bert-base-chinese-finetuned-question-answering-6

This model is a fine-tuned version of bert-base-chinese on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0618

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: 5e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
2.0209	0.0461	500	1.9120
1.8506	0.0921	1000	1.7149
1.6908	0.1382	1500	1.6126
1.7279	0.1842	2000	1.8186
1.6033	0.2303	2500	1.5719
1.4682	0.2763	3000	1.5929
1.7458	0.3224	3500	2.0739
1.575	0.3684	4000	1.5012
1.473	0.4145	4500	1.5199
1.5733	0.4605	5000	1.3922
1.8026	0.5066	5500	1.6235
1.3608	0.5526	6000	1.7175
1.4554	0.5987	6500	1.3453
1.7179	0.6447	7000	1.6828
1.6229	0.6908	7500	1.5436
1.4866	0.7369	8000	1.3952
1.5038	0.7829	8500	1.2955
1.5215	0.8290	9000	1.3297
1.5771	0.8750	9500	1.4685
1.4322	0.9211	10000	1.4607
1.3962	0.9671	10500	1.4697
1.0492	1.0132	11000	1.4867
1.29	1.0592	11500	1.7879
1.341	1.1053	12000	1.5917
1.3136	1.1513	12500	1.5838
1.3421	1.1974	13000	1.4495
1.2831	1.2434	13500	1.7703
1.118	1.2895	14000	1.4682
1.1808	1.3355	14500	1.3217
1.1677	1.3816	15000	1.4738
0.968	1.4277	15500	1.6698
1.294	1.4737	16000	1.7064
1.207	1.5198	16500	1.6069
1.0651	1.5658	17000	1.8631
1.0354	1.6119	17500	1.5430
1.4592	1.6579	18000	1.3579
1.2897	1.7040	18500	1.3598
1.2697	1.7500	19000	1.3874
1.0655	1.7961	19500	1.3918
1.2007	1.8421	20000	1.4897
1.0415	1.8882	20500	1.4199
1.2612	1.9342	21000	1.3972
1.3252	1.9803	21500	1.3493
0.7575	2.0263	22000	1.7524
0.9341	2.0724	22500	1.6567
0.6243	2.1184	23000	1.6430
0.8075	2.1645	23500	1.8267
0.8581	2.2106	24000	1.6460
0.9364	2.2566	24500	1.4578
0.9757	2.3027	25000	1.5213
0.6887	2.3487	25500	1.7984
0.9203	2.3948	26000	1.5756
0.8079	2.4408	26500	1.6416
0.836	2.4869	27000	1.7805
0.9916	2.5329	27500	1.2854
0.8501	2.5790	28000	1.5900
0.951	2.6250	28500	1.7041
0.725	2.6711	29000	1.6452
0.9249	2.7171	29500	1.6845
0.6042	2.7632	30000	1.7528
0.617	2.8092	30500	1.7251
0.9236	2.8553	31000	1.6484
0.8841	2.9014	31500	1.7583
0.7921	2.9474	32000	1.5881
0.657	2.9935	32500	1.8081
0.364	3.0395	33000	2.0073
0.3145	3.0856	33500	1.8009
0.4875	3.1316	34000	1.7690
0.7391	3.1777	34500	1.5941
0.4003	3.2237	35000	1.9043
0.5839	3.2698	35500	1.5942
0.3059	3.3158	36000	2.1032
0.7912	3.3619	36500	1.8461
0.4987	3.4079	37000	1.7626
0.4096	3.4540	37500	1.9525
0.4641	3.5000	38000	1.7831
0.6741	3.5461	38500	1.6394
0.5223	3.5922	39000	1.7295
0.6628	3.6382	39500	1.7417
0.3842	3.6843	40000	1.9575
0.5447	3.7303	40500	1.6962
0.5065	3.7764	41000	1.6205
0.4987	3.8224	41500	1.7965
0.4679	3.8685	42000	1.7241
0.4412	3.9145	42500	1.7947
0.5336	3.9606	43000	1.7249
0.4926	4.0066	43500	1.7266
0.3031	4.0527	44000	1.8313
0.1739	4.0987	44500	2.0269
0.1633	4.1448	45000	1.9412
0.2223	4.1908	45500	2.1326
0.2388	4.2369	46000	2.0716
0.297	4.2830	46500	2.0261
0.3006	4.3290	47000	2.0068
0.3573	4.3751	47500	1.8945
0.3003	4.4211	48000	2.0772
0.3278	4.4672	48500	1.9943
0.1343	4.5132	49000	2.0881
0.2136	4.5593	49500	2.1435
0.2846	4.6053	50000	1.9745
0.3605	4.6514	50500	2.0614
0.2491	4.6974	51000	1.9107
0.2531	4.7435	51500	2.0504
0.2409	4.7895	52000	1.9772
0.2536	4.8356	52500	1.8751
0.3425	4.8816	53000	1.8705
0.1654	4.9277	53500	1.9489
0.2758	4.9737	54000	1.9708
0.1577	5.0198	54500	1.9610
0.1067	5.0659	55000	2.0793
0.1657	5.1119	55500	1.9446
0.1461	5.1580	56000	1.9106
0.1248	5.2040	56500	2.0643
0.189	5.2501	57000	1.9927
0.1907	5.2961	57500	2.1214
0.1329	5.3422	58000	2.2351
0.0914	5.3882	58500	2.0377
0.0961	5.4343	59000	2.2045
0.0744	5.4803	59500	2.1818
0.1652	5.5264	60000	2.0111
0.1256	5.5724	60500	2.0353
0.1617	5.6185	61000	2.0892
0.0725	5.6645	61500	2.1369
0.2305	5.7106	62000	2.0559
0.1961	5.7567	62500	2.0562
0.2864	5.8027	63000	2.0555
0.0569	5.8488	63500	2.0838
0.0787	5.8948	64000	2.0614
0.112	5.9409	64500	2.0628
0.1097	5.9869	65000	2.0618

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1