learning_rate: 5e-05
train_batch_size: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Chrf++	Gen Len
46.9506	1.5161	6	2.7928	3.1601	190.25
46.9506	3.0	12	2.7878	2.6661	190.0625
46.9506	4.5161	18	2.7874	2.7449	189.875
46.9506	6.0	24	2.7829	2.9496	190.0625
46.9506	7.5161	30	2.7773	2.6831	190.25
46.9506	9.0	36	2.7763	2.7939	189.6875
46.9506	10.5161	42	2.7674	3.5057	189.625
46.9506	12.0	48	2.7546	3.1023	189.5625
46.9506	13.5161	54	2.7320	3.4061	189.625
46.9506	15.0	60	2.7042	2.7326	189.25
46.9506	16.5161	66	2.6829	3.1676	189.375
46.9506	18.0	72	2.6557	3.9708	189.0625
46.9506	19.5161	78	2.6299	4.5884	188.8125
46.9506	21.0	84	2.6178	4.2387	189.375
46.9506	22.5161	90	2.5844	5.2072	187.0
46.9506	24.0	96	2.5526	5.1571	188.4375
36.6303	25.5161	102	2.5161	5.1543	187.0
36.6303	27.0	108	2.4867	6.575	187.125
36.6303	28.5161	114	2.4595	6.6961	185.125
36.6303	30.0	120	2.4282	7.1433	187.0
36.6303	31.5161	126	2.4018	7.8063	187.0625
36.6303	33.0	132	2.3662	6.919	186.3125
36.6303	34.5161	138	2.3303	7.8335	187.375
36.6303	36.0	144	2.2944	7.9781	187.25
36.6303	37.5161	150	2.2623	7.954	185.75
36.6303	39.0	156	2.2320	8.5663	184.875
36.6303	40.5161	162	2.1724	8.7452	184.9375
36.6303	42.0	168	2.1573	7.7747	181.4375
36.6303	43.5161	174	2.1171	8.2491	180.0625
36.6303	45.0	180	2.0300	8.0288	175.1875
36.6303	46.5161	186	1.9725	9.633	180.125
36.6303	48.0	192	1.8980	9.719	181.8125
36.6303	49.5161	198	1.8246	9.5178	170.375
21.6742	51.0	204	1.7495	11.0916	168.0
21.6742	52.5161	210	1.6860	10.2375	169.0
21.6742	54.0	216	1.6192	11.7165	163.125
21.6742	55.5161	222	1.5772	10.9687	155.3125
21.6742	57.0	228	1.5176	13.5556	158.3125
21.6742	58.5161	234	1.4375	13.1794	135.0625
21.6742	60.0	240	1.3399	18.016	145.0625
21.6742	61.5161	246	1.2768	19.0614	112.3125
21.6742	63.0	252	1.1939	25.9686	88.8125
21.6742	64.5161	258	1.1055	29.6382	63.3125
21.6742	66.0	264	1.0491	34.0179	61.0625
21.6742	67.5161	270	0.9934	36.4617	66.875
21.6742	69.0	276	0.9478	38.3981	65.3125
21.6742	70.5161	282	0.9051	39.9668	59.25
21.6742	72.0	288	0.8738	40.837	62.25
21.6742	73.5161	294	0.8304	42.6046	63.0625
9.9293	75.0	300	0.8202	42.2442	60.9375
9.9293	76.5161	306	0.8143	42.6334	62.9375
9.9293	78.0	312	0.8019	44.2055	64.9375
9.9293	79.5161	318	0.7778	43.3095	63.75
9.9293	81.0	324	0.7382	46.9486	66.6875
9.9293	82.5161	330	0.7213	44.9093	63.6875
9.9293	84.0	336	0.7042	47.5838	68.125
9.9293	85.5161	342	0.6685	46.3312	66.4375
9.9293	87.0	348	0.6568	48.3356	65.75
9.9293	88.5161	354	0.6659	49.6912	69.4375
9.9293	90.0	360	0.6621	49.7831	70.3125
9.9293	91.5161	366	0.6563	49.4391	69.25
9.9293	93.0	372	0.6668	49.8116	70.875
9.9293	94.5161	378	0.7086	51.4134	71.375
9.9293	96.0	384	0.7165	50.58	69.75
9.9293	97.5161	390	0.7013	50.6562	71.0
9.9293	99.0	396	0.6794	50.7295	68.0
2.9055	100.0	400	0.6828	51.0762	70.1875

Framework versions

Transformers 5.12.1
Pytorch 2.6.0+cu124
Datasets 5.0.0
Tokenizers 0.22.2

Tsedeniya/finetuned_byT5large_pan_joint

finetuned_byT5large_pan_joint

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Tsedeniya/
finetuned_byT5large_pan_joint