Initial Commit

README.md

@@ -0,0 +1,74 @@
+---
+license: mit
+base_model: FacebookAI/xlm-roberta-base
+tags:
+- generated_from_trainer
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: scenario-kd-pre-ner-full-xlmr-halfen_data-univner_en66
+  results: []
+---
+
+<!-- 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. -->
+
+# scenario-kd-pre-ner-full-xlmr-halfen_data-univner_en66
+
+This model is a fine-tuned version of [FacebookAI/xlm-roberta-base](https://huggingface.co/FacebookAI/xlm-roberta-base) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 51.6812
+- Precision: 0.7610
+- Recall: 0.7712
+- F1: 0.7661
+- Accuracy: 0.9817
+
+## 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: 3e-05
+- train_batch_size: 8
+- eval_batch_size: 32
+- seed: 66
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 107.359       | 1.28  | 500  | 73.7771         | 0.7295    | 0.7008 | 0.7149 | 0.9786   |
+| 64.0888       | 2.55  | 1000 | 62.1069         | 0.7445    | 0.7298 | 0.7371 | 0.9808   |
+| 56.4531       | 3.83  | 1500 | 57.7062         | 0.7495    | 0.7340 | 0.7416 | 0.9805   |
+| 52.7856       | 5.1   | 2000 | 54.8858         | 0.7482    | 0.7598 | 0.7540 | 0.9814   |
+| 50.5437       | 6.38  | 2500 | 53.2901         | 0.7482    | 0.7505 | 0.7494 | 0.9811   |
+| 49.0825       | 7.65  | 3000 | 52.2441         | 0.7503    | 0.7557 | 0.7530 | 0.9815   |
+| 48.2494       | 8.93  | 3500 | 51.6812         | 0.7610    | 0.7712 | 0.7661 | 0.9817   |
+
+
+### Framework versions
+
+- Transformers 4.33.3
+- Pytorch 2.1.1+cu121
+- Datasets 2.14.5
+- Tokenizers 0.13.3

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "FacebookAI/xlm-roberta-base",
+  "architectures": [
+    "XLMRobertaForTokenClassificationKD"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

eval_result_ner.json

@@ -0,0 +1 @@
+{"ceb_gja": {"precision": 0.30612244897959184, "recall": 0.6122448979591837, "f1": 0.40816326530612246, "accuracy": 0.9135135135135135}, "en_pud": {"precision": 0.7570621468926554, "recall": 0.747906976744186, "f1": 0.7524567150210575, "accuracy": 0.9772383830751794}, "de_pud": {"precision": 0.7171620325982742, "recall": 0.7199230028873917, "f1": 0.7185398655139289, "accuracy": 0.9705123997937274}, "pt_pud": {"precision": 0.7593984962406015, "recall": 0.7352138307552321, "f1": 0.7471104946833103, "accuracy": 0.9766309223736489}, "ru_pud": {"precision": 0.6300692383778437, "recall": 0.6148648648648649, "f1": 0.6223742061553492, "accuracy": 0.9618703177473521}, "sv_pud": {"precision": 0.7959183673469388, "recall": 0.7580174927113703, "f1": 0.7765057242409158, "accuracy": 0.977877961836863}, "tl_trg": {"precision": 0.5882352941176471, "recall": 0.8695652173913043, "f1": 0.7017543859649124, "accuracy": 0.9782016348773842}, "tl_ugnayan": {"precision": 0.5454545454545454, "recall": 0.7272727272727273, "f1": 0.6233766233766234, "accuracy": 0.9699179580674567}, "zh_gsd": {"precision": 0.5703812316715543, "recall": 0.5071707953063885, "f1": 0.5369220151828847, "accuracy": 0.9423909423909423}, "zh_gsdsimp": {"precision": 0.5392441860465116, "recall": 0.48623853211009177, "f1": 0.5113714679531358, "accuracy": 0.9403096903096904}, "hr_set": {"precision": 0.7750188111361926, "recall": 0.7341411261582323, "f1": 0.7540263543191801, "accuracy": 0.9706512778235779}, "da_ddt": {"precision": 0.7530864197530864, "recall": 0.6823266219239373, "f1": 0.715962441314554, "accuracy": 0.9793475007482789}, "en_ewt": {"precision": 0.7727272727272727, "recall": 0.765625, "f1": 0.7691597414589104, "accuracy": 0.9775670398852453}, "pt_bosque": {"precision": 0.723205964585275, "recall": 0.6386831275720165, "f1": 0.6783216783216783, "accuracy": 0.9702579336328069}, "sr_set": {"precision": 0.797979797979798, "recall": 0.7461629279811098, "f1": 0.7712019524100062, "accuracy": 0.9648892391209176}, "sk_snk": {"precision": 0.6376811594202898, "recall": 0.5770491803278689, "f1": 0.6058519793459552, "accuracy": 0.946608040201005}, "sv_talbanken": {"precision": 0.7692307692307693, "recall": 0.8673469387755102, "f1": 0.8153477218225421, "accuracy": 0.9962703047553615}}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c51298260907efb0ad2f74629153dd9869bae5018cd7a98eb60bec37c754294a
+size 939760294

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f7898e2d99e3ac63cdefe684740b256c2e5deaf25bf7fd6d3ce32a2ca6acbbe
+size 4600