Initial Commit

README.md

@@ -0,0 +1,140 @@
+---
+library_name: transformers
+license: mit
+base_model: microsoft/mdeberta-v3-base
+tags:
+- generated_from_trainer
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: scenario-non-kd-scr-ner-half-mdeberta_data-univner_full55
+  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-non-kd-scr-ner-half-mdeberta_data-univner_full55
+
+This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3355
+- Precision: 0.6133
+- Recall: 0.5878
+- F1: 0.6003
+- Accuracy: 0.9612
+
+## 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: 32
+- eval_batch_size: 32
+- seed: 55
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 30
+
+### Training results
+
+| Training Loss | Epoch   | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.3541        | 0.2910  | 500   | 0.2833          | 0.3113    | 0.1163 | 0.1693 | 0.9287   |
+| 0.2371        | 0.5821  | 1000  | 0.2120          | 0.3611    | 0.2493 | 0.2950 | 0.9382   |
+| 0.1767        | 0.8731  | 1500  | 0.1741          | 0.4010    | 0.3865 | 0.3936 | 0.9464   |
+| 0.1351        | 1.1641  | 2000  | 0.1657          | 0.4667    | 0.4367 | 0.4512 | 0.9511   |
+| 0.1102        | 1.4552  | 2500  | 0.1532          | 0.4867    | 0.5260 | 0.5056 | 0.9542   |
+| 0.1002        | 1.7462  | 3000  | 0.1520          | 0.5199    | 0.5269 | 0.5234 | 0.9565   |
+| 0.0899        | 2.0373  | 3500  | 0.1584          | 0.5512    | 0.5116 | 0.5307 | 0.9574   |
+| 0.0681        | 2.3283  | 4000  | 0.1583          | 0.5564    | 0.5308 | 0.5433 | 0.9583   |
+| 0.0647        | 2.6193  | 4500  | 0.1532          | 0.5687    | 0.5555 | 0.5620 | 0.9595   |
+| 0.0637        | 2.9104  | 5000  | 0.1519          | 0.5798    | 0.5726 | 0.5762 | 0.9605   |
+| 0.05          | 3.2014  | 5500  | 0.1600          | 0.5696    | 0.5904 | 0.5798 | 0.9596   |
+| 0.0449        | 3.4924  | 6000  | 0.1660          | 0.5790    | 0.5754 | 0.5772 | 0.9602   |
+| 0.0437        | 3.7835  | 6500  | 0.1589          | 0.5820    | 0.5885 | 0.5853 | 0.9610   |
+| 0.0396        | 4.0745  | 7000  | 0.1690          | 0.5779    | 0.5926 | 0.5851 | 0.9608   |
+| 0.0293        | 4.3655  | 7500  | 0.1803          | 0.5815    | 0.5845 | 0.5830 | 0.9609   |
+| 0.0325        | 4.6566  | 8000  | 0.1726          | 0.5927    | 0.5953 | 0.5940 | 0.9615   |
+| 0.0314        | 4.9476  | 8500  | 0.1743          | 0.5756    | 0.6148 | 0.5945 | 0.9604   |
+| 0.0215        | 5.2386  | 9000  | 0.1860          | 0.5725    | 0.6006 | 0.5863 | 0.9604   |
+| 0.0205        | 5.5297  | 9500  | 0.1973          | 0.5849    | 0.5838 | 0.5843 | 0.9604   |
+| 0.0223        | 5.8207  | 10000 | 0.1943          | 0.6066    | 0.5917 | 0.5990 | 0.9612   |
+| 0.0205        | 6.1118  | 10500 | 0.2040          | 0.6086    | 0.5868 | 0.5975 | 0.9611   |
+| 0.0155        | 6.4028  | 11000 | 0.2090          | 0.5869    | 0.5963 | 0.5916 | 0.9607   |
+| 0.0163        | 6.6938  | 11500 | 0.2104          | 0.5972    | 0.5874 | 0.5922 | 0.9610   |
+| 0.0168        | 6.9849  | 12000 | 0.2088          | 0.5784    | 0.5976 | 0.5879 | 0.9603   |
+| 0.0106        | 7.2759  | 12500 | 0.2262          | 0.5997    | 0.5872 | 0.5934 | 0.9605   |
+| 0.0114        | 7.5669  | 13000 | 0.2251          | 0.6102    | 0.5842 | 0.5969 | 0.9616   |
+| 0.0122        | 7.8580  | 13500 | 0.2244          | 0.5989    | 0.5940 | 0.5965 | 0.9606   |
+| 0.0111        | 8.1490  | 14000 | 0.2333          | 0.5996    | 0.5825 | 0.5909 | 0.9612   |
+| 0.0086        | 8.4400  | 14500 | 0.2320          | 0.5881    | 0.5960 | 0.5920 | 0.9603   |
+| 0.0089        | 8.7311  | 15000 | 0.2440          | 0.6076    | 0.5852 | 0.5962 | 0.9610   |
+| 0.0087        | 9.0221  | 15500 | 0.2407          | 0.5978    | 0.5897 | 0.5937 | 0.9612   |
+| 0.0065        | 9.3132  | 16000 | 0.2479          | 0.6046    | 0.5827 | 0.5934 | 0.9613   |
+| 0.0074        | 9.6042  | 16500 | 0.2458          | 0.6007    | 0.5864 | 0.5934 | 0.9609   |
