ArabicNewSplits6_FineTuningAraBERTFreeze_run2_AugV5_k20_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: 0.6677
• Qwk: 0.5081
• Mse: 0.6677
• Rmse: 0.8172

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	0.04	2	6.3855	-0.0278	6.3855	2.5270
No log	0.08	4	4.2252	-0.0203	4.2252	2.0555
No log	0.12	6	2.9467	0.0180	2.9467	1.7166
No log	0.16	8	2.0438	0.0223	2.0438	1.4296
No log	0.2	10	1.4444	0.0567	1.4444	1.2018
No log	0.24	12	1.1075	0.0345	1.1075	1.0524
No log	0.28	14	0.9566	-0.0204	0.9566	0.9781
No log	0.32	16	0.9107	0.0243	0.9107	0.9543
No log	0.36	18	0.8562	0.0652	0.8562	0.9253
No log	0.4	20	0.8189	0.1712	0.8189	0.9049
No log	0.44	22	0.7848	0.2040	0.7848	0.8859
No log	0.48	24	0.7654	0.1913	0.7654	0.8749
No log	0.52	26	0.7625	0.2087	0.7625	0.8732
No log	0.56	28	0.7777	0.2574	0.7777	0.8819
No log	0.6	30	0.8428	0.1904	0.8428	0.9180
No log	0.64	32	0.8639	0.1833	0.8639	0.9295
No log	0.68	34	0.8096	0.1908	0.8096	0.8998
No log	0.72	36	0.7807	0.2794	0.7807	0.8836
No log	0.76	38	0.8472	0.2389	0.8472	0.9204
No log	0.8	40	0.9058	0.2237	0.9058	0.9517
No log	0.84	42	0.9014	0.2490	0.9014	0.9494
No log	0.88	44	0.8832	0.2729	0.8832	0.9398
No log	0.92	46	0.8637	0.2309	0.8637	0.9293
No log	0.96	48	0.8884	0.2654	0.8884	0.9426
No log	1.0	50	0.8887	0.2595	0.8887	0.9427
No log	1.04	52	0.8513	0.2099	0.8513	0.9226
No log	1.08	54	0.7599	0.2365	0.7599	0.8717
No log	1.12	56	0.6728	0.3345	0.6728	0.8202
No log	1.16	58	0.6584	0.3523	0.6584	0.8114
No log	1.2	60	0.6898	0.3600	0.6898	0.8306
No log	1.24	62	0.7641	0.3190	0.7641	0.8741
No log	1.28	64	0.8678	0.2694	0.8678	0.9316
No log	1.32	66	0.8241	0.2798	0.8241	0.9078
No log	1.3600	68	0.7670	0.2909	0.7670	0.8758
No log	1.4	70	0.6885	0.3208	0.6885	0.8297
No log	1.44	72	0.6433	0.3426	0.6433	0.8020
No log	1.48	74	0.6152	0.3392	0.6152	0.7844
No log	1.52	76	0.6263	0.3876	0.6263	0.7914
No log	1.56	78	0.6381	0.3809	0.6381	0.7988
No log	1.6	80	0.6506	0.3967	0.6506	0.8066
No log	1.6400	82	0.6837	0.3881	0.6837	0.8269
No log	1.6800	84	0.7458	0.3457	0.7458	0.8636
No log	1.72	86	0.7264	0.3852	0.7264	0.8523
No log	1.76	88	0.7099	0.3956	0.7099	0.8425
No log	1.8	90	0.6919	0.3885	0.6919	0.8318
No log	1.8400	92	0.6546	0.3346	0.6546	0.8091
No log	1.88	94	0.6071	0.3869	0.6071	0.7792
No log	1.92	96	0.5984	0.3833	0.5984	0.7736
No log	1.96	98	0.5751	0.3813	0.5751	0.7584
No log	2.0	100	0.5427	0.4694	0.5427	0.7367
No log	2.04	102	0.5151	0.4689	0.5151	0.7177
No log	2.08	104	0.5181	0.4835	0.5181	0.7198
No log	2.12	106	0.5304	0.4469	0.5304	0.7283
No log	2.16	108	0.5772	0.4410	0.5772	0.7598
No log	2.2	110	0.6655	0.4575	0.6655	0.8158
No log	2.24	112	0.7226	0.4406	0.7226	0.8501
No log	2.2800	114	0.7859	0.4339	0.7859	0.8865
No log	2.32	116	0.8286	0.4223	0.8286	0.9103
No log	2.36	118	0.7722	0.4497	0.7722	0.8787
No log	2.4	120	0.7617	0.4318	0.7617	0.8727
No log	2.44	122	0.7296	0.4628	0.7296	0.8542
No log	2.48	124	0.6645	0.4943	0.6645	0.8152
