tied linear layer experiment

tied_linear.py

import torch, torch.nn as nn, torch.nn.functional as F
import numpy as np
import torch.optim as optim

# tied autoencoder using off the shelf nn modules
class TiedAutoEncoderOffTheShelf(nn.Module):
    def __init__(self, inp, out, weight):
        super().__init__()
        self.encoder = nn.Linear(inp, out, bias=False)
        self.decoder = nn.Linear(out, inp, bias=False)

        # tie the weights
        #print(type(self.encoder.weight))
        self.encoder.weight = nn.Parameter(weight)
        self.decoder.weight = nn.Parameter(weight.transpose(0,1))

    def forward(self, input):
        encoded_feats = self.encoder(input)
        reconstructed_output = self.decoder(encoded_feats)
        return encoded_feats, reconstructed_output

# tied auto encoder using functional calls
class TiedAutoEncoderFunctional(nn.Module):
    def __init__(self, inp, out):
        super().__init__()
        self.param = nn.Parameter(torch.randn(out, inp))

    def forward(self, input):
        encoded_feats = F.linear(input, self.param)
        reconstructed_output = F.linear(encoded_feats, self.param.t())
        return encoded_feats, reconstructed_output

# mixed approach
class MixedAppraochTiedAutoEncoder(nn.Module):
    def __init__(self, inp, out, weight):
        super().__init__()
        self.encoder = nn.Linear(inp, out, bias=False)
        self.encoder.weight = nn.Parameter(weight)

    def forward(self, input):
        encoded_feats = self.encoder(input)
        reconstructed_output = F.linear(encoded_feats, self.encoder.weight.t())
        return encoded_feats, reconstructed_output

if __name__ == '__main__':
    tied_module_F = TiedAutoEncoderFunctional(5, 6)

    # instantiate off-the-shelf auto-encoder
    offshelf_weight = tied_module_F.param.data.clone()
    tied_module_offshelf = TiedAutoEncoderOffTheShelf(5, 6, offshelf_weight)

    # instantiate mixed type auto-encoder
    mixed_weight = tied_module_F.param.data.clone()
    tied_module_mixed = MixedAppraochTiedAutoEncoder(5, 6, mixed_weight)

    assert torch.equal(tied_module_offshelf.encoder.weight.data, tied_module_F.param.data), 'F vs offshelf: param not equal'
    assert torch.equal(tied_module_mixed.encoder.weight.data, tied_module_F.param.data), 'F vs mixed: param not equal'

    optim_F = optim.SGD(tied_module_F.parameters(), lr=1)
    optim_offshelf = optim.SGD(tied_module_offshelf.parameters(), lr=1)
    optim_mixed = optim.SGD(tied_module_mixed.parameters(), lr=1)
    # common input
    input = torch.rand(5, 5)

    # zero the gradients
    optim_F.zero_grad()
    optim_offshelf.zero_grad()
    optim_mixed.zero_grad()

    # get output from both modules      
    reconstruction_F = tied_module_F(input) 
    reconstruction_offshelf = tied_module_offshelf(input) 
    reconstruction_mixed = tied_module_mixed(input) 

    # back propagation
    reconstruction_F[1].sum().backward()
    reconstruction_offshelf[1].sum().backward()
    reconstruction_mixed[1].sum().backward()

    # step
    optim_F.step()
    optim_offshelf.step()
    optim_mixed.step()

    # check the equality of output and parameters
    assert torch.equal(reconstruction_offshelf[0], reconstruction_F[0]), 'F vs offshelf: bottleneck not equal'
    assert torch.equal(reconstruction_offshelf[1], reconstruction_F[1]), 'F vs offshelf: output not equal'
    assert (tied_module_offshelf.encoder.weight.data - tied_module_F.param.data).pow(2).sum() < 1e-10, 'F vs offshelf: param after step not equal'
    assert (tied_module_offshelf.encoder.weight.data - offshelf_weight).pow(2).sum() < 1e-10, 'F vs mixed: source weight tensor not equal'

    assert torch.equal(reconstruction_mixed[0], reconstruction_F[0]), 'F vs mixed: bottleneck not equal'
    assert torch.equal(reconstruction_mixed[1], reconstruction_F[1]), 'F vs mixed: output not equal'
    assert (tied_module_mixed.encoder.weight.data - tied_module_F.param.data).pow(2).sum() < 1e-10, 'F vs mixed: param after step not equal'
    assert (tied_module_mixed.encoder.weight.data - mixed_weight).pow(2).sum() < 1e-10, 'F vs mixed: param after step not equal'

    print('success!')

It seems L14 has to be changed as the following for for PyTorch 1.7.0 to be working:

self.decoder.weight = nn.Parameter(self.encoder.weight.transpose(0,1))

It seems L14 has to be changed as the following for for PyTorch 1.7.0 to be working:

self.decoder.weight = nn.Parameter(self.encoder.weight.transpose(0,1))

Thanks for letting me know. I have updated the gist to reflect it.

Note for myself: nn.Parameter shares the underlying memory with the given torch.Tensor

Hello, @InnovArul Thank you for this nice work. But may I ask why can't we just use this approach?

Hello, @InnovArul Thank you for this nice work. But may I ask why can't we just use this approach?

Ideally, the decoder.weight is the transpose of encoder.weight.
So, we need decoder.weight = encoder.weight.t(). However, this will throw the following error: cannot assign 'torch.FloatTensor' as parameter 'weight' (torch.nn.Parameter or None expected).

To overcome that, we need to wrap it with nn.Parameter().
Does this answer your question?

Yes. Thank you. Actually, that's what I did. But maybe my question should have been more on are there any merits to using the other approaches you have enlisted here?

To me, tied auto-encoder with functional calls looks clean without involving nn.Parameter(another_layer.weight) .
Apart from that, I do not see any particular merits in other approaches.

Hello, @InnovArul Thank you for this nice work! I am currently building an Autoencoder for dimensionality reduction with beginner level of knowledge in PyTorch. Sorry if my question is very trivial, but is the same concept can be applied to a non-linear model? I was thinking of putting gradient=False in the decoder layer so that the model only train the weights for encoder only. Is this a correct approach?

Hi, Sorry that I missed your message. I hope you already found the answer.

Just to answer your question, yes, in my understanding, setting decoder.requires_grad_(False) would not add the gradient from decoder to the weights. and it will let the weights to only receive gradients from encoder.