search

iSarmad / RL-GAN-Net

Official Repository of CVPR 2019 Paper : RL-GAN-Net: A Reinforcement Learning Agent Controlled GAN Network for Real-Time Point Cloud Shape Completion
MIT License
121 stars 42 forks source link

Training the autoencoder #3

Closed ghost closed 5 years ago

ghost commented 5 years ago

Hello thank you for providing this to accompany your work. Really cool stuff here.

I am trying to train the autoencoder and I encountered this error, I was wondering what could be the solution ERROR:visdom:[Errno 111] Connection refused Traceback (most recent call last): File "main2.py", line 484, in <module> main() File "main2.py", line 267, in main chamfer = ChamferLoss(args) File "/home/Documents/Git/RL-GAN-Net/models/lossess.py", line 201, in __init__ self.res.setTempMemoryFraction(0.1) File "/home/anaconda3/envs/pytorch-venv/lib/python3.7/site-packages/faiss/swigfaiss.py", line 1195, in <lambda> __getattr__ = lambda self, name: _swig_getattr(self, StandardGpuResources, name) File "/home/paolo/anaconda3/envs/pytorch-venv/lib/python3.7/site-packages/faiss/swigfaiss.py", line 80, in _swig_getattr raise AttributeError("'%s' object has no attribute '%s'" % (class_type.__name__, name)) AttributeError: 'StandardGpuResources' object has no attribute 'setTempMemoryFraction'

doudoulaile commented 5 years ago

I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1)

doudoulaile commented 5 years ago

RL-GAN-Net/models/lossess.py, line 201

---Original--- From: "rangchu"notifications@github.com Date: Mon, Aug 26, 2019 15:36 PM To: "iSarmad/RL-GAN-Net"RL-GAN-Net@noreply.github.com; Cc: "doudoulaile"2274570131@qq.com;"Comment"comment@noreply.github.com; Subject: Re: [iSarmad/RL-GAN-Net] Training the autoencoder (#3)

I solved this problem by commenting out this line of code.

self.res.setTempMemoryFraction(0.1)

That's great. Can I ask which file is this line of code located?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

ghost commented 5 years ago

RL-GAN-Net/models/lossess.py, line 201 … ---Original--- From: "rangchu"notifications@github.com Date: Mon, Aug 26, 2019 15:36 PM To: "iSarmad/RL-GAN-Net"RL-GAN-Net@noreply.github.com; Cc: "doudoulaile"2274570131@qq.com;"Comment"comment@noreply.github.com; Subject: Re: [iSarmad/RL-GAN-Net] Training the autoencoder (#3) I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1) That's great. Can I ask which file is this line of code located? — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1)

Unfortunately, it didn't work for me, I got this error instead ERROR:visdom:[Errno 111] Connection refused /home/myname/anaconda3/envs/rlgan-venv/lib/python3.7/site-packages/torch/optim/lr_scheduler.py:82: UserWarning: Detected call oflr_scheduler.step()beforeoptimizer.step(). In PyTorch 1.1.0 and later, you should call them in the opposite order:optimizer.step()beforelr_scheduler.step(). Failure to do this will result in PyTorch skipping the first value of the learning rate schedule.See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate "https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate", UserWarning) Traceback (most recent call last): File "main2.py", line 484, in <module> main() File "main2.py", line 285, in main train_loss, _, _ = train(train_loader,model,optimizer,epoch,args,chamfer,visualizer,train_writer) File "main2.py", line 444, in train loss_1 = chamfer(trans_input, pc_1) # File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 297, in __call__ loss = self.forward(predict_pc, gt_pc) File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 256, in forward selected_gt_by_predict = selected_gt_by_predict.to(self.opt.device) RuntimeError: CUDA error: invalid device ordinal

ghost commented 5 years ago

RL-GAN-Net/models/lossess.py, line 201 … ---Original--- From: "rangchu"notifications@github.com Date: Mon, Aug 26, 2019 15:36 PM To: "iSarmad/RL-GAN-Net"RL-GAN-Net@noreply.github.com; Cc: "doudoulaile"2274570131@qq.com;"Comment"comment@noreply.github.com; Subject: Re: [iSarmad/RL-GAN-Net] Training the autoencoder (#3) I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1) That's great. Can I ask which file is this line of code located? — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1)