+| 0.0076        | 9.8952  | 17000 | 0.2495          | 0.5944    | 0.5920 | 0.5932 | 0.9608   |
+| 0.0055        | 10.1863 | 17500 | 0.2517          | 0.6030    | 0.5933 | 0.5981 | 0.9610   |
+| 0.0048        | 10.4773 | 18000 | 0.2694          | 0.5975    | 0.5778 | 0.5875 | 0.9602   |
+| 0.0049        | 10.7683 | 18500 | 0.2642          | 0.6110    | 0.5882 | 0.5994 | 0.9608   |
+| 0.0061        | 11.0594 | 19000 | 0.2776          | 0.6150    | 0.5651 | 0.5890 | 0.9612   |
+| 0.0037        | 11.3504 | 19500 | 0.2723          | 0.6132    | 0.5842 | 0.5983 | 0.9613   |
+| 0.0045        | 11.6414 | 20000 | 0.2687          | 0.6065    | 0.5832 | 0.5946 | 0.9607   |
+| 0.0047        | 11.9325 | 20500 | 0.2776          | 0.6153    | 0.5673 | 0.5903 | 0.9610   |
+| 0.004         | 12.2235 | 21000 | 0.2806          | 0.6030    | 0.5763 | 0.5893 | 0.9612   |
+| 0.0033        | 12.5146 | 21500 | 0.2838          | 0.6173    | 0.5791 | 0.5976 | 0.9617   |
+| 0.0038        | 12.8056 | 22000 | 0.2884          | 0.6175    | 0.5705 | 0.5931 | 0.9611   |
+| 0.0034        | 13.0966 | 22500 | 0.2863          | 0.6082    | 0.5843 | 0.5960 | 0.9611   |
+| 0.0023        | 13.3877 | 23000 | 0.2905          | 0.6222    | 0.5806 | 0.6007 | 0.9618   |
+| 0.003         | 13.6787 | 23500 | 0.2897          | 0.6094    | 0.5885 | 0.5988 | 0.9612   |
+| 0.0034        | 13.9697 | 24000 | 0.2909          | 0.6126    | 0.5820 | 0.5969 | 0.9611   |
+| 0.0021        | 14.2608 | 24500 | 0.2951          | 0.5846    | 0.6029 | 0.5936 | 0.9604   |
+| 0.0028        | 14.5518 | 25000 | 0.2899          | 0.6086    | 0.5913 | 0.5998 | 0.9612   |
+| 0.0025        | 14.8428 | 25500 | 0.3014          | 0.6205    | 0.5719 | 0.5952 | 0.9610   |
+| 0.0024        | 15.1339 | 26000 | 0.3018          | 0.6173    | 0.5745 | 0.5951 | 0.9610   |
+| 0.0019        | 15.4249 | 26500 | 0.3058          | 0.6235    | 0.5738 | 0.5976 | 0.9614   |
+| 0.0021        | 15.7159 | 27000 | 0.3053          | 0.6220    | 0.5868 | 0.6039 | 0.9613   |
+| 0.0019        | 16.0070 | 27500 | 0.3142          | 0.6098    | 0.5689 | 0.5886 | 0.9608   |
+| 0.0018        | 16.2980 | 28000 | 0.2999          | 0.6057    | 0.5985 | 0.6021 | 0.9615   |
+| 0.0017        | 16.5891 | 28500 | 0.3096          | 0.6015    | 0.5822 | 0.5917 | 0.9605   |
+| 0.0017        | 16.8801 | 29000 | 0.3091          | 0.6159    | 0.5840 | 0.5995 | 0.9613   |
+| 0.0023        | 17.1711 | 29500 | 0.3051          | 0.6161    | 0.5913 | 0.6034 | 0.9615   |
+| 0.0012        | 17.4622 | 30000 | 0.3167          | 0.6283    | 0.5722 | 0.5990 | 0.9612   |
+| 0.0012        | 17.7532 | 30500 | 0.3246          | 0.6197    | 0.5682 | 0.5928 | 0.9612   |
+| 0.002         | 18.0442 | 31000 | 0.3197          | 0.6020    | 0.5887 | 0.5953 | 0.9608   |
+| 0.0013        | 18.3353 | 31500 | 0.3146          | 0.6031    | 0.5923 | 0.5977 | 0.9610   |
+| 0.0015        | 18.6263 | 32000 | 0.3228          | 0.6096    | 0.5827 | 0.5959 | 0.9612   |
+| 0.0011        | 18.9173 | 32500 | 0.3248          | 0.6178    | 0.5731 | 0.5946 | 0.9611   |
+| 0.0011        | 19.2084 | 33000 | 0.3195          | 0.6125    | 0.5904 | 0.6012 | 0.9611   |
+| 0.0011        | 19.4994 | 33500 | 0.3340          | 0.6205    | 0.5646 | 0.5912 | 0.9613   |
+| 0.0012        | 19.7905 | 34000 | 0.3270          | 0.6077    | 0.5839 | 0.5956 | 0.9612   |
+| 0.0012        | 20.0815 | 34500 | 0.3231          | 0.6135    | 0.5928 | 0.6030 | 0.9612   |
+| 0.0012        | 20.3725 | 35000 | 0.3282          | 0.6126    | 0.5803 | 0.5960 | 0.9612   |
+| 0.001         | 20.6636 | 35500 | 0.3340          | 0.5999    | 0.5851 | 0.5924 | 0.9605   |
+| 0.0009        | 20.9546 | 36000 | 0.3358          | 0.6126    | 0.5706 | 0.5909 | 0.9608   |
+| 0.001         | 21.2456 | 36500 | 0.3300          | 0.6039    | 0.5851 | 0.5943 | 0.9606   |
+| 0.0006        | 21.5367 | 37000 | 0.3355          | 0.6133    | 0.5878 | 0.6003 | 0.9612   |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.1.1+cu121
+- Datasets 2.14.5
+- Tokenizers 0.19.1

