ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k4_task5_organization

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6792
• Qwk: 0.7559
• Mse: 0.6792
• Rmse: 0.8241

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: 2e-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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.1053	2	2.2320	-0.0078	2.2320	1.4940
No log	0.2105	4	1.4604	0.1325	1.4604	1.2085
No log	0.3158	6	1.4258	0.2134	1.4258	1.1941
No log	0.4211	8	1.5284	0.3938	1.5284	1.2363
No log	0.5263	10	1.3844	0.3691	1.3844	1.1766
No log	0.6316	12	1.4103	0.3534	1.4103	1.1876
No log	0.7368	14	1.2183	0.3541	1.2183	1.1037
No log	0.8421	16	1.1544	0.2139	1.1544	1.0744
No log	0.9474	18	1.1493	0.3102	1.1493	1.0720
No log	1.0526	20	1.0901	0.3097	1.0901	1.0441
No log	1.1579	22	1.0754	0.2463	1.0754	1.0370
No log	1.2632	24	1.0631	0.3383	1.0631	1.0311
No log	1.3684	26	1.0769	0.4795	1.0769	1.0377
No log	1.4737	28	1.0611	0.5047	1.0611	1.0301
No log	1.5789	30	1.0376	0.4940	1.0376	1.0186
No log	1.6842	32	1.1758	0.5027	1.1758	1.0843
No log	1.7895	34	1.1510	0.5642	1.1510	1.0728
No log	1.8947	36	0.9683	0.6303	0.9683	0.9840
No log	2.0	38	0.8612	0.6564	0.8612	0.9280
No log	2.1053	40	0.8513	0.6547	0.8513	0.9226
No log	2.2105	42	0.9727	0.6574	0.9727	0.9862
No log	2.3158	44	1.3013	0.6079	1.3013	1.1407
No log	2.4211	46	1.2123	0.6060	1.2123	1.1010
No log	2.5263	48	0.8464	0.7005	0.8464	0.9200
No log	2.6316	50	0.7224	0.6854	0.7224	0.8500
No log	2.7368	52	0.7155	0.6949	0.7155	0.8458
No log	2.8421	54	0.7429	0.7035	0.7429	0.8619
No log	2.9474	56	0.8062	0.7071	0.8062	0.8979
No log	3.0526	58	1.0663	0.6564	1.0663	1.0326
No log	3.1579	60	1.1508	0.6301	1.1508	1.0728
No log	3.2632	62	0.9309	0.6784	0.9309	0.9648
No log	3.3684	64	0.8031	0.7141	0.8031	0.8962
No log	3.4737	66	0.7468	0.7229	0.7468	0.8642
No log	3.5789	68	0.6594	0.7146	0.6594	0.8120
No log	3.6842	70	0.6563	0.7192	0.6563	0.8101
No log	3.7895	72	0.7464	0.7245	0.7464	0.8640
No log	3.8947	74	0.8272	0.7275	0.8272	0.9095
No log	4.0	76	0.9378	0.6710	0.9378	0.9684
No log	4.1053	78	0.8662	0.6861	0.8662	0.9307
No log	4.2105	80	0.7481	0.7257	0.7481	0.8649
No log	4.3158	82	0.7448	0.7417	0.7448	0.8630
No log	4.4211	84	0.8113	0.7155	0.8113	0.9007
No log	4.5263	86	0.9442	0.6741	0.9442	0.9717
No log	4.6316	88	0.8944	0.6615	0.8944	0.9457
No log	4.7368	90	0.7466	0.7460	0.7466	0.8641
No log	4.8421	92	0.7001	0.7487	0.7001	0.8367
No log	4.9474	94	0.6790	0.7040	0.6790	0.8240
No log	5.0526	96	0.6738	0.7002	0.6738	0.8208
No log	5.1579	98	0.7076	0.7568	0.7076	0.8412
No log	5.2632	100	0.8350	0.6838	0.8350	0.9138
No log	5.3684	102	0.8389	0.6799	0.8389	0.9159
No log	5.4737	104	0.7344	0.7267	0.7344	0.8570
No log	5.5789	106	0.6785	0.7529	0.6785	0.8237
No log	5.6842	108	0.6893	0.7516	0.6893	0.8302
No log	5.7895	110	0.7846	0.7018	0.7846	0.8858
No log	5.8947	112	0.8343	0.6613	0.8343	0.9134
No log	6.0	114	0.8862	0.6598	0.8862	0.9414
No log	6.1053	116	0.8469	0.6688	0.8469	0.9203
No log	6.2105	118	0.7310	0.7237	0.7310	0.8550
No log	6.3158	120	0.6330	0.7536	0.6330	0.7956
No log	6.4211	122	0.6180	0.7290	0.6180	0.7861
No log	6.5263	124	0.6186	0.7513	0.6186	0.7865
No log	6.6316	126	0.6522	0.7406	0.6522	0.8076
No log	6.7368	128	0.7373	0.7239	0.7373	0.8586
No log	6.8421	130	0.7537	0.7449	0.7537	0.8682
No log	6.9474	132	0.7781	0.7262	0.7781	0.8821
No log	7.0526	134	0.7522	0.7436	0.7522	0.8673
No log	7.1579	136	0.6819	0.7257	0.6819	0.8258
No log	7.2632	138	0.6585	0.76	0.6585	0.8115
No log	7.3684	140	0.6617	0.7534	0.6617	0.8134
No log	7.4737	142	0.6681	0.7534	0.6681	0.8174
No log	7.5789	144	0.7053	0.7430	0.7053	0.8398
No log	7.6842	146	0.7274	0.7533	0.7274	0.8529
No log	7.7895	148	0.7273	0.7555	0.7273	0.8528
No log	7.8947	150	0.7338	0.7494	0.7338	0.8566
No log	8.0	152	0.7157	0.7577	0.7157	0.8460
No log	8.1053	154	0.7071	0.7524	0.7071	0.8409
No log	8.2105	156	0.7063	0.7436	0.7063	0.8404
No log	8.3158	158	0.7040	0.7436	0.7040	0.8391
No log	8.4211	160	0.7238	0.7577	0.7238	0.8508
No log	8.5263	162	0.7433	0.7494	0.7433	0.8622
No log	8.6316	164	0.7392	0.7494	0.7392	0.8598
No log	8.7368	166	0.7183	0.7462	0.7183	0.8475
No log	8.8421	168	0.6944	0.7550	0.6944	0.8333
No log	8.9474	170	0.6719	0.7334	0.6719	0.8197
No log	9.0526	172	0.6620	0.7650	0.6620	0.8136
No log	9.1579	174	0.6649	0.7544	0.6649	0.8154
No log	9.2632	176	0.6668	0.7544	0.6668	0.8166
No log	9.3684	178	0.6670	0.7459	0.6670	0.8167
No log	9.4737	180	0.6631	0.7650	0.6631	0.8143
No log	9.5789	182	0.6628	0.7650	0.6628	0.8141
No log	9.6842	184	0.6671	0.7437	0.6671	0.8167
No log	9.7895	186	0.6734	0.7371	0.6734	0.8206
No log	9.8947	188	0.6779	0.7559	0.6779	0.8233
No log	10.0	190	0.6792	0.7559	0.6792	0.8241

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1