018-microsoft-MiniLM-finetuned-yahoo-8000_2000

This model is a fine-tuned version of microsoft/MiniLM-L12-H384-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0511
• F1: 0.6984
• Accuracy: 0.701
• Precision: 0.7000
• Recall: 0.701
• System Ram Used: 4.0180
• System Ram Total: 83.4807
• Gpu Ram Allocated: 0.3995
• Gpu Ram Cached: 12.9297
• Gpu Ram Total: 39.5640
• Gpu Utilization: 35
• Disk Space Used: 26.2045
• Disk Space Total: 78.1898

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: 32
• eval_batch_size: 32
• 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	F1	Accuracy	Precision	Recall	System Ram Used	System Ram Total	Gpu Ram Allocated	Gpu Ram Cached	Gpu Ram Total	Gpu Utilization	Disk Space Used	Disk Space Total
2.1461	0.5	125	1.8487	0.4711	0.5465	0.5181	0.5465	3.8798	83.4807	0.3996	12.9297	39.5640	28	24.5841	78.1898
1.6793	1.0	250	1.5280	0.5799	0.615	0.6207	0.615	3.8827	83.4807	0.3996	12.9297	39.5640	28	24.5842	78.1898
1.4163	1.5	375	1.3396	0.6508	0.6675	0.6691	0.6675	3.8831	83.4807	0.3996	12.9297	39.5640	28	24.5842	78.1898
1.2855	2.0	500	1.2413	0.6633	0.6745	0.6742	0.6745	3.8975	83.4807	0.3996	12.9297	39.5640	30	24.5843	78.1898
1.1364	2.5	625	1.1795	0.6658	0.6725	0.6758	0.6725	4.0967	83.4807	0.3996	12.9297	39.5640	31	25.4571	78.1898
1.0569	3.0	750	1.1167	0.6785	0.6845	0.6841	0.6845	4.0923	83.4807	0.3996	12.9297	39.5640	29	25.4573	78.1898
0.9596	3.5	875	1.0866	0.6883	0.698	0.6920	0.698	3.8765	83.4807	0.3997	12.9297	39.5640	29	25.4573	78.1898
0.917	4.0	1000	1.0703	0.6796	0.6875	0.6841	0.6875	3.8976	83.4807	0.3996	12.9297	39.5640	29	25.4573	78.1898
0.8512	4.5	1125	1.0629	0.6913	0.6915	0.6945	0.6915	4.0600	83.4807	0.3997	12.9297	39.5640	28	25.8306	78.1898
0.8121	5.0	1250	1.0576	0.6838	0.691	0.6905	0.691	4.0432	83.4807	0.3996	12.9297	39.5640	31	25.8306	78.1898
0.7733	5.5	1375	1.0598	0.6774	0.6805	0.6838	0.6805	3.8379	83.4807	0.3996	12.9297	39.5640	28	25.8307	78.1898
0.7431	6.0	1500	1.0376	0.6974	0.702	0.6976	0.702	3.8546	83.4807	0.3996	12.9297	39.5640	31	25.8307	78.1898
0.7065	6.5	1625	1.0457	0.6990	0.6995	0.7014	0.6995	4.0339	83.4807	0.3996	12.9297	39.5640	28	26.2040	78.1898
0.671	7.0	1750	1.0396	0.6956	0.698	0.6966	0.698	4.0384	83.4807	0.3996	12.9297	39.5640	28	26.2040	78.1898
0.6438	7.5	1875	1.0474	0.6887	0.6925	0.6907	0.6925	3.8274	83.4807	0.3996	12.9297	39.5640	28	26.2040	78.1898
0.6326	8.0	2000	1.0384	0.6972	0.698	0.6983	0.698	3.8402	83.4807	0.3996	12.9297	39.5640	34	26.2041	78.1898
0.6121	8.5	2125	1.0440	0.6963	0.698	0.6976	0.698	4.0162	83.4807	0.3996	12.9297	39.5640	29	26.2042	78.1898
0.5911	9.0	2250	1.0518	0.6995	0.701	0.7006	0.701	4.0338	83.4807	0.3996	12.9297	39.5640	28	26.2043	78.1898
0.592	9.5	2375	1.0490	0.7023	0.7035	0.7025	0.7035	3.8126	83.4807	0.3996	12.9297	39.5640	27	26.2043	78.1898
0.5586	10.0	2500	1.0511	0.6984	0.701	0.7000	0.701	3.8448	83.4807	0.3996	12.9297	39.5640	27	26.2043	78.1898

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3