config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "microsoft/mdeberta-v3-base",
+  "architectures": [
+    "DebertaV2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "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": 1536,
+  "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-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 0,
+  "vocab_size": 251000
+}

eval_result_ner.json

@@ -0,0 +1 @@
+{"ceb_gja": {"precision": 0.25274725274725274, "recall": 0.46938775510204084, "f1": 0.32857142857142857, "accuracy": 0.9204633204633205}, "en_pud": {"precision": 0.4728682170542636, "recall": 0.3972093023255814, "f1": 0.43174924165824063, "accuracy": 0.9472988288628637}, "de_pud": {"precision": 0.14068054850177755, "recall": 0.2666025024061598, "f1": 0.1841755319148936, "accuracy": 0.8608597815386058}, "pt_pud": {"precision": 0.5847542627883651, "recall": 0.5304822565969063, "f1": 0.5562977099236641, "accuracy": 0.9609518520100825}, "ru_pud": {"precision": 0.019758507135016465, "recall": 0.08687258687258688, "f1": 0.03219459846181363, "accuracy": 0.59571170240248}, "sv_pud": {"precision": 0.4798061389337641, "recall": 0.2886297376093295, "f1": 0.3604368932038835, "accuracy": 0.943279513524848}, "tl_trg": {"precision": 0.2054794520547945, "recall": 0.6521739130434783, "f1": 0.3125, "accuracy": 0.9032697547683923}, "tl_ugnayan": {"precision": 0.07894736842105263, "recall": 0.2727272727272727, "f1": 0.12244897959183672, "accuracy": 0.8714676390154968}, "zh_gsd": {"precision": 0.5300613496932516, "recall": 0.5632333767926988, "f1": 0.5461441213653604, "accuracy": 0.9424741924741925}, "zh_gsdsimp": {"precision": 0.5666235446313066, "recall": 0.5740498034076016, "f1": 0.5703125, "accuracy": 0.9427239427239428}, "hr_set": {"precision": 0.7379061371841156, "recall": 0.7284390591589451, "f1": 0.7331420373027261, "accuracy": 0.97040395713108}, "da_ddt": {"precision": 0.6795580110497238, "recall": 0.5503355704697986, "f1": 0.6081582200247219, "accuracy": 0.9716651701087499}, "en_ewt": {"precision": 0.6033229491173416, "recall": 0.5340073529411765, "f1": 0.5665529010238908, "accuracy": 0.9583615571582261}, "pt_bosque": {"precision": 0.6476595744680851, "recall": 0.6263374485596708, "f1": 0.6368200836820085, "accuracy": 0.9669250833212578}, "sr_set": {"precision": 0.7925659472422062, "recall": 0.7804014167650531, "f1": 0.7864366448542534, "accuracy": 0.969442255494265}, "sk_snk": {"precision": 0.39658119658119656, "recall": 0.253551912568306, "f1": 0.3093333333333333, "accuracy": 0.9141802763819096}, "sv_talbanken": {"precision": 0.6534090909090909, "recall": 0.5867346938775511, "f1": 0.618279569892473, "accuracy": 0.9930804338224469}}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6adac80bd5d69dc916c84636812f19709a5af560b8490cd7625f3b4ab619bc8e
+size 428939068

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61147bd811ff164608f905bfb08fef21a8b921012605a9375ed95f20a59d60f6
+size 5304