Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add support for some missing modules #58

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Add support for some missing modules #58

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
191b902
Added tanh
LH23 Jul 10, 2023
39fdc1b
Added MulConstant
LH23 Jul 10, 2023
eda8861
Added SpatialZeroPadding
LH23 Jul 10, 2023
d8e88a7
Added InstanceNormalizarion (2d)
LH23 Jul 10, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 21 additions & 0 deletions convert_torch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,19 @@ def lua_recursive_model(module,seq):
n = LambdaReduce(lambda x,y,dim=dim: torch.cat((x,y),dim))
lua_recursive_model(m,n)
add_submodule(seq,n)
elif name == 'Tanh':
n = nn.Tanh()
add_submodule(seq, n)
elif name == 'MulConstant':
n = Lambda(lambda x: x * m.constant_scalar)
add_submodule(seq, n)
elif name == 'SpatialZeroPadding':
n = nn.ConstantPad2d((m.pad_l, m.pad_r, m.pad_t, m.pad_b), 0)
add_submodule(seq, n)
elif name == 'InstanceNormalization':
n = m._instance_norm
copy_param(m, n) # check this
add_submodule(seq, n)
elif name == 'TorchObject':
print('Not Implement',name,real._typename)
else:
Expand Down Expand Up @@ -229,6 +242,14 @@ def lua_recursive_source(module):
s += ['LambdaReduce(lambda x,y,dim={}: torch.cat((x,y),dim), # Concat'.format(m.dimension)]
s += lua_recursive_source(m)
s += [')']
elif name == 'Tanh':
s += ['nn.Tanh()']
elif name == 'MulConstant':
s += ['Lambda(lambda x: x*{}), # MulConstant'.format(m.constant_scalar)]
elif name == 'SpatialZeroPadding':
s += ['nn.ConstantPad2d({}, 0)'.format((m.pad_l, m.pad_r, m.pad_t, m.pad_b))]
elif name == 'InstanceNormalization':
s += ['nn.InstanceNorm2d({},{},{},{}), # InstanceNorm2d'.format(m._instance_norm.num_features, m.eps, m._instance_norm.momentum, m._instance_norm.affine)]
else:
s += '# ' + name + ' Not Implement,\n'
s = map(lambda x: '\t{}'.format(x),s)
Expand Down