You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I used TSAI to build a model and use the pretrained weights on a separate dataset to finetune. However, I am getting 100% accuracy when testing this dataset. Could anyone help me troubleshoot?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello, I used TSAI to build a model and use the pretrained weights on a separate dataset to finetune. However, I am getting 100% accuracy when testing this dataset. Could anyone help me troubleshoot?
code:
fineTuneName=f'{modelname}_fineTune'
dls3 = get_ts_dls(VLeftavgGut, yNew, splits=new_splits, tfms=tfms, batch_tfms=batch_tfms, path='/Users/nehakulkarni/Documents/GitHub/final-project-musculoskeletal-injury-location')
learn_pre_trained_FT = ts_learner(dls3, ResNetPlus, pretrained=True,\
learn_pre_trained_FT.fine_tune(5, 1e-3)
learn_pre_trained_FT.export(f'{fineTuneName}.pth')
learn_pre_trained_FT.save(fineTuneName)
output:
check unmatched_layers: ['backbone.0.convblock1.0.weight', 'backbone.0.convblock1.1.weight', 'backbone.0.convblock1.1.bias', 'backbone.0.convblock1.1.running_mean', 'backbone.0.convblock1.1.running_var', 'backbone.0.convblock2.0.weight', 'backbone.0.convblock2.1.weight', 'backbone.0.convblock2.1.bias', 'backbone.0.convblock2.1.running_mean', 'backbone.0.convblock2.1.running_var', 'backbone.0.convblock3.0.weight', 'backbone.0.convblock3.1.weight', 'backbone.0.convblock3.1.bias', 'backbone.0.convblock3.1.running_mean', 'backbone.0.convblock3.1.running_var', 'backbone.0.shortcut.0.weight', 'backbone.0.shortcut.1.weight', 'backbone.0.shortcut.1.bias', 'backbone.0.shortcut.1.running_mean', 'backbone.0.shortcut.1.running_var', 'backbone.1.convblock1.0.weight', 'backbone.1.convblock1.1.weight', 'backbone.1.convblock1.1.bias', 'backbone.1.convblock1.1.running_mean', 'backbone.1.convblock1.1.running_var', 'backbone.1.convblock2.0.weight', 'backbone.1.convblock2.1.weight', 'backbone.1.convblock2.1.bias', 'backbone.1.convblock2.1.running_mean', 'backbone.1.convblock2.1.running_var', 'backbone.1.convblock3.0.weight', 'backbone.1.convblock3.1.weight', 'backbone.1.convblock3.1.bias', 'backbone.1.convblock3.1.running_mean', 'backbone.1.convblock3.1.running_var', 'backbone.1.shortcut.0.weight', 'backbone.1.shortcut.1.weight', 'backbone.1.shortcut.1.bias', 'backbone.1.shortcut.1.running_mean', 'backbone.1.shortcut.1.running_var', 'backbone.2.convblock1.0.weight', 'backbone.2.convblock1.1.weight', 'backbone.2.convblock1.1.bias', 'backbone.2.convblock1.1.running_mean', 'backbone.2.convblock1.1.running_var', 'backbone.2.convblock2.0.weight', 'backbone.2.convblock2.1.weight', 'backbone.2.convblock2.1.bias', 'backbone.2.convblock2.1.running_mean', 'backbone.2.convblock2.1.running_var', 'backbone.2.convblock3.0.weight', 'backbone.2.convblock3.1.weight', 'backbone.2.convblock3.1.bias', 'backbone.2.convblock3.1.running_mean', 'backbone.2.convblock3.1.running_var', 'backbone.2.shortcut.weight', 'backbone.2.shortcut.bias', 'backbone.2.shortcut.running_mean', 'backbone.2.shortcut.running_var']
Beta Was this translation helpful? Give feedback.
All reactions