No log	2.52	126	0.5988	0.5297	0.5988	0.7738
No log	2.56	128	0.5952	0.5042	0.5952	0.7715
No log	2.6	130	0.6127	0.5147	0.6127	0.7828
No log	2.64	132	0.6186	0.5247	0.6186	0.7865
No log	2.68	134	0.6241	0.5240	0.6241	0.7900
No log	2.7200	136	0.6325	0.5259	0.6325	0.7953
No log	2.76	138	0.6537	0.4721	0.6537	0.8085
No log	2.8	140	0.6825	0.4619	0.6825	0.8262
No log	2.84	142	0.6542	0.4476	0.6542	0.8088
No log	2.88	144	0.6250	0.4548	0.6250	0.7906
No log	2.92	146	0.5599	0.5103	0.5599	0.7483
No log	2.96	148	0.5289	0.5680	0.5289	0.7273
No log	3.0	150	0.5381	0.5281	0.5381	0.7335
No log	3.04	152	0.5585	0.5229	0.5585	0.7474
No log	3.08	154	0.5799	0.5316	0.5799	0.7615
No log	3.12	156	0.5894	0.5076	0.5894	0.7678
No log	3.16	158	0.5901	0.5301	0.5901	0.7682
No log	3.2	160	0.6280	0.5071	0.6280	0.7924
No log	3.24	162	0.6301	0.5071	0.6301	0.7938
No log	3.2800	164	0.6112	0.4982	0.6112	0.7818
No log	3.32	166	0.5672	0.5354	0.5672	0.7531
No log	3.36	168	0.5312	0.5684	0.5312	0.7289
No log	3.4	170	0.5240	0.5609	0.5240	0.7239
No log	3.44	172	0.5339	0.5554	0.5339	0.7307
No log	3.48	174	0.5407	0.5664	0.5407	0.7353
No log	3.52	176	0.5558	0.5304	0.5558	0.7455
No log	3.56	178	0.5958	0.4611	0.5958	0.7719
No log	3.6	180	0.6127	0.4496	0.6127	0.7827
No log	3.64	182	0.6498	0.4669	0.6498	0.8061
No log	3.68	184	0.6335	0.4906	0.6335	0.7959
No log	3.7200	186	0.6025	0.5374	0.6025	0.7762
No log	3.76	188	0.5695	0.5370	0.5695	0.7546
No log	3.8	190	0.5550	0.5698	0.5550	0.7450
No log	3.84	192	0.5508	0.5351	0.5508	0.7422
No log	3.88	194	0.5657	0.5054	0.5657	0.7522
No log	3.92	196	0.5890	0.5192	0.5890	0.7675
No log	3.96	198	0.5924	0.4932	0.5924	0.7697
No log	4.0	200	0.5975	0.4843	0.5975	0.7730
No log	4.04	202	0.6065	0.4783	0.6065	0.7788
No log	4.08	204	0.5884	0.5199	0.5884	0.7671
No log	4.12	206	0.5832	0.5178	0.5832	0.7637
No log	4.16	208	0.6064	0.5117	0.6064	0.7787
No log	4.2	210	0.6043	0.5192	0.6043	0.7773
No log	4.24	212	0.5908	0.5192	0.5908	0.7686
No log	4.28	214	0.5825	0.5149	0.5825	0.7632
No log	4.32	216	0.5932	0.4925	0.5932	0.7702
No log	4.36	218	0.5973	0.5156	0.5973	0.7729
No log	4.4	220	0.5914	0.5406	0.5914	0.7690
No log	4.44	222	0.5967	0.5320	0.5967	0.7725
No log	4.48	224	0.6103	0.5194	0.6103	0.7812
No log	4.52	226	0.6353	0.5452	0.6353	0.7971
No log	4.5600	228	0.6590	0.5283	0.6590	0.8118
No log	4.6	230	0.6731	0.5392	0.6731	0.8204
No log	4.64	232	0.6824	0.5480	0.6824	0.8260
No log	4.68	234	0.6882	0.5466	0.6882	0.8296
No log	4.72	236	0.6867	0.5307	0.6867	0.8287
No log	4.76	238	0.6806	0.5153	0.6806	0.8250
No log	4.8	240	0.6709	0.5163	0.6709	0.8191
No log	4.84	242	0.6596	0.4865	0.6596	0.8121
No log	4.88	244	0.6762	0.5	0.6762	0.8223
No log	4.92	246	0.6437	0.5150	0.6437	0.8023
No log	4.96	248	0.6183	0.5373	0.6183	0.7863
No log	5.0	250	0.6054	0.5113	0.6054	0.7781
No log	5.04	252	0.5996	0.5133	0.5996	0.7743
No log	5.08	254	0.5900	0.5484	0.5900	0.7681
No log	5.12	256	0.5844	0.5257	0.5844	0.7645
