Faster-rcnn.pytorch: Pytorch 1.0 branch ImportError: torch.utils.ffi is deprecated

[andrewjong]

I followed the instructions for Pytorch 1.0: switched to pytorch-1.0 branch, went to lib and ran python setup.py build develop. Install succeeded with no errors.

However, when I try to import roi_pool in the Python prompt, I get "torch.utils.ffi is deprecated" error. Isn't this supposed to be fixed in the pytorch-1.0 branch?

Python 3.7.2 (default, Dec 29 2018, 06:19:36) 
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from model.roi_pooling.modules import roi_pool
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/remoteuser/faster-rcnn.pytorch/lib/model/roi_pooling/modules/roi_pool.py", line 2, in <module>
    from ..functions.roi_pool import RoIPoolFunction
  File "/home/remoteuser/faster-rcnn.pytorch/lib/model/roi_pooling/functions/roi_pool.py", line 3, in <module>
    from .._ext import roi_pooling
  File "/home/remoteuser/faster-rcnn.pytorch/lib/model/roi_pooling/_ext/roi_pooling/__init__.py", line 2, in <module>
    from torch.utils.ffi import _wrap_function
  File "/home/remoteuser/.miniconda3/envs/outfit/lib/python3.7/site-packages/torch/utils/ffi/__init__.py", line 1, in <module>
    raise ImportError("torch.utils.ffi is deprecated. Please use cpp extensions instead.")
ImportError: torch.utils.ffi is deprecated. Please use cpp extensions instead.

Thank you!

[ericsqxd]

I encountered the same problem. It is because the version of pytorch isn't compatible with the repo. You could change the pytorch to 0.4.0. I solve the problem in this way.

[andrewjong]

Thanks @ericsqxd , but the README says

Good news! This repo supports pytorch-1.0 now!!! We borrowed some code and techniques from maskrcnn-benchmark. Just go to pytorch-1.0 branch!

1.0 should be supported right?

[adityaarun1]

@andrewjong @ericsqxd You need to checkout the pytorch-1.0 branch. The master branch supports Pytorch 0.4.

[andrewjong]

@adityaarun1 hello, yes. As mentioned in my post, I first switched to the pytorch-1.0 branch, but still encountered the error. Maybe it's just me; did it work for you?

[adityaarun1]

Yes, it does work for me. Infact, I have successfully trained my model with the new branch. Ensure that you are on the correct branch with git branch.

[lzxzy]

@adityaarun1 hello, yes. As mentioned in my post, I first switched to the pytorch-1.0 branch, but still encountered the error. Maybe it's just me; did it work for you?

hi, I got the same error as you . So have you had solved the problem? thx.

[Worulz]

@lzxzy confirmed it works. using git checkout pytorch-1.0 will fix your problem.

[odellus]

If you look here you can see the ffi tools are still used on the pytorch-1.0 branch.

[adityaarun1]

@odellus If you look at the setup script lib/setup.py, it does not call the said file you mentioned. The ROI and NMS files that are used now reside here, source for the same are here. The files you pointed are required for PyTorch < 0.4.1 and have not been deleted from PyTorch 1.0 branch.

[odellus]

Okey dokey. Makes sense

On Wed, Jan 23, 2019, 10:08 AM Aditya Arun <[email protected] wrote:

@odellus https://github.com/odellus If you look at the setup script
lib/setup.py, it does not call the said file you mentioned. The ROI and
NMS files that are used now reside here
https://github.com/jwyang/faster-rcnn.pytorch/tree/pytorch-1.0/lib/model/roi_layers,
source for the same are here
https://github.com/jwyang/faster-rcnn.pytorch/tree/pytorch-1.0/lib/model/csrc.
The files you pointed are required for PyTorch < 0.4.1 and have not been
deleted from PyTorch 1.0 branch.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jwyang/faster-rcnn.pytorch/issues/412#issuecomment-456906714,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AEeEbDd1-eKRLo29JoxrdPBYHwo99T91ks5vGKUNgaJpZM4ZpAtd
.

[andrewjong]

Closing this issue since it seems it works for others.

[zcunyi]

@odellus If you look at the setup script lib/setup.py, it does not call the said file you mentioned. The ROI and NMS files that are used now reside here, source for the same are here. The files you pointed are required for PyTorch < 0.4.1 and have not been deleted from PyTorch 1.0 branch.

But, how to solve the error?

[adityaarun1]

@zcunyi After cloning the repo, you need to switch to the Pytorch-1.0 branch via git checkout pytorch-1.0. Then you can proceed with the installation instructions given in the README of pytorch-1.0 branch.

[13952522076]

@lzxzy Hi, I got the same problem as you mentioned, did you solve the problem ?

[andrewjong]

I figured it out! My goal was to import the roi layers for my project. The problem was I had the wrong import statement for PyTorch 1.0.

Before I had:
from model.roi_pooling.modules import roi_pool
It should be:
from model.roi_layers import ROIPool

Then it works with no problems.

Edit: if anyone else is looking to use ROI layers from this repo in the future, I wrote a mini tutorial.

[kangkang59812]

I clone the code from branch pytorch-1.0. But I still have this problem.

