finetuned-parsbert-uncased-ArmanEmo

This model is a fine-tuned version of HooshvareLab/bert-base-parsbert-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0060
• Accuracy: 1.0
• Precision Macro: 1.0
• Recall Macro: 1.0
• F1 Macro: 1.0
• F1 C0: 1.0
• F1 C1: 1.0
• F1 C2: 1.0
• F1 C3: 1.0
• F1 C4: 1.0
• F1 C5: 1.0
• F1 C6: 1.0
• Recall C0: 1.0
• Recall C1: 1.0
• Recall C2: 1.0
• Recall C3: 1.0
• Recall C4: 1.0
• Recall C5: 1.0
• Recall C6: 1.0
• Precision C0: 1.0
• Precision C1: 1.0
• Precision C2: 1.0
• Precision C3: 1.0
• Precision C4: 1.0
• Precision C5: 1.0
• Precision C6: 1.0

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	F1 C0	F1 C1	F1 C2	F1 C3	F1 C4	F1 C5	F1 C6	Recall C0	Recall C1	Recall C2	Recall C3	Recall C4	Recall C5	Recall C6	Precision C0	Precision C1	Precision C2	Precision C3	Precision C4	Precision C5	Precision C6
No log	1.0	144	0.3531	0.5065	0.3687	0.3986	0.3583	0.6766	0.3370	0.4966	0.0	0.4164	0.0	0.5817	0.9091	0.2366	0.7143	0.0	0.3862	0.0	0.5440	0.5388	0.5849	0.3806	0.0	0.4516	0.0	0.625
No log	2.0	288	0.2289	0.7298	0.5210	0.5631	0.5333	0.8451	0.7790	0.6063	0.0	0.6905	0.0	0.8120	0.9818	0.8206	0.5	0.0	0.8	0.0	0.8394	0.7418	0.7414	0.77	0.0	0.6073	0.0	0.7864
No log	3.0	432	0.1129	0.9270	0.9239	0.8955	0.9061	0.9762	0.9492	0.8910	0.9027	0.8792	0.8108	0.9333	0.9709	0.9618	0.9026	0.8947	0.9034	0.6923	0.9430	0.9816	0.9368	0.8797	0.9107	0.8562	0.9783	0.9239
0.3027	4.0	576	0.0567	0.9652	0.9491	0.9545	0.9507	0.9909	0.9847	0.9161	0.9402	0.9565	0.8872	0.9796	0.9927	0.9847	0.8506	0.9649	0.9862	0.9077	0.9948	0.9891	0.9847	0.9924	0.9167	0.9286	0.8676	0.9648
0.3027	5.0	720	0.0296	0.9844	0.9736	0.9819	0.9776	0.9964	0.9885	0.9673	0.9913	0.9862	0.9185	0.9948	0.9927	0.9847	0.9610	1.0	0.9862	0.9538	0.9948	1.0	0.9923	0.9737	0.9828	0.9862	0.8857	0.9948
0.3027	6.0	864	0.0144	0.9965	0.9939	0.9937	0.9938	0.9982	1.0	0.9935	0.9913	0.9965	0.9767	1.0	1.0	1.0	0.9935	1.0	0.9931	0.9692	1.0	0.9964	1.0	0.9935	0.9828	1.0	0.9844	1.0
0.0482	7.0	1008	0.0088	0.9991	0.9993	0.9990	0.9991	1.0	1.0	1.0	1.0	0.9965	1.0	0.9974	1.0	1.0	1.0	1.0	0.9931	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	0.9948
0.0482	8.0	1152	0.0069	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
0.0482	9.0	1296	0.0063	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
0.0482	10.0	1440	0.0060	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2