717358363018
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5582367980568581
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7560892328704758
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.744313725490196
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.7682433213831387
+  - 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: 0.6087425796006476
+    - name: Precision
+      type: precision
+      value: 0.6340640809443507
+    - name: Recall
+      type: recall
+      value: 0.5853658536585366
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5648877924450979
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.5930039411771633
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.5426595099078766
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.7242309767943875
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.7543563799887577
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.6964193046185781
+
+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-base-tweetner7-2020
+
+This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-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): 0.6008989019741707
+- Precision (micro): 0.591443610706686
+- Recall (micro): 0.6106614246068455
+- F1 (macro): 0.5467450408285621
+- Precision (macro): 0.537717358363018
+- Recall (macro): 0.5582367980568581
+
+
+
+The per-entity breakdown of the F1 score on the test set are below:
+- corporation: 0.4411294619072989
+- creative_work: 0.3751552795031057
+- event: 0.40279069767441866
+- group: 0.5576791808873721
+- location: 0.6179921773142112
+- person: 0.8051622154507977
+- product: 0.6273062730627307 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.5924664556782363, 0.6106294776916564]
+    - 95%: [0.5905572257793882, 0.6119935888266077] 
+- F1 (macro): 
+    - 90%: [0.5924664556782363, 0.6106294776916564]
+    - 95%: [0.5905572257793882, 0.6119935888266077] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/bert-base-tweetner7-2020/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/bert-base-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-base-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-base-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-base-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": 0.6070060207991242, "micro/f1_ci": {}, "micro/recall": 0.5794148380355277, "micro/precision": 0.6373563218390804, "macro/f1": 0.5467850521523531, "macro/f1_ci": {}, "macro/recall": 0.5207678704660276, "macro/precision": 0.5797488300321645, "per_entity_metric": {"corporation": {"f1": 0.448, "f1_ci": {}, "precision": 0.4883720930232558, "recall": 0.41379310344827586}, "creative_work": {"f1": 0.4136125654450261, "f1_ci": {}, "precision": 0.4540229885057471, "recall": 0.3798076923076923}, "event": {"f1": 0.3472803347280335, "f1_ci": {}, "precision": 0.3738738738738739, "recall": 0.32421875}, "group": {"f1": 0.5349397590361447, "f1_ci": {}, "precision": 0.5904255319148937, "recall": 0.4889867841409692}, "location": {"f1": 0.6047745358090184, "f1_ci": {}, "precision": 0.5816326530612245, "recall": 0.6298342541436464}, "person": {"f1": 0.8340283569641368, "f1_ci": {}, "precision": 0.831946755407654, "recall": 0.8361204013377926}, "product": {"f1": 0.644859813084112, "f1_ci": {}, "precision": 0.7379679144385026, "recall": 0.5726141078838174}}}, "2021.test": {"micro/f1": 0.6008989019741707, "micro/f1_ci": {"90": [0.5924664556782363, 0.6106294776916564], "95": [0.5905572257793882, 0.6119935888266077]}, "micro/recall": 0.6106614246068455, "micro/precision": 0.591443610706686, "macro/f1": 0.5467450408285621, "macro/f1_ci": {"90": [0.537317329281437, 0.5560337192806812], "95": [0.5354694155361338, 0.5580169049677575]}, "macro/recall": 0.5582367980568581, "macro/precision": 0.537717358363018, "per_entity_metric": {"corporation": {"f1": 0.4411294619072989, "f1_ci": {"90": [0.4151031001476261, 0.4668282751041523], "95": [0.4105067256738155, 0.47152140567132833]}, "precision": 0.42374616171954965, "recall": 0.46}, "creative_work": {"f1": 0.3751552795031057, "f1_ci": {"90": [0.3457281280218126, 0.405156902550491], "95": [0.34063869755792114, 0.41240861376957344]}, "precision": 0.3435722411831627, "recall": 0.4131326949384405}, "event": {"f1": 0.40279069767441866, "f1_ci": {"90": [0.3787119736963134, 0.4261736127060404], "95": [0.3742672669780581, 0.43103672973184526]}, "precision": 0.41198858230256896, "recall": 0.3939945404913558}, "group": {"f1": 0.5576791808873721, "f1_ci": {"90": [0.5357206877355878, 0.5791400107976302], "95": [0.5319548928759554, 0.5857991484361441]}, "precision": 0.5786118980169972, "recall": 0.5382081686429513}, "location": {"f1": 0.6179921773142112, "f1_ci": {"90": [0.5907379257184063, 0.6452965887051165], "95": [0.5849555120039597, 0.6503756726024427]}, "precision": 0.5794621026894865, "recall": 0.6620111731843575}, "person": {"f1": 0.8051622154507977, "f1_ci": {"90": [0.7940780338055736, 0.8169159097793982], "95": [0.7911344383076124, 0.8199079325067802]}, "precision": 0.7833972793861179, "recall": 0.8281710914454278}, "product": {"f1": 0.6273062730627307, "f1_ci": {"90": [0.6036743989614689, 0.6494340793310508], "95": [0.600650185458071, 0.6527402503143698]}, "precision": 0.6432432432432432, "recall": 0.6121399176954733}}}, "2020.test": {"micro/f1": 0.6087425796006476, "micro/f1_ci": {"90": [0.5869140222403038, 0.6303137007490964], "95": [0.583969030579955, 0.6335264711755912]}, "micro/recall": 0.5853658536585366, "micro/precision": 0.6340640809443507, "macro/f1": 0.5648877924450979, "macro/f1_ci": {"90": [0.541183700525316, 0.585728250601802], "95": [0.5380102168624624, 0.591001416229615]}, "macro/recall": 0.5426595099078766, "macro/precision": 0.5930039411771633, "per_entity_metric": {"corporation": {"f1": 0.5076923076923078, "f1_ci": {"90": [0.4456521739130435, 0.5648113256921546], "95": [0.4376895486195893, 0.5757609577922077]}, "precision": 0.49748743718592964, "recall": 0.518324607329843}, "creative_work": {"f1": 0.4406779661016949, "f1_ci": {"90": [0.3846046710949218, 0.5], "95": [0.3716590434523217, 0.5099408614379356]}, "precision": 0.44571428571428573, "recall": 0.43575418994413406}, "event": {"f1": 0.3834586466165414, "f1_ci": {"90": [0.3339654883922945, 0.4358004007303413], "95": [0.32399628008752734, 0.446370839473633]}, "precision": 0.38202247191011235, "recall": 0.3849056603773585}, "group": {"f1": 0.5318352059925093, "f1_ci": {"90": [0.47824719715613895, 0.5828882751694703], "95": [0.4639112831305961, 0.5931040718338841]}, "precision": 0.6367713004484304, "recall": 0.4565916398713826}, "location": {"f1": 0.6329113924050632, "f1_ci": {"90": [0.5619170872324033, 0.6942846790495442], "95": [0.5498484235005866, 0.7070288046928012]}, "precision": 0.6622516556291391, "recall": 0.6060606060606061}, "person": {"f1": 0.810580204778157, "f1_ci": {"90": [0.7831524826802577, 0.8346320308628404], "95": [0.7777679623085985, 0.8402630660959797]}, "precision": 0.8246527777777778, "recall": 0.7969798657718121}, "product": {"f1": 0.6470588235294118, "f1_ci": {"90": [0.591608369865995, 0.6975540206577531], "95": [0.5796820837872635, 0.711355295878703]}, "precision": 0.7021276595744681, "recall": 0.6}}}, "2021.test (span detection)": {"micro/f1": 0.7560892328704758, "micro/f1_ci": {}, "micro/recall": 0.7682433213831387, "micro/precision": 0.744313725490196, "macro/f1": 0.7560892328704758, "macro/f1_ci": {}, "macro/recall": 0.7682433213831387, "macro/precision": 0.744313725490196}, "2020.test (span detection)": {"micro/f1": 0.7242309767943875, "micro/f1_ci": {}, "micro/recall": 0.6964193046185781, "micro/precision": 0.7543563799887577, "macro/f1": 0.7242309767943875, "macro/f1_ci": {}, "macro/recall": 0.6964193046185781, "macro/precision": 0.7543563799887577}}

