Hello, Thank you for publishing this wonderful project.
I have a question about the answer to #25
I understood that the solution to #25 involves modifying generate.py to save the data from resnet_chonk into a pkl file, which can then be used for training. Consequently, I have altered generate.py and obtained the following data.
"{'resnet_chonk': tensor([[[[-3.1723e+00, -3.1356e+00, -3.0122e+00, ..., -3.0182e+00, -3.0302e+00, -3.0707e+00], [-3.1781e+00, -3.1504e+00, -2.9376e+00, ..., -2.8380e+00, -2.8835e+00, -3.1385e+00], [-3.1517e+00, -2.9036e+00, -2.5128e+00, ..., -1.6416e+00, -2.6505e+00, -3.0181e+00], ..., [ 1.1058e-01, 8.3331e-03, -6.3481e-01, ..., 1.1358e-01, -3.2545e-01, 2.4348e-02], [ 6.4464e-02, -8.6739e-02, -4.0153e-01, ..., 4.1071e-01, -2.2954e-01, -8.7458e-02], [ 1.4561e-01, -5.4080e-01, 3.4416e-02, ..., -1.7221e-01, 4.5234e-01, -4.2940e-01]]]], device='cuda:0')}
I tried to use the data but got the following error
Traceback (most recent call last): File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/trainers/train_eclustrousC.py", line 572, in <module> main() # pylint: disable=no-value-for-parameter File "/opt/conda/lib/python3.8/site-packages/click/core.py", line 1128, in __call__ return self.main(*args, **kwargs) File "/opt/conda/lib/python3.8/site-packages/click/core.py", line 1053, in main rv = self.invoke(ctx) File "/opt/conda/lib/python3.8/site-packages/click/core.py", line 1395, in invoke return ctx.invoke(self.callback, **ctx.params) File "/opt/conda/lib/python3.8/site-packages/click/core.py", line 754, in invoke return __callback(*args, **kwargs) File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/trainers/train_eclustrousC.py", line 567, in main launch_training(c=c, desc=desc, outdir=opts.outdir, dry_run=opts.dry_run) File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/trainers/train_eclustrousC.py", line 112, in launch_training subprocess_fn(rank=0, c=c, temp_dir=temp_dir) File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/trainers/train_eclustrousC.py", line 61, in subprocess_fn getattr(training_loop, c.training_loop_version)(rank=rank, **c) File "/home/dl/prodata/panic3d-anime-reconstruction/./_train/eg3dc/src/training/training_loop_v0.py", line 284, in training_loop grid_size, images, labels = setup_snapshot_image_grid(training_set=training_set) File "/home/dl/prodata/panic3d-anime-reconstruction/./_train/eg3dc/src/training/training_loop_v0.py", line 74, in setup_snapshot_image_grid images, labels = zip(*[unpk(training_set[i]) for i in grid_indices]) File "/home/dl/prodata/panic3d-anime-reconstruction/./_train/eg3dc/src/training/training_loop_v0.py", line 74, in <listcomp> images, labels = zip(*[unpk(training_set[i]) for i in grid_indices]) File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/datasets/ecrutileE.py", line 49, in __getitem__ x = self.ds[idx % len(self.ds)] File "/home/dl/prodata/panic3d-anime-reconstruction/_train/eg3dc/datasets/ecrutileE.py", line 368, in __getitem__ 'resnet_chonk': torch.tensor(uutil.pload( RuntimeError: Could not infer dtype of dict
Can you please tell me the correct file contents of front.pkl?
Or if my understanding is wrong, please point it out to me.