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
    results: []

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

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: 0.2250
• F1: 0.9209
• Jaccard: 0.6829

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: 9e-05
• 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: 65

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Jaccard
No log	1.0	57	0.3815	0.8283	0.4558
No log	2.0	114	0.3186	0.8287	0.4602
No log	3.0	171	0.2839	0.8781	0.5133
No log	4.0	228	0.2660	0.8890	0.5457
No log	5.0	285	0.2532	0.8996	0.5833
No log	6.0	342	0.2471	0.8966	0.5745
No log	7.0	399	0.2412	0.9066	0.6069
No log	8.0	456	0.2393	0.9065	0.5981
0.323	9.0	513	0.2354	0.9043	0.6025
0.323	10.0	570	0.2340	0.9087	0.6077
0.323	11.0	627	0.2326	0.9122	0.6283
0.323	12.0	684	0.2305	0.9161	0.6401
0.323	13.0	741	0.2297	0.9113	0.6350
0.323	14.0	798	0.2284	0.9138	0.6416
0.323	15.0	855	0.2281	0.9130	0.6497
0.323	16.0	912	0.2248	0.9164	0.6527
0.323	17.0	969	0.2228	0.9166	0.6527
0.2463	18.0	1026	0.2232	0.9169	0.6586
0.2463	19.0	1083	0.2243	0.9162	0.6571
0.2463	20.0	1140	0.2236	0.9147	0.6519
0.2463	21.0	1197	0.2255	0.9203	0.6637
0.2463	22.0	1254	0.2261	0.9177	0.6556
0.2463	23.0	1311	0.2226	0.9169	0.6637
0.2463	24.0	1368	0.2226	0.9175	0.6718
0.2463	25.0	1425	0.2246	0.9147	0.6571
0.2463	26.0	1482	0.2240	0.9147	0.6637
0.2313	27.0	1539	0.2237	0.9164	0.6622
0.2313	28.0	1596	0.2235	0.9176	0.6711
0.2313	29.0	1653	0.2228	0.9154	0.6689
0.2313	30.0	1710	0.2220	0.9165	0.6748
0.2313	31.0	1767	0.2231	0.9168	0.6696
0.2313	32.0	1824	0.2231	0.9176	0.6718
0.2313	33.0	1881	0.2241	0.9166	0.6704
0.2313	34.0	1938	0.2224	0.9167	0.6704
0.2313	35.0	1995	0.2219	0.9168	0.6748
0.2248	36.0	2052	0.2250	0.9209	0.6829
0.2248	37.0	2109	0.2235	0.9139	0.6593
0.2248	38.0	2166	0.2226	0.9146	0.6659
0.2248	39.0	2223	0.2228	0.9176	0.6748
0.2248	40.0	2280	0.2227	0.9152	0.6637
0.2248	41.0	2337	0.2225	0.9151	0.6652
0.2248	42.0	2394	0.2224	0.9134	0.6593
0.2248	43.0	2451	0.2225	0.9175	0.6748
0.2201	44.0	2508	0.2219	0.9158	0.6637
0.2201	45.0	2565	0.2222	0.9151	0.6659
0.2201	46.0	2622	0.2211	0.9160	0.6681
0.2201	47.0	2679	0.2214	0.9167	0.6696
0.2201	48.0	2736	0.2218	0.9163	0.6681
0.2201	49.0	2793	0.2217	0.9146	0.6615
0.2201	50.0	2850	0.2216	0.9135	0.6593
0.2201	51.0	2907	0.2217	0.9174	0.6770
0.2201	52.0	2964	0.2219	0.9166	0.6755
0.2162	53.0	3021	0.2219	0.9165	0.6748
0.2162	54.0	3078	0.2214	0.9182	0.6814
0.2162	55.0	3135	0.2211	0.9165	0.6792
0.2162	56.0	3192	0.2211	0.9169	0.6748
0.2162	57.0	3249	0.2211	0.9149	0.6726
0.2162	58.0	3306	0.2209	0.9167	0.6814
0.2162	59.0	3363	0.2213	0.9167	0.6726
0.2162	60.0	3420	0.2215	0.9158	0.6726
0.2162	61.0	3477	0.2211	0.9150	0.6681
0.2157	62.0	3534	0.2211	0.9166	0.6748
0.2157	63.0	3591	0.2209	0.9157	0.6770
0.2157	64.0	3648	0.2209	0.9165	0.6748
0.2157	65.0	3705	0.2209	0.9157	0.6726

Framework versions

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