led-risalah_data_v8

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0169
• Rouge1 Precision: 0.8329
• Rouge1 Recall: 0.135
• Rouge1 Fmeasure: 0.2293

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 Fmeasure	Rouge1 Precision	Rouge1 Recall
1.9061	1.0	15	1.9704	0.1528	0.5489	0.0894
1.8015	2.0	30	1.7979	0.2037	0.6934	0.1204
1.6484	3.0	45	1.7690	0.2107	0.72	0.1244
1.3656	4.0	60	1.7353	0.223	0.7526	0.1321
1.1833	5.0	75	1.7215	0.2172	0.7498	0.1283
1.1678	6.0	90	1.7365	0.2094	0.7063	0.1241
1.1258	7.0	105	1.7643	0.2193	0.7425	0.1299
1.0591	8.0	120	1.7697	0.2184	0.7328	0.1295
0.8896	9.0	135	1.7835	0.2207	0.7391	0.1306
1.0655	10.0	150	1.7985	0.2241	0.7559	0.1325
0.8386	11.0	165	1.8309	0.2217	0.7502	0.1314
0.8968	12.0	180	1.8377	0.2147	0.7179	0.1276
0.7863	13.0	195	1.8737	0.2172	0.7293	0.129
0.6942	14.0	210	1.8858	0.2185	0.7489	0.1291
0.6656	15.0	225	1.9181	0.2243	0.7566	0.1328
0.6672	16.0	240	1.9407	0.2224	0.7513	0.1315
0.6405	17.0	255	1.9416	0.2151	0.7369	0.1272
0.7382	18.0	270	1.9533	0.2214	0.7506	0.1311
0.6445	19.0	285	1.9605	0.2136	0.7292	0.1262

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1