eval/metric.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6087425796006476, "micro/f1_ci": {"90": [0.5869140222403038, 0.6303137007490964], "95": [0.583969030579955, 0.6335264711755912]}, "micro/recall": 0.5853658536585366, "micro/precision": 0.6340640809443507, "macro/f1": 0.5648877924450979, "macro/f1_ci": {"90": [0.541183700525316, 0.585728250601802], "95": [0.5380102168624624, 0.591001416229615]}, "macro/recall": 0.5426595099078766, "macro/precision": 0.5930039411771633, "per_entity_metric": {"corporation": {"f1": 0.5076923076923078, "f1_ci": {"90": [0.4456521739130435, 0.5648113256921546], "95": [0.4376895486195893, 0.5757609577922077]}, "precision": 0.49748743718592964, "recall": 0.518324607329843}, "creative_work": {"f1": 0.4406779661016949, "f1_ci": {"90": [0.3846046710949218, 0.5], "95": [0.3716590434523217, 0.5099408614379356]}, "precision": 0.44571428571428573, "recall": 0.43575418994413406}, "event": {"f1": 0.3834586466165414, "f1_ci": {"90": [0.3339654883922945, 0.4358004007303413], "95": [0.32399628008752734, 0.446370839473633]}, "precision": 0.38202247191011235, "recall": 0.3849056603773585}, "group": {"f1": 0.5318352059925093, "f1_ci": {"90": [0.47824719715613895, 0.5828882751694703], "95": [0.4639112831305961, 0.5931040718338841]}, "precision": 0.6367713004484304, "recall": 0.4565916398713826}, "location": {"f1": 0.6329113924050632, "f1_ci": {"90": [0.5619170872324033, 0.6942846790495442], "95": [0.5498484235005866, 0.7070288046928012]}, "precision": 0.6622516556291391, "recall": 0.6060606060606061}, "person": {"f1": 0.810580204778157, "f1_ci": {"90": [0.7831524826802577, 0.8346320308628404], "95": [0.7777679623085985, 0.8402630660959797]}, "precision": 0.8246527777777778, "recall": 0.7969798657718121}, "product": {"f1": 0.6470588235294118, "f1_ci": {"90": [0.591608369865995, 0.6975540206577531], "95": [0.5796820837872635, 0.711355295878703]}, "precision": 0.7021276595744681, "recall": 0.6}}}