No log	5.16	258	0.5733	0.5208	0.5733	0.7572
No log	5.2	260	0.5655	0.5262	0.5655	0.7520
No log	5.24	262	0.5616	0.5342	0.5616	0.7494
No log	5.28	264	0.5551	0.5662	0.5551	0.7450
No log	5.32	266	0.5994	0.4336	0.5994	0.7742
No log	5.36	268	0.6864	0.3835	0.6864	0.8285
No log	5.4	270	0.7011	0.4008	0.7011	0.8373
No log	5.44	272	0.6199	0.4311	0.6199	0.7874
No log	5.48	274	0.5970	0.5711	0.5970	0.7726
No log	5.52	276	0.6146	0.5270	0.6146	0.7840
No log	5.5600	278	0.6385	0.5357	0.6385	0.7991
No log	5.6	280	0.6520	0.5202	0.6520	0.8075
No log	5.64	282	0.6615	0.5231	0.6615	0.8133
No log	5.68	284	0.6749	0.5206	0.6749	0.8215
No log	5.72	286	0.6778	0.5045	0.6778	0.8233
No log	5.76	288	0.6773	0.4965	0.6773	0.8230
No log	5.8	290	0.6738	0.5392	0.6738	0.8209
No log	5.84	292	0.6634	0.5323	0.6634	0.8145
No log	5.88	294	0.6584	0.5665	0.6584	0.8114
No log	5.92	296	0.6524	0.5592	0.6524	0.8077
No log	5.96	298	0.6532	0.5811	0.6532	0.8082
No log	6.0	300	0.6560	0.5133	0.6560	0.8099
No log	6.04	302	0.6603	0.4798	0.6603	0.8126
No log	6.08	304	0.6462	0.4945	0.6462	0.8038
No log	6.12	306	0.6417	0.4945	0.6417	0.8011
No log	6.16	308	0.6382	0.5171	0.6382	0.7989
No log	6.2	310	0.6175	0.5017	0.6175	0.7858
No log	6.24	312	0.6117	0.5262	0.6117	0.7821
No log	6.28	314	0.6365	0.4814	0.6365	0.7978
No log	6.32	316	0.6507	0.4750	0.6507	0.8067
No log	6.36	318	0.6277	0.4852	0.6277	0.7923
No log	6.4	320	0.6116	0.5118	0.6116	0.7821
No log	6.44	322	0.6230	0.5199	0.6230	0.7893
No log	6.48	324	0.6277	0.5046	0.6277	0.7923
No log	6.52	326	0.6183	0.4911	0.6183	0.7863
No log	6.5600	328	0.6220	0.5449	0.6220	0.7887
No log	6.6	330	0.6357	0.4991	0.6357	0.7973
No log	6.64	332	0.6587	0.4797	0.6587	0.8116
No log	6.68	334	0.6676	0.4783	0.6676	0.8171
No log	6.72	336	0.6689	0.4809	0.6689	0.8179
No log	6.76	338	0.6475	0.4901	0.6475	0.8047
No log	6.8	340	0.6371	0.4978	0.6371	0.7982
No log	6.84	342	0.6619	0.5161	0.6619	0.8136
No log	6.88	344	0.6656	0.4895	0.6656	0.8158
No log	6.92	346	0.6647	0.4728	0.6647	0.8153
No log	6.96	348	0.6509	0.5273	0.6509	0.8068
No log	7.0	350	0.6539	0.5674	0.6539	0.8086
No log	7.04	352	0.6569	0.5502	0.6569	0.8105
No log	7.08	354	0.6556	0.5655	0.6556	0.8097
No log	7.12	356	0.6655	0.54	0.6655	0.8158
No log	7.16	358	0.6822	0.6018	0.6822	0.8259
No log	7.2	360	0.7258	0.5430	0.7258	0.8519
No log	7.24	362	0.7326	0.5335	0.7326	0.8559
No log	7.28	364	0.7036	0.5631	0.7036	0.8388
No log	7.32	366	0.6815	0.5874	0.6815	0.8255
No log	7.36	368	0.6764	0.6014	0.6764	0.8224
No log	7.4	370	0.6698	0.5947	0.6698	0.8184
No log	7.44	372	0.6655	0.5389	0.6655	0.8158
No log	7.48	374	0.6810	0.5373	0.6810	0.8252
No log	7.52	376	0.6798	0.5261	0.6798	0.8245
No log	7.5600	378	0.6602	0.5149	0.6602	0.8125
No log	7.6	380	0.6461	0.5477	0.6461	0.8038
No log	7.64	382	0.6284	0.5855	0.6284	0.7927
No log	7.68	384	0.6145	0.5828	0.6145	0.7839
No log	7.72	386	0.6030	0.5932	0.6030	0.7766
No log	7.76	388	0.5974	0.5900	0.5974	0.7729
