How do I run train.py?

[Tayfur26]

I want to train my data set. But I couldn't run the train.py file . I don't know how to proceed . Can you help me please ?

[haviduck]

Check out the params here https://github.com/svip-lab/impersonator/blob/master/options/train_options.py

[Tayfur26]

When I enter the arguments, it asks me for the 'pose.pkl' and 'train.txt' files.
When I define them as empty and add them to the dataset:

---

[TerminalIPythonApp] WARNING | GUI event loop or pylab initialization failed

UnknownBackendTraceback (most recent call last)
/usr/local/lib/python3.6/dist-packages/IPython/core/interactiveshell.py in enable_matplotlib(self, gui)
2953 # Now we must activate the gui pylab wants to use, and fix %run to take
2954 # plot updates into account
-> 2955 self.enable_gui(gui)
2956 self.magics_manager.registry['ExecutionMagics'].default_runner =
2957 pt.mpl_runner(self.safe_execfile)

/usr/local/lib/python3.6/dist-packages/IPython/terminal/interactiveshell.py in enable_gui(self, gui)
512 if gui:
513 self.active_eventloop, self._inputhook =
--> 514 get_inputhook_name_and_func(gui)
515 else:
516 self.active_eventloop = self._inputhook = None

/usr/local/lib/python3.6/dist-packages/IPython/terminal/pt_inputhooks/init.py in get_inputhook_name_and_func(gui)
36
37 if gui not in backends:
---> 38 raise UnknownBackend(gui)
39
40 if gui in aliases:

UnknownBackend: No event loop integration for 'inline'. Supported event loops are: qt, qt4, qt5, gtk, gtk2, gtk3, tk, wx, pyglet, glut, osx
/usr/local/lib/python3.6/dist-packages/IPython/utils/traitlets.py:5: UserWarning: IPython.utils.traitlets has moved to a top-level traitlets package.
warn("IPython.utils.traitlets has moved to a top-level traitlets package.")
------------ Options -------------
D_adam_b1: 0.5
D_adam_b2: 0.999
G_adam_b1: 0.5
G_adam_b2: 0.999
batch_size: 4
bg_replace: False
checkpoints_dir: ./outputs/checkpoints/
cond_nc: 3
data_dir: ./dataset/DatasetiPER
dataset_mode: iPER
debug: False
display_freq_s: 300
do_saturate_mask: False
face_model: assets/pretrains/sphere20a_20171020.pth
fashion_bs: 4
fashion_dir: /public/deep_fashion/intrinsic
final_lr: 2e-06
gen_name: impersonator
gpu_ids: 0
hmr_model: assets/pretrains/hmr_tf2pt.pth
image_size: 256
images_folder: images_HD
intervals: 10
is_train: True
lambda_D_prob: 1
lambda_face: 1
lambda_lp: 10
lambda_mask: 0.1
lambda_mask_smooth: 1e-05
lambda_style: 5
lambda_vgg: 10
load_epoch: 0
lr_D: 0.0002
lr_G: 0.0002
map_name: uv_seg
mask_bce: False
model: impersonator
n_threads_test: 2
n_threads_train: 4
name: running
nepochs_decay: 20
nepochs_no_decay: 10
norm_type: instance
num_iters_validate: 1
part_info: assets/pretrains/smpl_part_info.json
place_bs: 4
place_dir: ./dataset/DatasetiPER
print_freq_s: 60
repeat_num: 6
save_latest_freq_s: 3600
serial_batches: False
smpl_model: assets/pretrains/smpl_model.pkl
smpls_folder: smpls
test_ids_file: val.txt
tex_size: 3
time_step: 10
train_G_every_n_iterations: 1
train_ids_file: train.txt
use_face: False
use_style: False
use_vgg: False
uv_mapping: assets/pretrains/mapper.txt
-------------- End ----------------
./outputs/checkpoints/running
Traceback (most recent call last):
File "train.py", line 140, in
Train()
File "train.py", line 12, in init
data_loader_train = CustomDatasetDataLoader(self._opt, is_for_train=True)
File "/content/impersonator/data/custom_dataset_data_loader.py", line 10, in init
self._create_dataset()
File "/content/impersonator/data/custom_dataset_data_loader.py", line 13, in _create_dataset
self._dataset = DatasetFactory.get_by_name(self._opt.dataset_mode, self._opt, self._is_for_train)
File "/content/impersonator/data/dataset.py", line 23, in get_by_name
dataset = ImPerDataset(opt, is_for_train)
File "/content/impersonator/data/imper_dataset.py", line 131, in init
super(ImPerDataset, self).init(opt, is_for_train)
File "/content/impersonator/data/imper_dataset.py", line 22, in init
self._read_dataset_paths()
File "/content/impersonator/data/imper_dataset.py", line 56, in _read_dataset_paths
self._vids_info = self._read_vids_info(use_ids_filepath)
File "/content/impersonator/data/imper_dataset.py", line 77, in _read_vids_info
assert len(images_path) == len(cams), '{} != {}'.format(len(images_path), len(cams))
AssertionError: 0 != 961

How can i fix this ?

[StevenLiuWen]

@Tayfur26 @haviduck we update the additional files of iPER https://onedrive.live.com/?authkey=%21AJL_NAQMkdXGPlA&id=3705E349C336415F%2188052&cid=3705E349C336415F, and the training details are shown in https://github.com/svip-lab/impersonator/blob/master/doc/train.md. Welcome to clone the latest codes, and follow the training script to train the iPER from scratch.

[thalluripk]

Hi, I have cloned the code and built it on windows 10 surface machine. I experience the following error when execute python demo_imitator.py --gpu_ids 1. Please help

`./outputs/checkpoints/running
python :
At line:1 char:1

• python demo_imitator.py --gpu_ids 1

•   + CategoryInfo          : NotSpecified: (:String) [], RemoteException
    + FullyQualifiedErrorId : NativeCommandError
  
  0%|          | 0/3 [00:00<?, ?it/s]
  

Network impersonator was created

Traceback (most recent call last):

File "demo_imitator.py", line 111, in
main()

File "demo_imitator.py", line 102, in main
generate_actor_result(test_opt, src_img_path)

File "demo_imitator.py", line 44, in generate_actor_result
imitator = Imitator(test_opt)
File "C:\Praveen\Research\AI\impersonator\models\imitator.py", line 19, in init
self._create_networks()

File "C:\Praveen\Research\AI\impersonator\models\imitator.py", line 28, in _create_networks
self.generator = self._create_generator().cuda()

File "C:\Praveen\Research\AI\impersonator\models\imitator.py", line 59, in _create_generator
self._load_params(net, self._opt.load_path)
File "C:\Praveen\Research\AI\impersonator\models\models.py", line 161, in _load_params
load_path), 'Weights file not found. Have you trained a model!? We are not providing one %s' % load_path

AssertionError: Weights file not found. Have you trained a model!? We are not providing one ./outputs/checkpoints/lwb_imper_fashion_place/net_epoch_30_id_G.pth

If you suspect this is an IPython bug, please report it at:
https://github.com/ipython/ipython/issues
or send an email to the mailing list at ipython-dev@python.org

You can print a more detailed traceback right now with "%tb", or use "%debug"
to interactively debug it.

Extra-detailed tracebacks for bug-reporting purposes can be enabled via:
%config Application.verbose_crash=True
`