README.md

metadata

license: mit
base_model: xlnet/xlnet-base-cased
tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: xlnet-base-cased-airlines-news-multi-label-8-actions
    results: []

xlnet-base-cased-airlines-news-multi-label-8-actions

This model is a fine-tuned version of xlnet/xlnet-base-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1864
• F1: 0.9349
• Jaccard: 0.6652

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 150
• num_epochs: 45

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Jaccard
No log	1.0	57	0.3740	0.8583	0.4115
No log	2.0	114	0.2826	0.8583	0.4115
No log	3.0	171	0.2538	0.8890	0.4558
No log	4.0	228	0.2369	0.8936	0.5007
No log	5.0	285	0.2252	0.9036	0.5236
No log	6.0	342	0.2181	0.9070	0.5450
No log	7.0	399	0.2147	0.9168	0.5793
No log	8.0	456	0.2105	0.9172	0.5653
0.3052	9.0	513	0.2058	0.9199	0.5874
0.3052	10.0	570	0.2046	0.9204	0.6040
0.3052	11.0	627	0.2018	0.9214	0.5962
0.3052	12.0	684	0.1990	0.9242	0.6132
0.3052	13.0	741	0.1987	0.9223	0.6103
0.3052	14.0	798	0.1974	0.9235	0.6191
0.3052	15.0	855	0.1962	0.9213	0.6169
0.3052	16.0	912	0.1958	0.9255	0.6235
0.3052	17.0	969	0.1932	0.9264	0.6176
0.2221	18.0	1026	0.1927	0.9276	0.6423
0.2221	19.0	1083	0.1926	0.9279	0.6338
0.2221	20.0	1140	0.1910	0.9288	0.6434
0.2221	21.0	1197	0.1924	0.9271	0.6316
0.2221	22.0	1254	0.1904	0.9285	0.6353
0.2221	23.0	1311	0.1883	0.9288	0.6475
0.2221	24.0	1368	0.1877	0.9302	0.6504
0.2221	25.0	1425	0.1890	0.9291	0.6442
0.2221	26.0	1482	0.1878	0.9318	0.6659
0.2088	27.0	1539	0.1882	0.9308	0.6593
0.2088	28.0	1596	0.1867	0.9347	0.6597
0.2088	29.0	1653	0.1864	0.9349	0.6652
0.2088	30.0	1710	0.1866	0.9345	0.6681
0.2088	31.0	1767	0.1871	0.9341	0.6670
0.2088	32.0	1824	0.1862	0.9324	0.6622
0.2088	33.0	1881	0.1878	0.9325	0.6589
0.2088	34.0	1938	0.1866	0.9332	0.6633
0.2088	35.0	1995	0.1858	0.9330	0.6674
0.2021	36.0	2052	0.1863	0.9298	0.6479
0.2021	37.0	2109	0.1858	0.9325	0.6630
0.2021	38.0	2166	0.1861	0.9320	0.6652
0.2021	39.0	2223	0.1854	0.9325	0.6652
0.2021	40.0	2280	0.1852	0.9330	0.6674
0.2021	41.0	2337	0.1856	0.9318	0.6608
0.2021	42.0	2394	0.1857	0.9318	0.6608
0.2021	43.0	2451	0.1856	0.9324	0.6652
0.198	44.0	2508	0.1855	0.9325	0.6652
0.198	45.0	2565	0.1855	0.9325	0.6652

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1