ArabicNewSplits6_FineTuningAraBERTFreeze_run3_AugV5_k1_task2_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: 0.9072
• Qwk: 0.5442
• Mse: 0.9072
• Rmse: 0.9525

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.5	2	7.1753	0.0035	7.1753	2.6787
No log	1.0	4	5.6321	0.0112	5.6321	2.3732
No log	1.5	6	4.6829	0.0184	4.6829	2.1640
No log	2.0	8	3.9550	0.0102	3.9550	1.9887
No log	2.5	10	3.3013	0.0063	3.3013	1.8169
No log	3.0	12	2.6944	-0.0077	2.6944	1.6415
No log	3.5	14	2.1630	-0.0755	2.1630	1.4707
No log	4.0	16	1.7015	-0.0355	1.7015	1.3044
No log	4.5	18	1.3645	0.0972	1.3645	1.1681
No log	5.0	20	1.0882	0.1349	1.0882	1.0432
No log	5.5	22	0.8113	0.2508	0.8113	0.9007
No log	6.0	24	0.6504	0.3060	0.6504	0.8065
No log	6.5	26	0.6020	0.3386	0.6020	0.7759
No log	7.0	28	0.5723	0.3757	0.5723	0.7565
No log	7.5	30	0.5714	0.3841	0.5714	0.7559
No log	8.0	32	0.5847	0.3759	0.5847	0.7647
No log	8.5	34	0.6110	0.4089	0.6110	0.7817
No log	9.0	36	0.6544	0.3870	0.6544	0.8089
No log	9.5	38	0.7153	0.4838	0.7153	0.8458
No log	10.0	40	0.6804	0.4947	0.6804	0.8248
No log	10.5	42	0.6303	0.4951	0.6303	0.7939
No log	11.0	44	0.6253	0.4942	0.6253	0.7907
No log	11.5	46	0.6164	0.4763	0.6164	0.7851
No log	12.0	48	0.6512	0.5176	0.6512	0.8070
No log	12.5	50	0.7854	0.5346	0.7854	0.8862
No log	13.0	52	0.8671	0.4696	0.8671	0.9312
No log	13.5	54	0.8152	0.5439	0.8152	0.9029
No log	14.0	56	0.7063	0.5267	0.7063	0.8404
No log	14.5	58	0.7121	0.5219	0.7121	0.8438
No log	15.0	60	0.7185	0.5219	0.7185	0.8476
No log	15.5	62	0.7069	0.5097	0.7069	0.8408
No log	16.0	64	0.7514	0.5854	0.7514	0.8668
No log	16.5	66	0.7763	0.5995	0.7763	0.8811
No log	17.0	68	0.7885	0.5942	0.7885	0.8880
No log	17.5	70	0.7683	0.5649	0.7683	0.8765
No log	18.0	72	0.7338	0.5019	0.7338	0.8566
No log	18.5	74	0.7262	0.5030	0.7262	0.8522
No log	19.0	76	0.7029	0.5030	0.7029	0.8384
No log	19.5	78	0.7009	0.5185	0.7009	0.8372
No log	20.0	80	0.7814	0.5832	0.7814	0.8840
No log	20.5	82	0.8410	0.5688	0.8410	0.9171
No log	21.0	84	0.8235	0.5647	0.8235	0.9075
No log	21.5	86	0.7813	0.5619	0.7813	0.8839
No log	22.0	88	0.7823	0.5405	0.7823	0.8845
No log	22.5	90	0.8045	0.5393	0.8045	0.8969
No log	23.0	92	0.8215	0.5404	0.8215	0.9063
No log	23.5	94	0.8538	0.5449	0.8538	0.9240
No log	24.0	96	0.8995	0.5437	0.8995	0.9484
No log	24.5	98	0.8771	0.5474	0.8771	0.9365
No log	25.0	100	0.8649	0.5186	0.8649	0.9300
No log	25.5	102	0.8682	0.5624	0.8682	0.9318
No log	26.0	104	0.8652	0.5413	0.8652	0.9302
No log	26.5	106	0.8543	0.5472	0.8543	0.9243
No log	27.0	108	0.8401	0.5316	0.8401	0.9166
No log	27.5	110	0.8413	0.5462	0.8413	0.9172
No log	28.0	112	0.8124	0.5889	0.8124	0.9013
No log	28.5	114	0.8094	0.5601	0.8094	0.8996
No log	29.0	116	0.8506	0.5444	0.8506	0.9223
No log	29.5	118	0.9106	0.5104	0.9106	0.9542
No log	30.0	120	0.8931	0.5343	0.8931	0.9451
No log	30.5	122	0.8817	0.5375	0.8817	0.9390
No log	31.0	124	0.8524	0.5637	0.8524	0.9232
No log	31.5	126	0.8394	0.5719	0.8394	0.9162
No log	32.0	128	0.8311	0.5717	0.8311	0.9117
No log	32.5	130	0.8445	0.5799	0.8445	0.9190
