learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use adamw_torch 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: 1000
num_epochs: 10

Training Loss	Epoch	Step	Validation Loss	Accuracy	Micro Precision	Micro Recall	Micro F1	Macro Precision	Macro Recall	Macro F1
0.7154	1.0	4758	0.6927	0.7892	0.7892	0.7892	0.7892	0.7708	0.7432	0.7224
0.5336	2.0	9516	0.5828	0.8216	0.8216	0.8216	0.8216	0.8005	0.7771	0.7747
0.3957	3.0	14274	0.5817	0.8383	0.8383	0.8383	0.8383	0.8339	0.7910	0.7974
0.2801	4.0	19032	0.6542	0.8353	0.8353	0.8353	0.8353	0.8254	0.7975	0.8005
0.1928	5.0	23790	0.6737	0.8462	0.8462	0.8462	0.8462	0.8205	0.8195	0.8127
0.1219	6.0	28548	0.8407	0.8431	0.8431	0.8431	0.8431	0.8156	0.8227	0.8091
0.0668	7.0	33306	0.9256	0.8564	0.8564	0.8564	0.8564	0.8466	0.8332	0.8311
0.0329	8.0	38064	1.0479	0.8544	0.8544	0.8544	0.8544	0.8308	0.8427	0.8285
0.0145	9.0	42822	1.0892	0.8581	0.8581	0.8581	0.8581	0.8375	0.8493	0.8370
0.0083	10.0	47580	1.0962	0.8571	0.8571	0.8571	0.8571	0.8355	0.8465	0.8345

stulcrad/XLM-RoBERTa-PRETRAINED4-CERED3

XLM-RoBERTa-PRETRAINED4-CERED3