metadata

license: apache-2.0
base_model: bert-large-uncased
tags:
  - generated_from_trainer
datasets:
  - wnut_17
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: sample_data
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: wnut_17
          type: wnut_17
          config: wnut_17
          split: validation
          args: wnut_17
        metrics:
          - name: Precision
            type: precision
            value: 0.7290715372907154
          - name: Recall
            type: recall
            value: 0.5729665071770335
          - name: F1
            type: f1
            value: 0.6416610850636303
          - name: Accuracy
            type: accuracy
            value: 0.9602644796236252

sample_data

This model is a fine-tuned version of bert-large-uncased on the wnut_17 dataset. It achieves the following results on the evaluation set:

Loss: 0.2684
Precision: 0.7291
Recall: 0.5730
F1: 0.6417
Accuracy: 0.9603

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: 2e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6403	0.12	25	0.4914	0.0	0.0	0.0	0.9205
0.334	0.23	50	0.4539	0.0	0.0	0.0	0.9205
0.2346	0.35	75	0.3556	0.4118	0.0419	0.0760	0.9236
0.2352	0.47	100	0.2936	0.4337	0.2464	0.3143	0.9341
0.1725	0.59	125	0.2898	0.4983	0.3421	0.4057	0.9372
0.1449	0.7	150	0.2858	0.4606	0.3493	0.3973	0.9399
0.1548	0.82	175	0.2487	0.5699	0.3900	0.4631	0.9435
0.1429	0.94	200	0.3071	0.6888	0.3469	0.4614	0.9415
0.1506	1.06	225	0.2252	0.4820	0.4952	0.4885	0.9465
0.1196	1.17	250	0.2512	0.5463	0.4940	0.5188	0.9485
0.1062	1.29	275	0.2916	0.6395	0.4605	0.5355	0.9495
0.0983	1.41	300	0.2402	0.6199	0.5443	0.5796	0.9497
0.1068	1.53	325	0.2470	0.6018	0.4773	0.5324	0.9504
0.0879	1.64	350	0.2360	0.6468	0.5586	0.5995	0.9511
0.0928	1.76	375	0.2267	0.6126	0.5467	0.5777	0.9514
0.1045	1.88	400	0.2258	0.6934	0.5060	0.5851	0.9542
0.0933	2.0	425	0.2403	0.6954	0.5108	0.5890	0.9547
0.0497	2.11	450	0.2539	0.6460	0.5371	0.5865	0.9554
0.0607	2.23	475	0.3065	0.7293	0.4737	0.5743	0.9523
0.0857	2.35	500	0.2565	0.6770	0.4964	0.5728	0.9545
0.0513	2.46	525	0.2569	0.6931	0.5323	0.6022	0.9569
0.0697	2.58	550	0.2273	0.7193	0.5670	0.6341	0.9566
0.0446	2.7	575	0.2361	0.6348	0.5634	0.5970	0.9580
0.0498	2.82	600	0.2544	0.7109	0.5323	0.6088	0.9579
0.0464	2.93	625	0.2576	0.7237	0.5514	0.6259	0.9589
0.0441	3.05	650	0.2691	0.7321	0.5490	0.6275	0.9586
0.0524	3.17	675	0.2368	0.6947	0.5825	0.6337	0.9603
0.0335	3.29	700	0.2488	0.6991	0.5670	0.6262	0.9594
0.0349	3.4	725	0.2564	0.7084	0.5347	0.6094	0.9580
0.026	3.52	750	0.2523	0.7085	0.5610	0.6262	0.9594
0.0314	3.64	775	0.2647	0.7335	0.5467	0.6265	0.9584
0.0213	3.76	800	0.2551	0.7032	0.5754	0.6329	0.9603
0.0312	3.87	825	0.2470	0.7034	0.5957	0.6451	0.9606
0.0313	3.99	850	0.2693	0.7421	0.5610	0.6390	0.9598
0.0243	4.11	875	0.2699	0.7345	0.5658	0.6392	0.9598
0.0289	4.23	900	0.2535	0.7143	0.5682	0.6329	0.9603
0.0226	4.34	925	0.2581	0.7205	0.5706	0.6368	0.9602
0.0173	4.46	950	0.2644	0.7145	0.5718	0.6352	0.9601
0.0139	4.58	975	0.2705	0.7164	0.5682	0.6338	0.9600
0.0243	4.69	1000	0.2615	0.7116	0.5813	0.6399	0.9606
0.0222	4.81	1025	0.2642	0.7229	0.5742	0.64	0.9606
0.0112	4.93	1050	0.2684	0.7291	0.5730	0.6417	0.9603

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3