dementia-baseline-model

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4056
• Accuracy: 0.6667
• F1: 0.6667
• Precision: 0.7048
• Recall: 0.6667

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	2.2759	34	0.6908	0.5833	0.4298	0.3403	0.5833
No log	4.5517	68	0.6915	0.5833	0.4298	0.3403	0.5833
No log	6.8276	102	0.6674	0.5833	0.4298	0.3403	0.5833
0.6969	9.0690	136	0.6722	0.6667	0.6246	0.6917	0.6667
0.6969	11.3448	170	0.7017	0.4167	0.2451	0.1736	0.4167
0.6969	13.6207	204	0.6845	0.5833	0.4298	0.3403	0.5833
0.6928	15.8966	238	0.6751	0.5833	0.4298	0.3403	0.5833
0.6928	18.1379	272	0.6708	0.5833	0.4298	0.3403	0.5833
0.6928	20.4138	306	0.6703	0.5833	0.4298	0.3403	0.5833
0.6807	22.6897	340	0.5833	0.7708	0.7534	0.8045	0.7708
0.6807	24.9655	374	0.5408	0.7083	0.7099	0.7139	0.7083
0.6807	27.2069	408	0.9694	0.6042	0.5929	0.6813	0.6042
0.4643	29.4828	442	0.8592	0.6875	0.6885	0.6900	0.6875
0.4643	31.7586	476	0.9248	0.7083	0.7057	0.7056	0.7083
0.4643	34.0	510	1.0527	0.7292	0.7309	0.7384	0.7292
0.4643	36.2759	544	1.2136	0.625	0.6276	0.6389	0.625
0.1691	38.5517	578	1.4056	0.6667	0.6667	0.7048	0.6667

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2