No log	33.0	132	0.8919	0.5441	0.8919	0.9444
No log	33.5	134	0.9039	0.5441	0.9039	0.9507
No log	34.0	136	0.8765	0.5621	0.8765	0.9362
No log	34.5	138	0.8764	0.5750	0.8764	0.9362
No log	35.0	140	0.8854	0.5692	0.8854	0.9410
No log	35.5	142	0.8798	0.5753	0.8798	0.9380
No log	36.0	144	0.9066	0.5526	0.9066	0.9522
No log	36.5	146	0.9819	0.5173	0.9819	0.9909
No log	37.0	148	0.9502	0.5273	0.9502	0.9748
No log	37.5	150	0.8850	0.5682	0.8850	0.9408
No log	38.0	152	0.8563	0.5745	0.8563	0.9254
No log	38.5	154	0.8739	0.5625	0.8739	0.9348
No log	39.0	156	0.9505	0.5616	0.9505	0.9750
No log	39.5	158	0.9863	0.5214	0.9863	0.9931
No log	40.0	160	0.9481	0.5461	0.9481	0.9737
No log	40.5	162	0.9059	0.5423	0.9059	0.9518
No log	41.0	164	0.9145	0.5482	0.9145	0.9563
No log	41.5	166	0.9749	0.5431	0.9749	0.9874
No log	42.0	168	1.0398	0.5096	1.0398	1.0197
No log	42.5	170	1.0444	0.4933	1.0444	1.0220
No log	43.0	172	1.0197	0.5050	1.0197	1.0098
No log	43.5	174	0.9561	0.5406	0.9561	0.9778
No log	44.0	176	0.9075	0.5729	0.9075	0.9526
No log	44.5	178	0.8831	0.5741	0.8831	0.9398
No log	45.0	180	0.9258	0.5721	0.9258	0.9622
No log	45.5	182	0.9113	0.5500	0.9113	0.9546
No log	46.0	184	0.9490	0.5422	0.9490	0.9742
No log	46.5	186	0.9889	0.5144	0.9889	0.9944
No log	47.0	188	0.9690	0.5356	0.9690	0.9844
No log	47.5	190	0.9189	0.5478	0.9189	0.9586
No log	48.0	192	0.8710	0.5654	0.8710	0.9333
No log	48.5	194	0.8391	0.5715	0.8391	0.9160
No log	49.0	196	0.8492	0.5715	0.8492	0.9215
No log	49.5	198	0.9053	0.5735	0.9053	0.9515
No log	50.0	200	0.9975	0.5243	0.9975	0.9987
No log	50.5	202	1.0129	0.5231	1.0129	1.0064
No log	51.0	204	0.9591	0.5726	0.9591	0.9793
No log	51.5	206	0.8853	0.5847	0.8853	0.9409
No log	52.0	208	0.8495	0.5402	0.8495	0.9217
No log	52.5	210	0.8446	0.5567	0.8446	0.9190
No log	53.0	212	0.8697	0.5876	0.8697	0.9326
No log	53.5	214	0.9053	0.5571	0.9053	0.9515
No log	54.0	216	0.9710	0.5382	0.9710	0.9854
No log	54.5	218	1.0227	0.5119	1.0227	1.0113
No log	55.0	220	0.9860	0.5208	0.9860	0.9930
No log	55.5	222	0.9077	0.5559	0.9077	0.9527
No log	56.0	224	0.8499	0.5821	0.8499	0.9219
No log	56.5	226	0.8339	0.5786	0.8339	0.9132
No log	57.0	228	0.8591	0.5779	0.8591	0.9269
No log	57.5	230	0.9467	0.5371	0.9467	0.9730
No log	58.0	232	1.0194	0.5029	1.0194	1.0097
No log	58.5	234	1.0036	0.5089	1.0036	1.0018
No log	59.0	236	0.9316	0.5605	0.9316	0.9652
No log	59.5	238	0.9141	0.5583	0.9141	0.9561
No log	60.0	240	0.9265	0.5661	0.9265	0.9625
No log	60.5	242	0.9619	0.5211	0.9619	0.9808
No log	61.0	244	0.9302	0.5616	0.9302	0.9645
No log	61.5	246	0.8969	0.5822	0.8969	0.9470
No log	62.0	248	0.8896	0.5904	0.8896	0.9432
No log	62.5	250	0.9104	0.5790	0.9104	0.9541
No log	63.0	252	0.9532	0.5527	0.9532	0.9763
No log	63.5	254	0.9942	0.5296	0.9942	0.9971
No log	64.0	256	0.9656	0.5360	0.9656	0.9827
No log	64.5	258	0.8930	0.5760	0.8930	0.9450
No log	65.0	260	0.8697	0.5876	0.8697	0.9326
No log	65.5	262	0.8804	0.5760	0.8804	0.9383
No log	66.0	264	0.9210	0.5652	0.9210	0.9597
No log	66.5	266	0.9967	0.5241	0.9967	0.9983