[Qlulu]

copy the following code to torch.utils.ffi.__init__.py,these codes are copied from torch.utils.ffi.__init__.py of pytorch 0.4.1,and then the raise ImportError can be solved.But I don't know whether the change will influence the next process.

raise ImportError("torch.utils.ffi is deprecated. Please use cpp extensions instead.")

from functools import wraps
def _wrap_function(function, ffi):
@wraps(function)
def safe_call(args, *kwargs):
args = tuple(ffi.cast(_torch_to_cffi.get(arg.type(), 'void') + '', arg._cdata)
if isinstance(arg, torch.Tensor) or torch.is_storage(arg)
else arg
for arg in args)
args = (function,) + args
result = torch._C._safe_call(args, **kwargs)
if isinstance(result, ffi.CData):
typeof = ffi.typeof(result)
if typeof.kind == 'pointer':
cdata = int(ffi.cast('uintptr_t', result))
cname = typeof.item.cname
if cname in _cffi_to_torch:
# TODO: Maybe there is a less janky way to eval
# off of this
return eval(_cffi_to_torch[cname])(cdata=cdata)
return result
return safe_call

[E-Dreamer-LQ]

Does this program work in pytorch1.0, how to fix the error or replace the error reporting module????

[marcunzueta]

@zcunyi After cloning the repo, you need to switch to the Pytorch-1.0 branch via git checkout pytorch-1.0. Then you can proceed with the installation instructions given in the README of pytorch-1.0 branch.

Awesome! Shall we add it specifically in the readme.md?

[rpfly3]

If you look here you can see the ffi tools are still used on the pytorch-1.0 branch.

For my case, ffi is already deprecated in PyTorch 1.0.0

ERROR MSG:
Traceback (most recent call last):
File "waymo.py", line 30, in
from model.nms.nms_wrapper import nms
File "Workspace/Code/faster-rcnn.pytorch-pytorch-1.0/lib/model/nms/nms_wrapper.py", line 10, in
from model.nms.nms_gpu import nms_gpu
File "Workspace/Code/faster-rcnn.pytorch-pytorch-1.0/lib/model/nms/nms_gpu.py", line 4, in
from ._ext import nms
File "Workspace/Code/faster-rcnn.pytorch-pytorch-1.0/lib/model/nms/_ext/nms/__init__.py", line 2, in
from torch.utils.ffi import _wrap_function
File "python-environments/pytorch_1_0/lib/python3.6/site-packages/torch/utils/ffi/__init__.py", line 1, in
raise ImportError("torch.utils.ffi is deprecated. Please use cpp extensions instead.")
ImportError: torch.utils.ffi is deprecated. Please use cpp extensions instead.

[lhaippp]

check git branch to pytorch-1.0 worked for me

[jity16]

