anton-l/xtreme_s_xlsr_300m_fleurs_langid_quicker_warmup

Source: 🤗audio-classificationapache-2.01.3 GBother✓ 10 checksumsupdated 1,685d ago

xtreme_s_xlsr_300m_fleurs_langid_quicker_warmup

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

Loss: 1.9765
Accuracy: 0.6199

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: 0.0003
train_batch_size: 4
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 2
total_train_batch_size: 64
total_eval_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 10.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
0.6644	0.26	1000	0.3071	3.2482
0.394	0.52	2000	0.5948	1.8833
0.1034	0.78	3000	0.6297	1.5852
0.1088	1.04	4000	0.5992	1.7903
0.0032	1.3	5000	0.6356	1.6219
0.1813	1.56	6000	0.5788	1.8168
0.0654	1.82	7000	0.6234	1.6089
0.0144	2.08	8000	0.6424	1.6071
0.0019	2.34	9000	0.5822	1.7820
0.0159	2.6	10000	0.6043	1.8407
0.0029	2.86	11000	0.5845	1.8600
0.0458	3.12	12000	0.6299	1.6591
0.013	3.38	13000	0.5903	2.0788
0.003	3.64	14000	0.6188	1.7645
0.0015	3.9	15000	0.6328	1.7739
0.0003	4.16	16000	0.6072	1.8742
0.0005	4.42	17000	0.6231	1.7102
0.006	4.68	18000	0.6122	1.6909
0.2367	4.93	19000	0.6029	1.9891
0.005	5.19	20000	0.6220	1.7245
0.0813	5.45	21000	0.5739	2.0495
0.1233	5.71	22000	0.6104	1.9601
0.0003	5.97	23000	0.5924	1.8881
0.0003	6.23	24000	0.6055	1.9568
0.0001	6.49	25000	0.6086	1.8489
0.2198	6.75	26000	0.6292	1.8048
0.0261	7.01	27000	2.0284	0.5989
0.0001	7.27	28000	1.7323	0.6431
0.0001	7.53	29000	1.9329	0.6310
0.0011	7.79	30000	1.9256	0.6107
0.0933	8.05	31000	2.3915	0.5896
0.0001	8.31	32000	1.9948	0.6021
0.0003	8.57	33000	1.9518	0.6126
0.0005	8.83	34000	1.8935	0.6243
0.0	9.09	35000	2.0177	0.6144
0.0002	9.35	36000	2.0234	0.6174
0.0	9.61	37000	1.9568	0.6216
0.0	9.87	38000	1.9765	0.6199

Framework versions

Transformers 4.18.0.dev0
Pytorch 1.11.0+cu113
Datasets 1.18.4.dev0
Tokenizers 0.11.6

Seed this model

Have the files? Submit a torrent.

Submit a torrent

Package the files with one command

Have the complete files and permission to share them? Replace the example folder path. The script checks the files, makes a torrent and submits its magnet and checksums. Keep your torrent client seeding. Get your community key from your settings, and read the script before running it.

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo anton-l/xtreme_s_xlsr_300m_fleurs_langid_quicker_warmup ./model-folder