albert-base-ours-run-5

This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.9766	1.0	200	0.8865	0.645	0.5935	0.5872	0.5881
0.7725	2.0	400	1.0650	0.665	0.7143	0.5936	0.5556
0.6018	3.0	600	0.8558	0.7	0.6637	0.6444	0.6456
0.3838	4.0	800	0.9796	0.67	0.6220	0.6219	0.6218
0.2135	5.0	1000	1.4533	0.675	0.6611	0.5955	0.6055
0.1209	6.0	1200	1.4688	0.67	0.6392	0.6474	0.6398
0.072	7.0	1400	1.8395	0.695	0.6574	0.6540	0.6514
0.0211	8.0	1600	2.0849	0.7	0.6691	0.6607	0.6603
0.0102	9.0	1800	2.3042	0.695	0.6675	0.6482	0.6533
0.0132	10.0	2000	2.2390	0.685	0.6472	0.6423	0.6439
0.004	11.0	2200	2.3779	0.68	0.6435	0.6481	0.6443
0.0004	12.0	2400	2.4575	0.675	0.6397	0.6352	0.6357
0.0003	13.0	2600	2.4676	0.675	0.6356	0.6360	0.6356
0.0003	14.0	2800	2.5109	0.68	0.6427	0.6424	0.6422
0.0002	15.0	3000	2.5470	0.675	0.6356	0.6360	0.6356
0.0002	16.0	3200	2.5674	0.675	0.6356	0.6360	0.6356
0.0001	17.0	3400	2.5889	0.685	0.6471	0.6488	0.6474
0.0001	18.0	3600	2.6016	0.675	0.6356	0.6360	0.6356
0.0001	19.0	3800	2.6108	0.675	0.6356	0.6360	0.6356
0.0001	20.0	4000	2.6151	0.675	0.6356	0.6360	0.6356