No log	67.0	268	1.0385	0.4868	1.0385	1.0191
No log	67.5	270	1.0149	0.5	1.0149	1.0074
No log	68.0	272	0.9551	0.5162	0.9551	0.9773
No log	68.5	274	0.9227	0.5616	0.9227	0.9606
No log	69.0	276	0.9063	0.5571	0.9063	0.9520
No log	69.5	278	0.8753	0.5954	0.8753	0.9356
No log	70.0	280	0.8560	0.5847	0.8560	0.9252
No log	70.5	282	0.8733	0.5868	0.8733	0.9345
No log	71.0	284	0.9294	0.5616	0.9294	0.9641
No log	71.5	286	1.0005	0.5013	1.0005	1.0003
No log	72.0	288	1.0219	0.4889	1.0219	1.0109
No log	72.5	290	1.0003	0.4943	1.0003	1.0002
No log	73.0	292	0.9701	0.5164	0.9701	0.9849
No log	73.5	294	0.9055	0.5703	0.9055	0.9516
No log	74.0	296	0.8458	0.5779	0.8458	0.9197
No log	74.5	298	0.8213	0.5893	0.8213	0.9063
No log	75.0	300	0.8245	0.5893	0.8245	0.9080
No log	75.5	302	0.8475	0.5839	0.8475	0.9206
No log	76.0	304	0.9015	0.5595	0.9015	0.9494
No log	76.5	306	0.9724	0.5050	0.9724	0.9861
No log	77.0	308	1.0442	0.4810	1.0442	1.0219
No log	77.5	310	1.0894	0.4545	1.0894	1.0437
No log	78.0	312	1.0776	0.4553	1.0776	1.0381
No log	78.5	314	1.0337	0.4873	1.0337	1.0167
No log	79.0	316	0.9697	0.5315	0.9697	0.9847
No log	79.5	318	0.8946	0.5735	0.8946	0.9458
No log	80.0	320	0.8370	0.5947	0.8370	0.9149
No log	80.5	322	0.8158	0.5885	0.8158	0.9032
No log	81.0	324	0.8137	0.5885	0.8137	0.9021
No log	81.5	326	0.8281	0.5951	0.8281	0.9100
No log	82.0	328	0.8581	0.5670	0.8581	0.9264
No log	82.5	330	0.8855	0.5477	0.8855	0.9410
No log	83.0	332	0.9128	0.5318	0.9128	0.9554
No log	83.5	334	0.9412	0.5204	0.9412	0.9701
No log	84.0	336	0.9933	0.4782	0.9933	0.9966
No log	84.5	338	1.0215	0.4782	1.0215	1.0107
No log	85.0	340	1.0147	0.4782	1.0147	1.0073
No log	85.5	342	0.9901	0.4833	0.9901	0.9951
No log	86.0	344	0.9549	0.5007	0.9549	0.9772
No log	86.5	346	0.9177	0.5348	0.9177	0.9579
No log	87.0	348	0.8706	0.5703	0.8706	0.9331
No log	87.5	350	0.8443	0.5697	0.8443	0.9188
No log	88.0	352	0.8313	0.6011	0.8313	0.9117
No log	88.5	354	0.8347	0.5871	0.8347	0.9136
No log	89.0	356	0.8456	0.5697	0.8456	0.9196
No log	89.5	358	0.8586	0.5731	0.8586	0.9266
No log	90.0	360	0.8708	0.5703	0.8708	0.9332
No log	90.5	362	0.8868	0.5453	0.8868	0.9417
No log	91.0	364	0.9100	0.5429	0.9100	0.9539
No log	91.5	366	0.9385	0.4962	0.9385	0.9687
No log	92.0	368	0.9627	0.4791	0.9627	0.9812
No log	92.5	370	0.9787	0.4791	0.9787	0.9893
No log	93.0	372	0.9849	0.4838	0.9849	0.9924
No log	93.5	374	0.9847	0.4786	0.9847	0.9923
No log	94.0	376	0.9742	0.4791	0.9742	0.9870
No log	94.5	378	0.9650	0.4791	0.9650	0.9823
No log	95.0	380	0.9614	0.4791	0.9614	0.9805
No log	95.5	382	0.9581	0.4791	0.9581	0.9788
No log	96.0	384	0.9519	0.4791	0.9519	0.9757
No log	96.5	386	0.9403	0.5136	0.9403	0.9697
No log	97.0	388	0.9312	0.5193	0.9312	0.9650
No log	97.5	390	0.9224	0.5193	0.9224	0.9604
No log	98.0	392	0.9138	0.5431	0.9138	0.9559
No log	98.5	394	0.9100	0.5442	0.9100	0.9539
No log	99.0	396	0.9080	0.5442	0.9080	0.9529
No log	99.5	398	0.9072	0.5442	0.9072	0.9525
No log	100.0	400	0.9072	0.5442	0.9072	0.9525

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1