Git Product home page Git Product logo

cca_series's People

Watchers

avatar

cca_series's Issues

请问这份代码运行DCCAE时，loss是负数？ 

cpu
Using 0 GPUs
loading data ...
loading data ...
[ INFO : 2023-04-06 12:01:32,336 ] - DataParallel(
  (module): DCCAE(
    (encoder1): MlpNet(
      (layers): ModuleList(
        (0): Sequential(
          (0): Linear(in_features=784, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (1): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (2): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (3): Sequential(
          (0): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
          (1): Linear(in_features=1024, out_features=10, bias=True)
        )
      )
    )
    (encoder2): MlpNet(
      (layers): ModuleList(
        (0): Sequential(
          (0): Linear(in_features=784, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (1): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (2): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (3): Sequential(
          (0): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
          (1): Linear(in_features=1024, out_features=10, bias=True)
        )
      )
    )
    (decoder1): MlpNet(
      (layers): ModuleList(
        (0): Sequential(
          (0): Linear(in_features=10, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (1): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (2): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (3): Sequential(
          (0): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
          (1): Linear(in_features=1024, out_features=784, bias=True)
        )
      )
    )
    (decoder2): MlpNet(
      (layers): ModuleList(
        (0): Sequential(
          (0): Linear(in_features=10, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (1): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (2): Sequential(
          (0): Linear(in_features=1024, out_features=1024, bias=True)
          (1): Sigmoid()
          (2): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
        )
        (3): Sequential(
          (0): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=False, track_running_stats=True)
          (1): Linear(in_features=1024, out_features=784, bias=True)
        )
      )
    )
  )
)
[ INFO : 2023-04-06 12:01:32,337 ] - RMSprop (
Parameter Group 0
    alpha: 0.99
    centered: False
    eps: 1e-08
    foreach: None
    lr: 0.001
    momentum: 0
    weight_decay: 1e-05
)
/Users/zhao/Projects/ai/CCA_Series/Dccae/../objectives.py:141: UserWarning: torch.symeig is deprecated in favor of torch.linalg.eigh and will be removed in a future PyTorch release.
The default behavior has changed from using the upper triangular portion of the matrix by default to using the lower triangular portion.
L, _ = torch.symeig(A, upper=upper)
should be replaced with
L = torch.linalg.eigvalsh(A, UPLO='U' if upper else 'L')
and
L, V = torch.symeig(A, eigenvectors=True)
should be replaced with
L, V = torch.linalg.eigh(A, UPLO='U' if upper else 'L') (Triggered internally at  /Users/runner/work/_temp/anaconda/conda-bld/pytorch_1659484780698/work/aten/src/ATen/native/BatchLinearAlgebra.cpp:3041.)
  [D1, V1] = torch.symeig(SigmaHat11, eigenvectors=True)
[ INFO : 2023-04-06 12:01:47,858 ] - Epoch 1: val_loss improved from 0.0000 to -5.7144, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:01:47,923 ] - Epoch 1/20 - time: 15.59 - training_loss: -4.9601 - val_loss: -5.7144
[ INFO : 2023-04-06 12:02:02,979 ] - Epoch 2: val_loss improved from -5.7144 to -6.2400, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:02:03,039 ] - Epoch 2/20 - time: 15.12 - training_loss: -5.6013 - val_loss: -6.2400
[ INFO : 2023-04-06 12:02:18,342 ] - Epoch 3: val_loss improved from -6.2400 to -7.0498, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:02:18,404 ] - Epoch 3/20 - time: 15.37 - training_loss: -6.0406 - val_loss: -7.0498
