model.no3_expe.dia.1.A_data_ESLO_06.05.25

This model is a fine-tuned version of pyannote/segmentation-3.0 on the CAENNAIS dataset. It achieves the following results on the evaluation set:

• Loss: 0.7507
• Model Preparation Time: 0.0039
• Der: 0.4650
• False Alarm: 0.1514
• Missed Detection: 0.2224
• Confusion: 0.0912

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: 0.001
• 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: cosine
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Der	False Alarm	Missed Detection	Confusion
0.8537	1.0	270	0.7807	0.0039	0.4947	0.1419	0.2527	0.1001
0.7932	2.0	540	0.7687	0.0039	0.4851	0.1467	0.2319	0.1065
0.783	3.0	810	0.7574	0.0039	0.4800	0.1488	0.2284	0.1027
0.7628	4.0	1080	0.7507	0.0039	0.4730	0.1524	0.2207	0.0999
0.7601	5.0	1350	0.7505	0.0039	0.4622	0.1620	0.1999	0.1003
0.7531	6.0	1620	0.7540	0.0039	0.4701	0.1449	0.2302	0.0950
0.7122	7.0	1890	0.7627	0.0039	0.4670	0.1507	0.2258	0.0904
0.7356	8.0	2160	0.7352	0.0039	0.4608	0.1402	0.2357	0.0848
0.7036	9.0	2430	0.7440	0.0039	0.4625	0.1589	0.2083	0.0953
0.7218	10.0	2700	0.7388	0.0039	0.4620	0.1562	0.2122	0.0936
0.7139	11.0	2970	0.7487	0.0039	0.4677	0.1504	0.2231	0.0942
0.7113	12.0	3240	0.7397	0.0039	0.4632	0.1481	0.2225	0.0927
0.6512	13.0	3510	0.7490	0.0039	0.4632	0.1548	0.2164	0.0920
0.6812	14.0	3780	0.7451	0.0039	0.4615	0.1477	0.2233	0.0905
0.6718	15.0	4050	0.7461	0.0039	0.4654	0.1568	0.2155	0.0932
0.6758	16.0	4320	0.7486	0.0039	0.4649	0.1518	0.2201	0.0931
0.6784	17.0	4590	0.7515	0.0039	0.4651	0.1499	0.2243	0.0910
0.6736	18.0	4860	0.7520	0.0039	0.4655	0.1516	0.2221	0.0918
0.6832	19.0	5130	0.7496	0.0039	0.4644	0.1515	0.2221	0.0907
0.7066	20.0	5400	0.7507	0.0039	0.4650	0.1514	0.2224	0.0912

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0