README.md

metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - crows_pairs
metrics:
  - accuracy
model-index:
  - name: xlnet-base-cased_crows_pairs_finetuned
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: crows_pairs
          type: crows_pairs
          config: crows_pairs
          split: test
          args: crows_pairs
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7119205298013245

xlnet-base-cased_crows_pairs_finetuned

This model is a fine-tuned version of xlnet-base-cased on the crows_pairs dataset. It achieves the following results on the evaluation set:

• Loss: 2.5652
• Accuracy: 0.7119

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: 64
• eval_batch_size: 64
• seed: 42
• 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	Accuracy
0.728	0.53	10	0.6939	0.4901
0.6914	1.05	20	0.6939	0.4901
0.705	1.58	30	0.6925	0.5066
0.6993	2.11	40	0.6949	0.5066
0.6979	2.63	50	0.6996	0.5066
0.7152	3.16	60	0.6940	0.4901
0.7158	3.68	70	0.7007	0.4934
0.6968	4.21	80	0.6999	0.5066
0.7164	4.74	90	0.6977	0.4934
0.6698	5.26	100	0.7079	0.4536
0.611	5.79	110	0.8882	0.5099
0.6487	6.32	120	0.8360	0.5066
0.5223	6.84	130	0.8047	0.5728
0.2879	7.37	140	1.1483	0.5795
0.2369	7.89	150	1.1773	0.5993
0.2542	8.42	160	0.9170	0.6424
0.1743	8.95	170	1.3674	0.6424
0.1307	9.47	180	1.0740	0.7152
0.0718	10.0	190	1.4397	0.6424
0.0278	10.53	200	1.9821	0.6523
0.0519	11.05	210	1.6970	0.6755
0.0269	11.58	220	1.8299	0.6656
0.0556	12.11	230	1.9459	0.7086
0.0455	12.63	240	1.6443	0.6854
0.0665	13.16	250	1.9887	0.6821
0.009	13.68	260	2.0236	0.6788
0.0146	14.21	270	1.8515	0.7152
0.0034	14.74	280	1.9315	0.7252
0.0248	15.26	290	2.0754	0.7119
0.0536	15.79	300	2.0371	0.7053
0.0393	16.32	310	1.9381	0.6987
0.0255	16.84	320	1.9074	0.6788
0.0116	17.37	330	2.2182	0.6623
0.0128	17.89	340	2.3002	0.6689
0.0006	18.42	350	2.2353	0.6788
0.0053	18.95	360	2.4277	0.6755
0.0013	19.47	370	2.5156	0.6490
0.0004	20.0	380	2.5091	0.6689
0.0003	20.53	390	2.4096	0.6854
0.0017	21.05	400	2.3497	0.6921
0.0001	21.58	410	2.3376	0.6854
0.012	22.11	420	2.3832	0.6854
0.0002	22.63	430	2.4388	0.7053
0.0001	23.16	440	2.4821	0.7152
0.0001	23.68	450	2.5027	0.7119
0.0001	24.21	460	2.5105	0.7152
0.0001	24.74	470	2.5145	0.7152
0.0002	25.26	480	2.5143	0.6954
0.0001	25.79	490	2.5629	0.6821
0.0002	26.32	500	2.5414	0.6887
0.0001	26.84	510	2.5301	0.7119
0.0012	27.37	520	2.5360	0.7020
0.0	27.89	530	2.5428	0.6921
0.0117	28.42	540	2.5455	0.6954
0.0001	28.95	550	2.5598	0.7086
0.0001	29.47	560	2.5648	0.7119
0.0001	30.0	570	2.5652	0.7119

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1
• Datasets 2.10.1
• Tokenizers 0.13.2