learning_rate: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Validation Loss
4.1053	1.0	57	0.0	0.0	0.0	0.0	0.06	0.3402
0.5072	2.0	114	24.9319	4.5161	24.9296	24.6395	121.94	0.2054
0.3207	3.0	171	27.9657	6.3338	27.7444	27.4873	124.36	0.1624
0.2669	4.0	228	28.3503	6.1986	28.1966	28.2528	124.48	0.1468
0.2171	5.0	285	29.5230	7.1443	29.5008	29.3333	124.5	0.1369
0.1792	6.0	342	32.1619	7.5140	31.9657	31.4165	124.48	0.1314
0.177	7.0	399	31.6686	6.8931	31.4190	31.3302	126.96	0.1256
0.1399	8.0	456	29.5794	5.7714	29.6270	29.2778	124.48	0.1282
0.1229	9.0	513	27.3419	5.5325	27.5182	27.3056	124.5	0.1211
0.1308	10.0	570	26.7754	5.9141	26.6904	26.5173	124.46	0.1259
0.1164	11.0	627	29.6974	6.5350	29.9128	29.5000	124.48	0.1207
0.0994	12.0	684	29.6974	6.5350	29.8615	29.4308	123.98	0.1187
0.1116	13.0	741	27.7020	6.8939	27.5474	27.4379	126.98	0.1227
0.1096	14.0	798	29.3949	5.6064	29.4974	29.0667	124.5	0.1217
0.0968	15.0	855	29.1429	6.3077	29.3641	28.8260	124.5	0.1209
0.0908	16.0	912	27.2810	5.7125	27.2952	27.2857	124.5	0.1227
0.0839	17.0	969	28.1333	7.6077	27.8667	27.8667	124.5	0.1216
0.0839	18.0	1026	28.1333	7.6077	27.8667	27.8667	124.5	0.1219
0.0979	19.0	1083	27.7333	7.1192	27.5333	27.5667	124.5	0.1214
0.0868	20.0	1140	27.6308	7.0773	27.4000	27.4282	124.5	0.1210

uzorovK/ruT5_hd_tickets_cleaner

ruT5_hd_tickets_cleaner