update from Yaxin

README.md

@@ -0,0 +1,84 @@
+---
+license: mit
+tags:
+- generated_from_trainer
+datasets:
+- conll2003
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: test-conll2003-ner
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: conll2003
+      type: conll2003
+      args: conll2003
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.9459188783174762
+    - name: Recall
+      type: recall
+      value: 0.9537192864355436
+    - name: F1
+      type: f1
+      value: 0.94980306712478
+    - name: Accuracy
+      type: accuracy
+      value: 0.9911218410498034
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# test-conll2003-ner
+
+This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the conll2003 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0470
+- Precision: 0.9459
+- Recall: 0.9537
+- F1: 0.9498
+- Accuracy: 0.9911
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.18.0.dev0
+- Pytorch 1.10.0
+- Datasets 1.18.3
+- Tokenizers 0.11.0

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9911218410498034,
+    "eval_f1": 0.94980306712478,
+    "eval_loss": 0.0470016747713089,
+    "eval_precision": 0.9459188783174762,
+    "eval_recall": 0.9537192864355436,
+    "eval_runtime": 8.6425,
+    "eval_samples": 3251,
+    "eval_samples_per_second": 376.165,
+    "eval_steps_per_second": 47.093,
+    "train_loss": 0.05759347815719987,
+    "train_runtime": 571.906,
+    "train_samples": 14042,
+    "train_samples_per_second": 73.659,
+    "train_steps_per_second": 9.211
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9911218410498034,
+    "eval_f1": 0.94980306712478,
+    "eval_loss": 0.0470016747713089,
+    "eval_precision": 0.9459188783174762,
+    "eval_recall": 0.9537192864355436,
+    "eval_runtime": 8.6425,
+    "eval_samples": 3251,
+    "eval_samples_per_second": 376.165,
+    "eval_steps_per_second": 47.093
+}

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "xlm-roberta-base", "tokenizer_class": "XLMRobertaTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.05759347815719987,
+    "train_runtime": 571.906,
+    "train_samples": 14042,
+    "train_samples_per_second": 73.659,
+    "train_steps_per_second": 9.211
+}

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 5268,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.525436598329537e-05,
+      "loss": 0.2137,
+      "step": 500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.050873196659074e-05,
+      "loss": 0.0901,
+      "step": 1000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.5763097949886106e-05,
+      "loss": 0.0723,
+      "step": 1500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 3.1017463933181475e-05,
+      "loss": 0.0491,
+      "step": 2000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 2.6271829916476843e-05,
+      "loss": 0.0436,
+      "step": 2500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.152619589977221e-05,
+      "loss": 0.0397,
+      "step": 3000
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.678056188306758e-05,
+      "loss": 0.0331,
+      "step": 3500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 1.2034927866362947e-05,
+      "loss": 0.0241,
+      "step": 4000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 7.289293849658315e-06,
+      "loss": 0.0177,
+      "step": 4500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 2.5436598329536827e-06,
+      "loss": 0.0158,
+      "step": 5000
+    },
+    {
+      "epoch": 3.0,
+      "step": 5268,
+      "total_flos": 1035876517340328.0,
+      "train_loss": 0.05759347815719987,
+      "train_runtime": 571.906,
+      "train_samples_per_second": 73.659,
+      "train_steps_per_second": 9.211
+    }
+  ],
+  "max_steps": 5268,
+  "num_train_epochs": 3,
+  "total_flos": 1035876517340328.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0137cb859f2e54236b7d37db34b9bd8498374a82ac0a92769fb43e6cf2a8a6a2
+size 2991