model update

README.md

@@ -0,0 +1,135 @@
+---
+datasets:
+- ttc
+metrics:
+- f1
+- precision
+- recall
+model-index:
+- name: tner/roberta-large-ttc
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: ttc
+      type: ttc
+      args: ttc
+    metrics:
+    - name: F1
+      type: f1
+      value: 0.8314534321624235
+    - name: Precision
+      type: precision
+      value: 0.8269230769230769
+    - name: Recall
+      type: recall
+      value: 0.8360337005832793
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.8317396497007042
+    - name: Precision (macro)
+      type: precision_macro
+      value: 0.8296690551538254
+    - name: Recall (macro)
+      type: recall_macro
+      value: 0.8340850231639706
+    - name: F1 (entity span)
+      type: f1_entity_span
+      value: 0.8739929100870126
+    - name: Precision (entity span)
+      type: precision_entity_span
+      value: 0.8692307692307693
+    - name: Recall (entity span)
+      type: recall_entity_span
+      value: 0.8788075178224238
+
+pipeline_tag: token-classification
+widget:
+- text: "Jacob Collier is a Grammy awarded artist from England."
+  example_title: "NER Example 1"
+---
+# tner/roberta-large-ttc
+
+This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the 
+[tner/ttc](https://huggingface.co/datasets/tner/ttc) dataset.
+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:
+- F1 (micro): 0.8314534321624235
+- Precision (micro): 0.8269230769230769
+- Recall (micro): 0.8360337005832793
+- F1 (macro): 0.8317396497007042
+- Precision (macro): 0.8296690551538254
+- Recall (macro): 0.8340850231639706
+
+The per-entity breakdown of the F1 score on the test set are below:
+- location: 0.7817403708987161
+- organization: 0.7737656595431097
+- person: 0.939712918660287 
+
+For F1 scores, the confidence interval is obtained by bootstrap as below:
+- F1 (micro): 
+    - 90%: [0.8153670265512099, 0.8476331336073506]
+    - 95%: [0.8126974643551524, 0.8505459585794019] 
+- F1 (macro): 
+    - 90%: [0.8153670265512099, 0.8476331336073506]
+    - 95%: [0.8126974643551524, 0.8505459585794019] 
+
+Full evaluation can be found at [metric file of NER](https://huggingface.co/tner/roberta-large-ttc/raw/main/eval/metric.json) 
+and [metric file of entity span](https://huggingface.co/tner/roberta-large-ttc/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-large-ttc")
+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/ttc']
+ - dataset_split: train
+ - dataset_name: None
+ - local_dataset: None
+ - model: roberta-large
+ - crf: True
+ - max_length: 128
+ - epoch: 16
+ - batch_size: 64
+ - lr: 1e-05
+ - random_seed: 42
+ - gradient_accumulation_steps: 2
+ - weight_decay: None
+ - lr_warmup_step_ratio: 0.1
+ - max_grad_norm: None
+
+The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/tner/roberta-large-ttc/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.",
+}
+
+```

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "tner_ckpt/ttc_roberta_large/best_model",
+  "_name_or_path": "tner_ckpt/ttc_roberta_large/model_hyypuo/epoch_15",
   "architectures": [
     "RobertaForTokenClassification"
   ],

eval/metric.json

@@ -0,0 +1 @@
+{"micro/f1": 0.8314534321624235, "micro/f1_ci": {"90": [0.8153670265512099, 0.8476331336073506], "95": [0.8126974643551524, 0.8505459585794019]}, "micro/recall": 0.8360337005832793, "micro/precision": 0.8269230769230769, "macro/f1": 0.8317396497007042, "macro/f1_ci": {"90": [0.8164152043391184, 0.8483857154851796], "95": [0.8121405650665432, 0.8514439162114872]}, "macro/recall": 0.8340850231639706, "macro/precision": 0.8296690551538254, "per_entity_metric": {"location": {"f1": 0.7817403708987161, "f1_ci": {"90": [0.7488828052379153, 0.8171473330846698], "95": [0.7417012065948236, 0.8221992266543787]}, "precision": 0.7806267806267806, "recall": 0.7828571428571428}, "organization": {"f1": 0.7737656595431097, "f1_ci": {"90": [0.7491889128022777, 0.8000146842878121], "95": [0.7428538612994967, 0.8038162886292103]}, "precision": 0.7586705202312138, "recall": 0.7894736842105263}, "person": {"f1": 0.939712918660287, "f1_ci": {"90": [0.9228016331473269, 0.9576607614300043], "95": [0.9193529240572251, 0.9603688809668725]}, "precision": 0.9497098646034816, "recall": 0.9299242424242424}}}

eval/metric_span.json

@@ -0,0 +1 @@
+{"micro/f1": 0.8739929100870126, "micro/f1_ci": {"90": [0.8607175432025641, 0.8879493299842456], "95": [0.8581408590354063, 0.8902564058328115]}, "micro/recall": 0.8788075178224238, "micro/precision": 0.8692307692307693, "macro/f1": 0.8739929100870126, "macro/f1_ci": {"90": [0.8607175432025641, 0.8879493299842456], "95": [0.8581408590354063, 0.8902564058328115]}, "macro/recall": 0.8788075178224238, "macro/precision": 0.8692307692307693}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:67700356ecb6a6b0119d844b6389abb3fefeb1caffcfd4713a044f05731b5ee9
-size 1417400369
+oid sha256:9c610c34480a6d26278ed2b6da33799d1a0bfe9eafb7df2078b2320b7f58e28a
+size 1417405809

tokenizer_config.json

@@ -6,7 +6,7 @@
   "errors": "replace",
   "mask_token": "<mask>",
   "model_max_length": 512,
-  "name_or_path": "tner_ckpt/ttc_roberta_large/best_model",
+  "name_or_path": "tner_ckpt/ttc_roberta_large/model_hyypuo/epoch_15",
   "pad_token": "<pad>",
   "sep_token": "</s>",
   "special_tokens_map_file": "tner_ckpt/ttc_roberta_large/model_hyypuo/epoch_5/special_tokens_map.json",

trainer_config.json

@@ -0,0 +1 @@
+{"dataset": ["tner/ttc"], "dataset_split": "train", "dataset_name": null, "local_dataset": null, "model": "roberta-large", "crf": true, "max_length": 128, "epoch": 16, "batch_size": 64, "lr": 1e-05, "random_seed": 42, "gradient_accumulation_steps": 2, "weight_decay": null, "lr_warmup_step_ratio": 0.1, "max_grad_norm": null}