No log	7.8	390	0.5994	0.5900	0.5994	0.7742
No log	7.84	392	0.6093	0.5915	0.6093	0.7806
No log	7.88	394	0.6124	0.5806	0.6124	0.7826
No log	7.92	396	0.6171	0.5898	0.6171	0.7856
No log	7.96	398	0.6192	0.5764	0.6192	0.7869
No log	8.0	400	0.6111	0.5533	0.6111	0.7817
No log	8.04	402	0.6062	0.5187	0.6062	0.7786
No log	8.08	404	0.6016	0.5106	0.6016	0.7756
No log	8.12	406	0.6033	0.5106	0.6033	0.7767
No log	8.16	408	0.6099	0.5159	0.6099	0.7809
No log	8.2	410	0.6266	0.5485	0.6266	0.7916
No log	8.24	412	0.6403	0.5466	0.6403	0.8002
No log	8.28	414	0.6416	0.5533	0.6416	0.8010
No log	8.32	416	0.6491	0.5716	0.6491	0.8057
No log	8.36	418	0.6710	0.5506	0.6710	0.8191
No log	8.4	420	0.6933	0.4991	0.6933	0.8326
No log	8.44	422	0.6895	0.5614	0.6895	0.8303
No log	8.48	424	0.6886	0.5324	0.6886	0.8298
No log	8.52	426	0.6865	0.5035	0.6865	0.8285
No log	8.56	428	0.6668	0.4959	0.6668	0.8166
No log	8.6	430	0.6476	0.4836	0.6476	0.8048
No log	8.64	432	0.6367	0.5036	0.6367	0.7979
No log	8.68	434	0.6247	0.4965	0.6247	0.7904
No log	8.72	436	0.6191	0.5367	0.6191	0.7868
No log	8.76	438	0.6236	0.5549	0.6236	0.7897
No log	8.8	440	0.6270	0.5140	0.6270	0.7918
No log	8.84	442	0.6329	0.5103	0.6329	0.7955
No log	8.88	444	0.6294	0.5242	0.6294	0.7933
No log	8.92	446	0.6404	0.4870	0.6404	0.8002
No log	8.96	448	0.6450	0.4863	0.6450	0.8031
No log	9.0	450	0.6467	0.5243	0.6467	0.8042
No log	9.04	452	0.6482	0.5374	0.6482	0.8051
No log	9.08	454	0.6586	0.5538	0.6586	0.8116
No log	9.12	456	0.6653	0.5070	0.6653	0.8157
No log	9.16	458	0.6595	0.4972	0.6595	0.8121
No log	9.2	460	0.6487	0.5151	0.6487	0.8054
No log	9.24	462	0.6451	0.5551	0.6451	0.8032
No log	9.28	464	0.6473	0.5205	0.6473	0.8045
No log	9.32	466	0.6676	0.5517	0.6676	0.8171
No log	9.36	468	0.7224	0.5216	0.7224	0.8500
No log	9.4	470	0.7749	0.4915	0.7749	0.8803
No log	9.44	472	0.7573	0.5393	0.7573	0.8702
No log	9.48	474	0.7210	0.4816	0.7210	0.8491
No log	9.52	476	0.7227	0.4785	0.7227	0.8501
No log	9.56	478	0.7527	0.4695	0.7527	0.8676
No log	9.6	480	0.7308	0.5258	0.7308	0.8548
No log	9.64	482	0.6996	0.5414	0.6996	0.8364
No log	9.68	484	0.6904	0.54	0.6904	0.8309
No log	9.72	486	0.6922	0.5416	0.6922	0.8320
No log	9.76	488	0.7136	0.5589	0.7136	0.8448
No log	9.8	490	0.7183	0.5518	0.7183	0.8475
No log	9.84	492	0.7033	0.5187	0.7033	0.8386
No log	9.88	494	0.6827	0.5306	0.6827	0.8263
No log	9.92	496	0.6698	0.5165	0.6698	0.8184
No log	9.96	498	0.6558	0.5521	0.6558	0.8098
0.5567	10.0	500	0.6639	0.5561	0.6639	0.8148
0.5567	10.04	502	0.6994	0.5295	0.6994	0.8363
0.5567	10.08	504	0.7092	0.5222	0.7092	0.8421
0.5567	10.12	506	0.6720	0.5618	0.6720	0.8197
0.5567	10.16	508	0.6309	0.5349	0.6309	0.7943
0.5567	10.2	510	0.6416	0.5211	0.6416	0.8010
0.5567	10.24	512	0.6697	0.4625	0.6697	0.8184
0.5567	10.28	514	0.6801	0.4575	0.6801	0.8247
0.5567	10.32	516	0.6677	0.5081	0.6677	0.8172

Framework versions

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