learning_rate: 0.00015
train_batch_size: 64
eval_batch_size: 1024
seed: 2
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: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 5
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Mae	Rounded Mae
0.3623	0.1411	500	0.2169	0.5747	0.4917
0.2310	0.2822	1000	0.1939	0.5188	0.4427
0.2933	0.4233	1500	0.2048	0.5561	0.4650
0.2215	0.5643	2000	0.1852	0.4930	0.4254
0.2097	0.7054	2500	0.1951	0.5325	0.5121
0.2067	0.8465	3000	0.2008	0.5133	0.4467
0.3312	0.9876	3500	0.7125	1.2558	1.2107
0.2876	1.1287	4000	0.2389	0.5945	0.5495
0.4925	1.2698	4500	0.1948	0.5244	0.4437
0.2557	1.4108	5000	0.1942	0.5242	0.4515
0.2617	1.5519	5500	0.1922	0.5237	0.4388
0.2471	1.6930	6000	0.1810	0.4922	0.4151
0.2432	1.8341	6500	0.1781	0.4740	0.4141
0.1809	1.9752	7000	0.4463	0.8649	0.8574
0.3098	2.1163	7500	0.1808	0.4988	0.4184
0.1929	2.2573	8000	0.5667	1.0336	1.0250
0.2770	2.3984	8500	0.5864	1.0561	1.0060
0.3224	2.5395	9000	0.1744	0.4812	0.4069
0.1696	2.6806	9500	0.1700	0.4626	0.4043
0.2389	2.8217	10000	0.1696	0.4644	0.3995
0.2788	2.9628	10500	0.2998	0.6529	0.5941
0.3413	3.1038	11000	0.1919	0.5414	0.5256
0.3397	3.2449	11500	0.1986	0.5256	0.4876
0.3546	3.3860	12000	0.5394	1.0256	0.8448
0.2682	3.5271	12500	0.5264	0.9696	0.8949
0.2849	3.6682	13000	0.3474	0.7435	0.6889
0.3388	3.8093	13500	0.5434	1.0155	0.9122
0.2947	3.9503	14000	0.1872	0.5054	0.4900
0.3379	4.0914	14500	0.3559	0.7390	0.6704
0.3151	4.2325	15000	0.5394	1.0020	1.0125
0.4669	4.3736	15500	0.8444	1.4326	1.3828
0.5330	4.5147	16000	0.3602	0.7536	0.7432
0.5989	4.6558	16500	0.5220	0.9786	0.9442
0.5571	4.7968	17000	0.9736	1.5950	1.5801
0.8639	4.9379	17500	0.9659	1.5901	1.5486
0.8639	5.0	17720	0.9641	1.5879	1.5487

MichaHenh/cil-sentiment-anylsis-seed2

cil-sentiment-anylsis-seed2