eval/metric.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.6008989019741707, "micro/f1_ci": {"90": [0.5924664556782363, 0.6106294776916564], "95": [0.5905572257793882, 0.6119935888266077]}, "micro/recall": 0.6106614246068455, "micro/precision": 0.591443610706686, "macro/f1": 0.5467450408285621, "macro/f1_ci": {"90": [0.537317329281437, 0.5560337192806812], "95": [0.5354694155361338, 0.5580169049677575]}, "macro/recall": 0.5582367980568581, "macro/precision": 0.537717358363018, "per_entity_metric": {"corporation": {"f1": 0.4411294619072989, "f1_ci": {"90": [0.4151031001476261, 0.4668282751041523], "95": [0.4105067256738155, 0.47152140567132833]}, "precision": 0.42374616171954965, "recall": 0.46}, "creative_work": {"f1": 0.3751552795031057, "f1_ci": {"90": [0.3457281280218126, 0.405156902550491], "95": [0.34063869755792114, 0.41240861376957344]}, "precision": 0.3435722411831627, "recall": 0.4131326949384405}, "event": {"f1": 0.40279069767441866, "f1_ci": {"90": [0.3787119736963134, 0.4261736127060404], "95": [0.3742672669780581, 0.43103672973184526]}, "precision": 0.41198858230256896, "recall": 0.3939945404913558}, "group": {"f1": 0.5576791808873721, "f1_ci": {"90": [0.5357206877355878, 0.5791400107976302], "95": [0.5319548928759554, 0.5857991484361441]}, "precision": 0.5786118980169972, "recall": 0.5382081686429513}, "location": {"f1": 0.6179921773142112, "f1_ci": {"90": [0.5907379257184063, 0.6452965887051165], "95": [0.5849555120039597, 0.6503756726024427]}, "precision": 0.5794621026894865, "recall": 0.6620111731843575}, "person": {"f1": 0.8051622154507977, "f1_ci": {"90": [0.7940780338055736, 0.8169159097793982], "95": [0.7911344383076124, 0.8199079325067802]}, "precision": 0.7833972793861179, "recall": 0.8281710914454278}, "product": {"f1": 0.6273062730627307, "f1_ci": {"90": [0.6036743989614689, 0.6494340793310508], "95": [0.600650185458071, 0.6527402503143698]}, "precision": 0.6432432432432432, "recall": 0.6121399176954733}}}

eval/metric_span.test_2020.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7242309767943875, "micro/f1_ci": {}, "micro/recall": 0.6964193046185781, "micro/precision": 0.7543563799887577, "macro/f1": 0.7242309767943875, "macro/f1_ci": {}, "macro/recall": 0.6964193046185781, "macro/precision": 0.7543563799887577}

eval/metric_span.test_2021.json

@@ -0,0 +1 @@
+{"micro/f1": 0.7560892328704758, "micro/f1_ci": {}, "micro/recall": 0.7682433213831387, "micro/precision": 0.744313725490196, "macro/f1": 0.7560892328704758, "macro/f1_ci": {}, "macro/recall": 0.7682433213831387, "macro/precision": 0.744313725490196}

trainer_config.json

@@ -1 +1 @@
-{"data_split": "2020.train", "model": "bert-base-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_2020", "dataset_name": null, "local_dataset": null, "model": "bert-base-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}