ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k3_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: 0.8467
• Qwk: 0.5477
• Mse: 0.8467
• Rmse: 0.9202

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	3.9673	0.0094	3.9673	1.9918
No log	0.2105	4	2.6822	0.1025	2.6822	1.6377
No log	0.3158	6	1.3908	0.1538	1.3908	1.1793
No log	0.4211	8	0.7873	0.1007	0.7873	0.8873
No log	0.5263	10	0.6311	0.2730	0.6311	0.7944
No log	0.6316	12	0.6469	0.2243	0.6469	0.8043
No log	0.7368	14	0.7144	0.3627	0.7144	0.8452
No log	0.8421	16	0.8002	0.3241	0.8002	0.8945
No log	0.9474	18	0.7107	0.4081	0.7107	0.8430
No log	1.0526	20	0.7170	0.4110	0.7170	0.8468
No log	1.1579	22	0.7221	0.4498	0.7221	0.8498
No log	1.2632	24	0.6036	0.4620	0.6036	0.7769
No log	1.3684	26	0.5649	0.4986	0.5649	0.7516
No log	1.4737	28	0.5911	0.5505	0.5911	0.7688
No log	1.5789	30	0.6415	0.5633	0.6415	0.8009
No log	1.6842	32	0.6119	0.5738	0.6119	0.7823
No log	1.7895	34	0.6802	0.5354	0.6802	0.8247
No log	1.8947	36	0.6793	0.5625	0.6793	0.8242
No log	2.0	38	0.6515	0.6376	0.6515	0.8072
No log	2.1053	40	0.5963	0.5919	0.5963	0.7722
No log	2.2105	42	0.5536	0.5850	0.5536	0.7440
No log	2.3158	44	0.5082	0.5736	0.5082	0.7129
No log	2.4211	46	0.5296	0.5802	0.5296	0.7278
No log	2.5263	48	0.6308	0.5490	0.6308	0.7942
No log	2.6316	50	0.5533	0.5682	0.5533	0.7438
No log	2.7368	52	0.5295	0.5682	0.5295	0.7277
No log	2.8421	54	0.6095	0.5580	0.6095	0.7807
No log	2.9474	56	0.5851	0.5702	0.5851	0.7649
No log	3.0526	58	0.6289	0.5605	0.6289	0.7930
No log	3.1579	60	0.7864	0.5339	0.7864	0.8868
No log	3.2632	62	0.6782	0.5373	0.6782	0.8235
No log	3.3684	64	0.6301	0.5836	0.6301	0.7938
No log	3.4737	66	0.6849	0.5883	0.6849	0.8276
No log	3.5789	68	0.7523	0.6040	0.7523	0.8673
No log	3.6842	70	0.8357	0.5869	0.8357	0.9142
No log	3.7895	72	0.8797	0.6072	0.8797	0.9379
No log	3.8947	74	0.9402	0.5402	0.9402	0.9696
No log	4.0	76	0.9999	0.5100	0.9999	1.0000
No log	4.1053	78	0.9803	0.5443	0.9803	0.9901
No log	4.2105	80	0.9204	0.5600	0.9204	0.9594
No log	4.3158	82	0.9263	0.5294	0.9263	0.9625
No log	4.4211	84	0.8952	0.5571	0.8952	0.9462
No log	4.5263	86	0.8823	0.5688	0.8823	0.9393
No log	4.6316	88	0.8412	0.5495	0.8412	0.9172
No log	4.7368	90	0.8315	0.5624	0.8315	0.9119
No log	4.8421	92	0.8203	0.5565	0.8203	0.9057
No log	4.9474	94	0.7567	0.5837	0.7567	0.8699
No log	5.0526	96	0.7108	0.5976	0.7108	0.8431
No log	5.1579	98	0.6743	0.6234	0.6743	0.8211
No log	5.2632	100	0.6841	0.6083	0.6841	0.8271
No log	5.3684	102	0.7509	0.5865	0.7509	0.8665
No log	5.4737	104	0.7275	0.5728	0.7275	0.8530
No log	5.5789	106	0.7217	0.5693	0.7217	0.8495
No log	5.6842	108	0.6804	0.6388	0.6804	0.8249
No log	5.7895	110	0.6936	0.5888	0.6936	0.8328
No log	5.8947	112	0.6944	0.5851	0.6944	0.8333
No log	6.0	114	0.6965	0.6303	0.6965	0.8346
No log	6.1053	116	0.7656	0.5464	0.7656	0.8750
No log	6.2105	118	0.8533	0.5130	0.8533	0.9238
No log	6.3158	120	0.9362	0.5214	0.9362	0.9676
No log	6.4211	122	0.9121	0.5257	0.9121	0.9550
No log	6.5263	124	0.8437	0.5639	0.8437	0.9185
No log	6.6316	126	0.8342	0.5697	0.8342	0.9134
No log	6.7368	128	0.8350	0.5778	0.8350	0.9138
No log	6.8421	130	0.8175	0.5773	0.8175	0.9041
No log	6.9474	132	0.8173	0.5795	0.8173	0.9041
No log	7.0526	134	0.8256	0.5562	0.8256	0.9086
No log	7.1579	136	0.8222	0.5440	0.8222	0.9067
No log	7.2632	138	0.8116	0.5488	0.8116	0.9009
No log	7.3684	140	0.8196	0.5306	0.8196	0.9053
No log	7.4737	142	0.7951	0.5601	0.7951	0.8917
No log	7.5789	144	0.7689	0.5963	0.7689	0.8769
No log	7.6842	146	0.7616	0.5854	0.7616	0.8727
No log	7.7895	148	0.7695	0.5958	0.7695	0.8772
No log	7.8947	150	0.7971	0.5326	0.7971	0.8928
No log	8.0	152	0.8415	0.5296	0.8415	0.9173
No log	8.1053	154	0.8715	0.5187	0.8715	0.9335
No log	8.2105	156	0.8557	0.5249	0.8557	0.9251
No log	8.3158	158	0.8248	0.5380	0.8248	0.9082
No log	8.4211	160	0.7961	0.5640	0.7961	0.8922
No log	8.5263	162	0.7866	0.5841	0.7866	0.8869
No log	8.6316	164	0.7879	0.5977	0.7879	0.8876
No log	8.7368	166	0.8029	0.5977	0.8029	0.8961
No log	8.8421	168	0.8055	0.5850	0.8055	0.8975
No log	8.9474	170	0.8175	0.5763	0.8175	0.9041
No log	9.0526	172	0.8334	0.5514	0.8334	0.9129
No log	9.1579	174	0.8388	0.5490	0.8388	0.9159
No log	9.2632	176	0.8452	0.5160	0.8452	0.9193
No log	9.3684	178	0.8450	0.5272	0.8450	0.9193
No log	9.4737	180	0.8456	0.5272	0.8456	0.9196
No log	9.5789	182	0.8487	0.5272	0.8487	0.9212
No log	9.6842	184	0.8460	0.5346	0.8460	0.9198
No log	9.7895	186	0.8449	0.5477	0.8449	0.9192
No log	9.8947	188	0.8456	0.5477	0.8456	0.9196
No log	10.0	190	0.8467	0.5477	0.8467	0.9202

Framework versions

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