Unfortunately, it didn't work for me, I got this error instead ERROR:visdom:[Errno 111] Connection refused /home/myname/anaconda3/envs/rlgan-venv/lib/python3.7/site-packages/torch/optim/lr_scheduler.py:82: UserWarning: Detected call oflr_scheduler.step()beforeoptimizer.step(). In PyTorch 1.1.0 and later, you should call them in the opposite order:optimizer.step()beforelr_scheduler.step(). Failure to do this will result in PyTorch skipping the first value of the learning rate schedule.See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate "https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate", UserWarning) Traceback (most recent call last): File "main2.py", line 484, in <module> main() File "main2.py", line 285, in main train_loss, _, _ = train(train_loader,model,optimizer,epoch,args,chamfer,visualizer,train_writer) File "main2.py", line 444, in train loss_1 = chamfer(trans_input, pc_1) # File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 297, in __call__ loss = self.forward(predict_pc, gt_pc) File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 256, in forward selected_gt_by_predict = selected_gt_by_predict.to(self.opt.device) RuntimeError: CUDA error: invalid device ordinal

Silly me. I forgot to start the visdom server

doudoulaile commented 5 years ago

have you finished running the code? 

---Original--- From: "rangchu"<notifications@github.com> Date: Tue, Sep 24, 2019 14:15 PM To: "iSarmad/RL-GAN-Net"<RL-GAN-Net@noreply.github.com>; Cc: "doudoulaile"<2274570131@qq.com>;"Comment"<comment@noreply.github.com>; Subject: Re: [iSarmad/RL-GAN-Net] Training the autoencoder (#3)

RL-GAN-Net/models/lossess.py, line 201 … ---Original--- From: "rangchu"notifications@github.com Date: Mon, Aug 26, 2019 15:36 PM To: "iSarmad/RL-GAN-Net"RL-GAN-Net@noreply.github.com; Cc: "doudoulaile"2274570131@qq.com;"Comment"comment@noreply.github.com; Subject: Re: [iSarmad/RL-GAN-Net] Training the autoencoder (#3) I solved this problem by commenting out this line of code. # self.res.setTempMemoryFraction(0.1) That's great. Can I ask which file is this line of code located? — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

I solved this problem by commenting out this line of code.

self.res.setTempMemoryFraction(0.1)

Unfortunately, it didn't work for me, I got this error instead ERROR:visdom:[Errno 111] Connection refused /home/myname/anaconda3/envs/rlgan-venv/lib/python3.7/site-packages/torch/optim/lr_scheduler.py:82: UserWarning: Detected call of lr_scheduler.step()beforeoptimizer.step(). In PyTorch 1.1.0 and later, you should call them in the opposite order: optimizer.step()beforelr_scheduler.step(). Failure to do this will result in PyTorch skipping the first value of the learning rate schedule.See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate "https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate", UserWarning) Traceback (most recent call last): File "main2.py", line 484, in <module> main() File "main2.py", line 285, in main trainloss, , _ = train(train_loader,model,optimizer,epoch,args,chamfer,visualizer,train_writer) File "main2.py", line 444, in train loss_1 = chamfer(trans_input, pc_1) # File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 297, in call loss = self.forward(predict_pc, gt_pc) File "/home/myname/Documents/Git/RL-GAN-Net/models/lossess.py", line 256, in forward selected_gt_by_predict = selected_gt_by_predict.to(self.opt.device) RuntimeError: CUDA error: invalid device ordinal

Silly me. I forgot to start the visdom server

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.