CoolFace
Modelpublic

yenpolin/wav2vec2-tiny-1-cnn-fp16-demo

sourceHugging Faceupdated 4y agoView on Hugging Face
1likes13downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

wav2vec2-tiny-1-cnn-fp16-demo

This model is a fine-tuned version of yenpolin/wav2vec2-tiny-1-cnn on the DNA_R9.4.1 - NA dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.2063
  • —Mean Acc: 2.4426
  • —Median Acc: 0.0

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
  • —trainbatchsize: 100
  • —evalbatchsize: 200
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 500
  • —num_epochs: 100.0
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossMean AccMedian Acc
0.51461.010000.36730.00.0
0.37852.020000.34470.00.0
0.3643.030000.33170.00.0
0.35234.040000.32490.00270.0
0.34475.050000.31810.00270.0
0.33866.060000.31230.00270.0
0.33347.070000.30850.00270.0
0.32948.080000.30460.00.0
0.3269.090000.30270.00.0
0.32310.0100000.29780.00.0
0.3211.0110000.29630.00270.0
0.306112.0120000.26160.04650.0
0.281813.0130000.25750.04860.0
0.276814.0140000.25170.10750.0
0.273115.0150000.24920.03970.0
0.269716.0160000.24480.06480.0
0.265817.0170000.24440.11710.0
0.261918.0180000.23850.13050.0
0.258519.0190000.23890.09600.0
0.254620.0200000.23260.09090.0
0.250521.0210000.22870.06350.0
0.246222.0220000.22400.10700.0
0.240823.0230000.21900.14290.0
0.233424.0240000.21090.72120.0
0.22525.0250000.20270.73050.0
0.217826.0260000.19800.81860.0
0.211527.0270000.19370.97430.0
0.206528.0280000.18920.82660.0
0.202629.0290000.18900.16150.0
0.198730.0300000.18361.00210.0
0.195331.0310000.18300.80090.0
0.192132.0320000.18211.28370.0
0.189333.0330000.18190.59870.0
0.186534.0340000.18350.93600.0
0.183535.0350000.17961.34520.0
0.180836.0360000.18161.46690.0
0.177937.0370000.18062.42690.0
0.175538.0380000.17870.78430.0
0.172639.0390000.18071.86500.0
0.169940.0400000.18112.18930.0
0.16741.0410000.17991.72850.0
0.164442.0420000.17921.58620.0
0.161743.0430000.17851.51650.0
0.15944.0440000.18061.15420.0
0.156345.0450000.18041.83340.0
0.153946.0460000.18302.14500.0
0.151547.0470000.18352.29050.0
0.148948.0480000.18212.48790.0
0.146549.0490000.18061.61130.0
0.144150.0500000.18571.71320.0
0.141851.0510000.18471.28300.0
0.139452.0520000.18621.85480.0
0.13753.0530000.18152.15850.0
0.134554.0540000.18961.19180.0
0.132555.0550000.18921.55080.0
0.130456.0560000.18791.51800.0
0.128257.0570000.18680.80480.0
0.12658.0580000.19061.65210.0
0.124359.0590000.18791.32550.0
0.122260.0600000.18841.29700.0
0.120261.0610000.19051.53700.0
0.118462.0620000.19361.74080.0
0.116763.0630000.19221.55560.0
0.114964.0640000.19601.51760.0
0.113365.0650000.19661.85770.0
0.111766.0660000.19422.18860.0
0.110267.0670000.19611.45470.0
0.108768.0680000.19651.54820.0
0.107269.0690000.19651.56440.0
0.105870.0700000.19821.54360.0
0.104571.0710000.19562.23190.0
0.103372.0720000.20372.53930.0
0.101973.0730000.19771.90890.0
0.100874.0740000.19831.69570.0
0.099775.0750000.19803.01640.0
0.098676.0760000.20041.80430.0
0.097677.0770000.20251.82340.0
0.096678.0780000.20062.17190.0
0.095779.0790000.19911.85940.0
0.094880.0800000.20371.96600.0
0.09481.0810000.20002.26300.0
0.093282.0820000.20682.31640.0
0.092583.0830000.20042.01100.0
0.091884.0840000.20342.44190.0
0.091285.0850000.20312.08230.0
0.090686.0860000.20392.39550.0
0.090187.0870000.20422.49070.0
0.089788.0880000.20522.33990.0
0.089289.0890000.20502.32140.0
0.088890.0900000.20522.13390.0
0.088491.0910000.20502.34320.0
0.088192.0920000.20522.31610.0
0.087893.0930000.20502.42580.0
0.087694.0940000.20512.27420.0
0.087495.0950000.20652.34000.0
0.087296.0960000.20632.40990.0
0.087197.0970000.20602.42490.0
0.08798.0980000.20642.48260.0
0.086999.0990000.20642.44470.0
0.0869100.01000000.20632.44260.0

Framework versions

  • —Transformers 4.26.0
  • —Pytorch 1.12.1
  • —Datasets 2.9.0
  • —Tokenizers 0.13.2