ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k5_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: 0.9630
• Qwk: 0.6500
• Mse: 0.9630
• Rmse: 0.9813

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	0.1	2	2.3859	0.0072	2.3859	1.5446
No log	0.2	4	1.5992	0.2149	1.5992	1.2646
No log	0.3	6	1.4759	0.0936	1.4759	1.2149
No log	0.4	8	1.6131	0.1827	1.6131	1.2701
No log	0.5	10	1.6897	0.2472	1.6897	1.2999
No log	0.6	12	1.8132	0.2789	1.8132	1.3466
No log	0.7	14	1.6192	0.3067	1.6192	1.2725
No log	0.8	16	1.3408	0.2326	1.3408	1.1579
No log	0.9	18	1.2313	0.2729	1.2313	1.1096
No log	1.0	20	1.2652	0.1865	1.2652	1.1248
No log	1.1	22	1.3484	0.2132	1.3484	1.1612
No log	1.2	24	1.4316	0.3282	1.4316	1.1965
No log	1.3	26	1.4458	0.4170	1.4458	1.2024
No log	1.4	28	1.4350	0.4393	1.4350	1.1979
No log	1.5	30	1.2673	0.3994	1.2673	1.1257
No log	1.6	32	1.2046	0.4109	1.2046	1.0976
No log	1.7	34	1.1748	0.4754	1.1748	1.0839
No log	1.8	36	1.0393	0.4166	1.0393	1.0195
No log	1.9	38	0.9877	0.4389	0.9877	0.9938
No log	2.0	40	0.9554	0.5204	0.9554	0.9774
No log	2.1	42	0.9378	0.5308	0.9378	0.9684
No log	2.2	44	0.9363	0.5437	0.9363	0.9676
No log	2.3	46	0.9626	0.5724	0.9626	0.9811
No log	2.4	48	0.9242	0.5681	0.9242	0.9613
No log	2.5	50	0.9259	0.5575	0.9259	0.9622
No log	2.6	52	1.0395	0.5580	1.0395	1.0195
No log	2.7	54	1.0156	0.5580	1.0156	1.0078
No log	2.8	56	0.9720	0.5729	0.9720	0.9859
No log	2.9	58	0.9466	0.5479	0.9466	0.9729
No log	3.0	60	0.9445	0.5576	0.9445	0.9718
No log	3.1	62	0.9324	0.5519	0.9324	0.9656
No log	3.2	64	0.9355	0.5741	0.9355	0.9672
No log	3.3	66	0.9312	0.5818	0.9312	0.9650
No log	3.4	68	1.1087	0.5302	1.1087	1.0529
No log	3.5	70	1.2534	0.4806	1.2534	1.1195
No log	3.6	72	1.1622	0.5059	1.1622	1.0781
No log	3.7	74	1.0616	0.5355	1.0616	1.0303
No log	3.8	76	0.8959	0.6048	0.8959	0.9465
No log	3.9	78	0.8535	0.6197	0.8535	0.9239
No log	4.0	80	0.9748	0.6057	0.9748	0.9873
No log	4.1	82	1.2370	0.5151	1.2370	1.1122
No log	4.2	84	1.2585	0.5175	1.2585	1.1218
No log	4.3	86	1.0706	0.6199	1.0706	1.0347
No log	4.4	88	0.9442	0.6040	0.9442	0.9717
No log	4.5	90	0.9065	0.6429	0.9065	0.9521
No log	4.6	92	0.8442	0.6676	0.8442	0.9188
No log	4.7	94	0.8155	0.6638	0.8155	0.9031
No log	4.8	96	0.7726	0.6675	0.7726	0.8790
No log	4.9	98	0.9287	0.6653	0.9287	0.9637
No log	5.0	100	1.1187	0.6480	1.1187	1.0577
No log	5.1	102	1.2398	0.6049	1.2398	1.1135
No log	5.2	104	1.1418	0.6154	1.1418	1.0685
No log	5.3	106	0.9132	0.6562	0.9132	0.9556
No log	5.4	108	0.8386	0.6541	0.8386	0.9158
No log	5.5	110	0.7385	0.6703	0.7385	0.8594
No log	5.6	112	0.7469	0.6703	0.7469	0.8642
No log	5.7	114	0.9043	0.6521	0.9043	0.9509
No log	5.8	116	1.1678	0.6076	1.1678	1.0807
No log	5.9	118	1.5330	0.5358	1.5330	1.2382
No log	6.0	120	1.6834	0.5123	1.6834	1.2975
No log	6.1	122	1.5150	0.5387	1.5150	1.2309
No log	6.2	124	1.1426	0.6182	1.1426	1.0689
No log	6.3	126	0.8874	0.6590	0.8874	0.9420
No log	6.4	128	0.8601	0.6734	0.8601	0.9274
No log	6.5	130	0.9788	0.6421	0.9788	0.9894
No log	6.6	132	1.2205	0.5996	1.2205	1.1047
No log	6.7	134	1.3857	0.5509	1.3857	1.1772
No log	6.8	136	1.3853	0.5412	1.3853	1.1770
No log	6.9	138	1.2685	0.5805	1.2685	1.1263
No log	7.0	140	1.0411	0.6357	1.0411	1.0203
No log	7.1	142	0.8856	0.6643	0.8856	0.9410
No log	7.2	144	0.8601	0.6641	0.8601	0.9274
No log	7.3	146	0.8935	0.6963	0.8935	0.9452
No log	7.4	148	0.8756	0.6720	0.8756	0.9358
No log	7.5	150	0.8444	0.6683	0.8444	0.9189
No log	7.6	152	0.8512	0.6691	0.8512	0.9226
No log	7.7	154	0.8936	0.6442	0.8936	0.9453
No log	7.8	156	0.9734	0.6474	0.9734	0.9866
No log	7.9	158	1.0758	0.6488	1.0758	1.0372
No log	8.0	160	1.1103	0.6406	1.1103	1.0537
No log	8.1	162	1.1033	0.6488	1.1033	1.0504
No log	8.2	164	1.0394	0.6538	1.0394	1.0195
No log	8.3	166	1.0313	0.6538	1.0313	1.0156
No log	8.4	168	0.9729	0.6450	0.9729	0.9863
No log	8.5	170	0.9440	0.6436	0.9440	0.9716
No log	8.6	172	0.9060	0.6668	0.9060	0.9519
No log	8.7	174	0.8640	0.6494	0.8640	0.9295
No log	8.8	176	0.8721	0.6625	0.8721	0.9339
No log	8.9	178	0.9020	0.6575	0.9020	0.9497
No log	9.0	180	0.9117	0.6559	0.9117	0.9548
No log	9.1	182	0.9104	0.6559	0.9104	0.9541
No log	9.2	184	0.9189	0.6467	0.9189	0.9586
No log	9.3	186	0.9146	0.6636	0.9146	0.9563
No log	9.4	188	0.9167	0.6636	0.9167	0.9574
No log	9.5	190	0.9220	0.6501	0.9220	0.9602
No log	9.6	192	0.9286	0.6432	0.9286	0.9636
No log	9.7	194	0.9443	0.6432	0.9443	0.9717
No log	9.8	196	0.9522	0.6500	0.9522	0.9758
No log	9.9	198	0.9606	0.6500	0.9606	0.9801
No log	10.0	200	0.9630	0.6500	0.9630	0.9813

Framework versions

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