Misinformation-Covid-LowLearningRatebert-base-german-cased

This model is a fine-tuned version of bert-base-german-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5151
• F1: 0.3793

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-07
• 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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.6534	1.0	189	0.6298	0.1000
0.6467	2.0	378	0.6222	0.1379
0.6302	3.0	567	0.6121	0.0784
0.6259	4.0	756	0.6042	0.0870
0.6255	5.0	945	0.5987	0.0870
0.6091	6.0	1134	0.5922	0.0909
0.6237	7.0	1323	0.5881	0.1224
0.6019	8.0	1512	0.5826	0.1277
0.6038	9.0	1701	0.5779	0.2
0.5996	10.0	1890	0.5730	0.1961
0.5858	11.0	2079	0.5678	0.2353
0.5794	12.0	2268	0.5636	0.24
0.5806	13.0	2457	0.5587	0.2264
0.5586	14.0	2646	0.5548	0.24
0.5682	15.0	2835	0.5514	0.24
0.5631	16.0	3024	0.5471	0.2353
0.5603	17.0	3213	0.5425	0.2593
0.5437	18.0	3402	0.5393	0.2593
0.5439	19.0	3591	0.5368	0.2642
0.547	20.0	3780	0.5329	0.2909
0.5408	21.0	3969	0.5297	0.3158
0.5327	22.0	4158	0.5270	0.3158
0.5194	23.0	4347	0.5256	0.3214
0.5206	24.0	4536	0.5227	0.3214
0.516	25.0	4725	0.5205	0.3214
0.5103	26.0	4914	0.5191	0.3214
0.5037	27.0	5103	0.5172	0.3214
0.4974	28.0	5292	0.5180	0.3214
0.5116	29.0	5481	0.5156	0.3214
0.5006	30.0	5670	0.5150	0.3214
0.509	31.0	5859	0.5141	0.3214
0.4832	32.0	6048	0.5150	0.3273
0.4877	33.0	6237	0.5133	0.3214
0.49	34.0	6426	0.5131	0.3158
0.4827	35.0	6615	0.5143	0.3214
0.4986	36.0	6804	0.5125	0.3214
0.4794	37.0	6993	0.5131	0.3793
0.4809	38.0	7182	0.5137	0.3793
0.4929	39.0	7371	0.5114	0.3793
0.465	40.0	7560	0.5135	0.3793
0.4867	41.0	7749	0.5121	0.3793
0.4685	42.0	7938	0.5129	0.3793
0.4643	43.0	8127	0.5142	0.3793
0.4804	44.0	8316	0.5144	0.3793
0.4779	45.0	8505	0.5141	0.3793
0.4701	46.0	8694	0.5139	0.3793
0.4619	47.0	8883	0.5146	0.3793
0.4558	48.0	9072	0.5151	0.3793
0.4824	49.0	9261	0.5152	0.3793
0.4758	50.0	9450	0.5151	0.3793

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.2
• Datasets 2.12.0
• Tokenizers 0.13.3