[ INFO : 2023-04-06 12:02:33,191 ] - Epoch 4: val_loss improved from -7.0498 to -7.3024, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:02:33,254 ] - Epoch 4/20 - time: 14.85 - training_loss: -6.4123 - val_loss: -7.3024
[ INFO : 2023-04-06 12:02:48,255 ] - Epoch 5: val_loss improved from -7.3024 to -7.4365, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:02:48,325 ] - Epoch 5/20 - time: 15.07 - training_loss: -6.6911 - val_loss: -7.4365
[ INFO : 2023-04-06 12:03:03,354 ] - Epoch 6: val_loss improved from -7.4365 to -7.5211, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:03:03,408 ] - Epoch 6/20 - time: 15.08 - training_loss: -6.9115 - val_loss: -7.5211
[ INFO : 2023-04-06 12:03:18,382 ] - Epoch 7: val_loss improved from -7.5211 to -7.6173, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:03:18,440 ] - Epoch 7/20 - time: 15.03 - training_loss: -7.0902 - val_loss: -7.6173
[ INFO : 2023-04-06 12:03:33,442 ] - Epoch 8: val_loss improved from -7.6173 to -7.6933, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:03:33,514 ] - Epoch 8/20 - time: 15.07 - training_loss: -7.2408 - val_loss: -7.6933
[ INFO : 2023-04-06 12:03:48,399 ] - Epoch 9: val_loss improved from -7.6933 to -7.7581, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:03:48,461 ] - Epoch 9/20 - time: 14.95 - training_loss: -7.3704 - val_loss: -7.7581
[ INFO : 2023-04-06 12:04:03,122 ] - Epoch 10: val_loss improved from -7.7581 to -7.9082, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:04:03,181 ] - Epoch 10/20 - time: 14.72 - training_loss: -7.4816 - val_loss: -7.9082
[ INFO : 2023-04-06 12:04:18,192 ] - Epoch 11: val_loss did not improve from -7.9082
[ INFO : 2023-04-06 12:04:18,193 ] - Epoch 11/20 - time: 15.01 - training_loss: -7.5805 - val_loss: -7.8961
[ INFO : 2023-04-06 12:04:33,850 ] - Epoch 12: val_loss improved from -7.9082 to -7.9150, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:04:33,905 ] - Epoch 12/20 - time: 15.71 - training_loss: -7.6694 - val_loss: -7.9150
[ INFO : 2023-04-06 12:04:49,641 ] - Epoch 13: val_loss improved from -7.9150 to -7.9408, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:04:49,701 ] - Epoch 13/20 - time: 15.80 - training_loss: -7.7485 - val_loss: -7.9408
[ INFO : 2023-04-06 12:05:05,306 ] - Epoch 14: val_loss did not improve from -7.9408
[ INFO : 2023-04-06 12:05:05,307 ] - Epoch 14/20 - time: 15.61 - training_loss: -7.8191 - val_loss: -7.8902
[ INFO : 2023-04-06 12:05:20,623 ] - Epoch 15: val_loss improved from -7.9408 to -7.9923, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:05:20,715 ] - Epoch 15/20 - time: 15.41 - training_loss: -7.8826 - val_loss: -7.9923
[ INFO : 2023-04-06 12:05:35,985 ] - Epoch 16: val_loss did not improve from -7.9923
[ INFO : 2023-04-06 12:05:35,988 ] - Epoch 16/20 - time: 15.27 - training_loss: -7.9414 - val_loss: -7.9750
[ INFO : 2023-04-06 12:05:51,699 ] - Epoch 17: val_loss improved from -7.9923 to -8.0683, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:05:51,761 ] - Epoch 17/20 - time: 15.77 - training_loss: -8.0033 - val_loss: -8.0683
[ INFO : 2023-04-06 12:06:06,561 ] - Epoch 18: val_loss improved from -8.0683 to -8.0793, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:06:06,691 ] - Epoch 18/20 - time: 14.93 - training_loss: -8.0722 - val_loss: -8.0793
[ INFO : 2023-04-06 12:06:21,838 ] - Epoch 19: val_loss improved from -8.0793 to -8.1336, saving model to DCCAE_checkpoint.model
[ INFO : 2023-04-06 12:06:21,904 ] - Epoch 19/20 - time: 15.21 - training_loss: -8.1463 - val_loss: -8.1336
[ INFO : 2023-04-06 12:06:36,972 ] - Epoch 20: val_loss did not improve from -8.1336
[ INFO : 2023-04-06 12:06:36,973 ] - Epoch 20/20 - time: 15.07 - training_loss: -8.2202 - val_loss: -8.1012
[ INFO : 2023-04-06 12:06:37,898 ] - loss on validation data: -8.1336
[ INFO : 2023-04-06 12:06:38,828 ] - loss on test data: -8.0896

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.