I solved by adding the source code of torch.utils.ffi to my project without changing my pytorch version.(I use torch 1.4.0).
For example, save the following code to ffiext.py, then you can replace from torch.utils.ffi import _wrap_function with from ffiext import _wrap_function.
You can also find the following source code from 'https://s0pytorch0org.icopy.site/docs/0.4.1/_modules/torch/utils/ffi.html'.

~~~python
import os
import glob
import tempfile
import shutil
from functools import wraps, reduce
from string import Template
import torch
import torch.cuda
from torch._utils import _accumulate

try:
import cffi
except ImportError:
raise ImportError("torch.utils.ffi requires the cffi package")

if cffi.__version_info__ < (1, 4, 0):
raise ImportError("torch.utils.ffi requires cffi version >= 1.4, but "
"got " + '.'.join(map(str, cffi.__version_info__)))

def _generate_typedefs():
typedefs = []
for t in ['Double', 'Float', 'Long', 'Int', 'Short', 'Char', 'Byte']:
for lib in ['TH', 'THCuda']:
for kind in ['Tensor', 'Storage']:
python_name = t + kind
if t == 'Float' and lib == 'THCuda':
th_name = 'THCuda' + kind
else:
th_name = lib + t + kind
th_struct = 'struct ' + th_name

            typedefs += ['typedef {} {};'.format(th_struct, th_name)]
            # We have to assemble a string here, because we're going to
            # do this lookup based on tensor.type(), which returns a
            # string (not a type object, as this code was before)
            python_module = 'torch.cuda' if lib == 'THCuda' else 'torch'
            python_class = python_module + '.' + python_name
            _cffi_to_torch[th_struct] = python_class
            _torch_to_cffi[python_class] = th_struct
return '\n'.join(typedefs) + '\n'

_cffi_to_torch = {}
_torch_to_cffi = {}
_typedefs = _generate_typedefs()

PY_MODULE_TEMPLATE = Template("""
from torch.utils.ffi import _wrap_function
from .$cffi_wrapper_name import lib as _lib, ffi as _ffi

__all__ = []
def _import_symbols(locals):
for symbol in dir(_lib):
fn = getattr(_lib, symbol)
if callable(fn):
locals[symbol] = _wrap_function(fn, _ffi)
else:
locals[symbol] = fn
__all__.append(symbol)

_import_symbols(locals())
""")

def _setup_wrapper(with_cuda):
here = os.path.abspath(os.path.dirname(__file__))
lib_dir = os.path.join(here, '..', '..', 'lib')
include_dirs = [
os.path.join(lib_dir, 'include'),
os.path.join(lib_dir, 'include', 'TH'),
]

wrapper_source = '#include <TH/TH.h>\n'
if with_cuda:
    import torch.cuda
    wrapper_source += '#include <THC/THC.h>\n'
    if os.sys.platform == 'win32':
        cuda_include_dirs = glob.glob(os.getenv('CUDA_PATH', '') + '/include')
        cuda_include_dirs += glob.glob(os.getenv('NVTOOLSEXT_PATH', '') + '/include')
    else:
        cuda_include_dirs = glob.glob('/usr/local/cuda/include')
        cuda_include_dirs += glob.glob('/Developer/NVIDIA/CUDA-*/include')
    include_dirs.append(os.path.join(lib_dir, 'include', 'THC'))
    include_dirs.extend(cuda_include_dirs)
return wrapper_source, include_dirs

def _create_module_dir(base_path, fullname):
module, _, name = fullname.rpartition('.')
if not module:
target_dir = name
else:
target_dir = reduce(os.path.join, fullname.split('.'))
target_dir = os.path.join(base_path, target_dir)
try:
os.makedirs(target_dir)
except os.error:
pass
for dirname in _accumulate(fullname.split('.'), os.path.join):
init_file = os.path.join(base_path, dirname, '__init__.py')
open(init_file, 'a').close() # Create file if it doesn't exist yet
return name, target_dir

def _build_extension(ffi, cffi_wrapper_name, target_dir, verbose):
try:
tmpdir = tempfile.mkdtemp()
ext_suf = '.pyd' if os.sys.platform == 'win32' else '.so'
libname = cffi_wrapper_name + ext_suf
outfile = ffi.compile(tmpdir=tmpdir, verbose=verbose, target=libname)
shutil.copy(outfile, os.path.join(target_dir, libname))
finally:
shutil.rmtree(tmpdir)

def _make_python_wrapper(name, cffi_wrapper_name, target_dir):
py_source = PY_MODULE_TEMPLATE.substitute(name=name,
cffi_wrapper_name=cffi_wrapper_name)
with open(os.path.join(target_dir, '__init__.py'), 'w') as f:
f.write(py_source)

def create_extension(name, headers, sources, verbose=True, with_cuda=False,
package=False, relative_to='.', **kwargs):
base_path = os.path.abspath(os.path.dirname(relative_to))
name_suffix, target_dir = _create_module_dir(base_path, name)
if not package:
cffi_wrapper_name = '_' + name_suffix
else:
cffi_wrapper_name = (name.rpartition('.')[0] +
'.{0}._{0}'.format(name_suffix))

wrapper_source, include_dirs = _setup_wrapper(with_cuda)
include_dirs.extend(kwargs.pop('include_dirs', []))

if os.sys.platform == 'win32':
    library_dirs = glob.glob(os.getenv('CUDA_PATH', '') + '/lib/x64')
    library_dirs += glob.glob(os.getenv('NVTOOLSEXT_PATH', '') + '/lib/x64')

    here = os.path.abspath(os.path.dirname(__file__))
    lib_dir = os.path.join(here, '..', '..', 'lib')

    library_dirs.append(os.path.join(lib_dir))
else:
    library_dirs = []
library_dirs.extend(kwargs.pop('library_dirs', []))

if isinstance(headers, str):
    headers = [headers]
all_headers_source = ''
for header in headers:
    with open(os.path.join(base_path, header), 'r') as f:
        all_headers_source += f.read() + '\n\n'

ffi = cffi.FFI()
sources = [os.path.join(base_path, src) for src in sources]
# NB: TH headers are C99 now
kwargs['extra_compile_args'] = ['-std=c99'] + kwargs.get('extra_compile_args', [])
ffi.set_source(cffi_wrapper_name, wrapper_source + all_headers_source,
               sources=sources,
               include_dirs=include_dirs,
               library_dirs=library_dirs, **kwargs)
ffi.cdef(_typedefs + all_headers_source)

_make_python_wrapper(name_suffix, '_' + name_suffix, target_dir)

def build():
    _build_extension(ffi, cffi_wrapper_name, target_dir, verbose)
ffi.build = build
return ffi

def _wrap_function(function, ffi):
@wraps(function)
def safe_call(args, *kwargs):
args = tuple(ffi.cast(_torch_to_cffi.get(arg.type(), 'void') + '', arg._cdata)
if isinstance(arg, torch.Tensor) or torch.is_storage(arg)
else arg
for arg in args)
args = (function,) + args
result = torch._C._safe_call(args, **kwargs)
if isinstance(result, ffi.CData):
typeof = ffi.typeof(result)
if typeof.kind == 'pointer':
cdata = int(ffi.cast('uintptr_t', result))
cname = typeof.item.cname
if cname in _cffi_to_torch:
# TODO: Maybe there is a less janky way to eval
# off of this
return eval(_cffi_to_torch[cname])(cdata=cdata)
return result
return safe_call
~~~