Train: 0%

[760677482]

In the beginning，I noticed that I didn‘t have "pytorch3d",so I used "pip install pytorch3d"，but it showed an error.then I used"pip unintall pytorch3d"and downloaded it from https://anaconda.org/pytorch3d/pytorch3d/files. But now, when training,it's always "Epoch: 1/25. Loop: Train: 0% 0/11579 [03:37<?, ?it/s]". I found the program stopped at this line:loss.backward().

What could be the problem?And I am using cuda 9.2 because my Driver version is outdated.looking forword to your help,thanks ！

[eugenelyj]

You can try unset the flag of args.distributed .

[shariqfarooq123]

Please refer to instructions provided here to install pytorch3d.

If you can't install pytorch3d for your driver version, you may also give a try to pytorch3d-nightly.

As @eugenelyj pointed out, try unsetting the distributed flag. You may get a better traceback.

[libetter0913]

Hello，I had the same problem. When I ran ‘python train.py args_train_nyu.txt’，The program stops here. Can you help me?

[9796l]

Hello，I had the same problem. When I ran ‘python train.py args_train_nyu.txt’，The program stops here. Can you help me?

[9796l]

excuse me,did you solve this problem?

[zhangbaijin]

excuse me,did you solve this problem?

Did you sovle the problem?

[9796l]

我好像是换了个服务器，用了4块显卡的服务器就没有报错了。

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "shariqfarooq123/AdaBins" ***@***.***&gt;; 发送时间:&nbsp;2023年3月24日(星期五) 上午8:40 ***@***.***&gt;; ***@***.******@***.***&gt;; 主题:&nbsp;Re: [shariqfarooq123/AdaBins] Train: 0% (Issue #53) excuse me,did you solve this problem? Did you sovle the problem? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***&gt;

[zhangbaijin]

我好像是换了个服务器，用了4块显卡的服务器就没有报错了。
…
------------------ 原始邮件 ------------------ 发件人: "shariqfarooq123/AdaBins" @.>; 发送时间: 2023年3月24日(星期五) 上午8:40 @.>; @.@.>; 主题: Re: [shariqfarooq123/AdaBins] Train: 0% (Issue #53) excuse me,did you solve this problem? Did you sovle the problem? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>
我这边用的自己的数据集，raw_image，和depth image。作者说的input.txt是指的哪个文件呢，后面857.47又代表啥意思呢？
`
Traceback (most recent call last):
File "train.py", line 403, in
mp.spawn(main_worker, nprocs=ngpus_per_node, args=(ngpus_per_node, args))
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 200, in spawn
return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 158, in start_processes
while not context.join():
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 119, in join
raise Exception(msg)
Exception:

-- Process 1 terminated with the following error:
Traceback (most recent call last):
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 20, in _wrap
fn(i, *args)
File "/root/autodl-tmp/UDepth-master/train.py", line 109, in main_worker
experiment_name=args.name, optimizer_state_dict=None)
File "/root/autodl-tmp/UDepth-master/train.py", line 178, in train
args) else enumerate(train_loader):
File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 345, in next
data = self._next_data()
File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 385, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/root/autodl-tmp/UDepth-master/dataloader.py", line 87, in getitem
focal = float(sample_path.split()[2])
IndexError: list index out of range
`

[jasdkfj]

我好像是换了个服务器，用了4块显卡的服务器就没有报错了。
…
------------------ 原始邮件 ------------------ 发件人: "shariqfarooq123/AdaBins" @.>; 发送时间: 2023年3月24日(星期五) 上午8:40 _@**._>; _@.@._>; 主题: Re: [shariqfarooq123/AdaBins] Train: 0% (Issue #53) excuse me,did you solve this problem? Did you sovle the problem? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: _@_.*>
我这边用的自己的数据集，raw_image，和depth image。作者说的input.txt是指的哪个文件呢，后面857.47又代表啥意思呢？
`
Traceback (most recent call last):
File "train.py", line 403, in
mp.spawn(main_worker, nprocs=ngpus_per_node, args=(ngpus_per_node, args))
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 200, in spawn
return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 158, in start_processes
while not context.join():
File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 119, in join
raise Exception(msg)
Exception:

-- Process 1 terminated with the following error: Traceback (most recent call last): File "/root/miniconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 20, in _wrap fn(i, *args) File "/root/autodl-tmp/UDepth-master/train.py", line 109, in main_worker experiment_name=args.name, optimizer_state_dict=None) File "/root/autodl-tmp/UDepth-master/train.py", line 178, in train args) else enumerate(train_loader): File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 345, in next data = self._next_data() File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 385, in _next_data data = self._dataset_fetcher.fetch(index) # may raise StopIteration File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/root/miniconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/root/autodl-tmp/UDepth-master/dataloader.py", line 87, in getitem focal = float(sample_path.split()[2]) IndexError: list index out of range `

focal的问题，你分割的文件里后面肯定没有标focal length in pixels，读不出来就报错了