search

mihirp1998 / EmbLang

Embodied Language Grounding With 3D Visual Feature Representations
Apache License 2.0
4 stars 1 forks source link

FileNotFoundError: [Errno 2] No such file or directory: 'CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p' #1

Open wenhan-wu-stephen opened 3 years ago

wenhan-wu-stephen commented 3 years ago

Details: Traceback (most recent call last): File "EmbLang-master/vis_imagine_static_voxels/main.py", line 44, in model = models.GQN3D() File "/content/EmbLang-master/vis_imagine_static_voxels/models.py", line 269, in init return super().init(nets.GQN3D) File "/content/EmbLang-master/vis_imagine_staticvoxels/models.py", line 252, in init input = inputs.GQNShapenet() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 592, in init self.child = MultiViewReconstructionInput2() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 143, in init self.train_data = self.make_data(const.train_file, True) File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 176, in make_data fns = pickle.load(open("CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p","rb")) FileNotFoundError: [Errno 2] No such file or directory: 'CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p'

Can you tell me where is the file "single_obj.p" or how to solve the error,please?

mihirp1998 commented 3 years ago

Hi,

Ill update it by this week as I'm out of town. Meanwhile you can take a look at https://github.com/mihirp1998/Disentangling-3D-Prototypical-Nets as this is our updated iclr version/paper of the Embodied language paper.

On Sun, 18 Jul, 2021, 11:30 AM wenhan-wu-stephen, @.***> wrote:

Details: Traceback (most recent call last): File "EmbLang-master/vis_imagine_static_voxels/main.py", line 44, in model = models.GQN3D() File "/content/EmbLang-master/vis_imagine_static_voxels/models.py", line 269, in init return super().init(nets.GQN3D) File "/content/EmbLang-master/vis_imagine_staticvoxels/models.py", line 252, in init input = inputs.GQNShapenet() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 592, in init self.child = MultiViewReconstructionInput2() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 143, in init self.train_data = self.make_data(const.train_file, True) File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 176, in make_data fns = pickle.load(open("CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p","rb")) FileNotFoundError: [Errno 2] No such file or directory: 'CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p'

Can you tell me where is the file "single_obj.p" or how to solve the error,please?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mihirp1998/EmbLang/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE4C5LBD5I4M26BKF6IKWHLTYJUXJANCNFSM5ARZSRFA .

wenhan-wu-stephen commented 3 years ago

Hi, Ill update it by this week as I'm out of town. Meanwhile you can take a look at https://github.com/mihirp1998/Disentangling-3D-Prototypical-Nets as this is our updated iclr version/paper of the Embodied language paper. … On Sun, 18 Jul, 2021, 11:30 AM wenhan-wu-stephen, @.**> wrote: Details: Traceback (most recent call last): File "EmbLang-master/vis_imagine_static_voxels/main.py", line 44, in model = models.GQN3D() File "/content/EmbLang-master/vis_imagine_static_voxels/models.py", line 269, in init return super().init(nets.GQN3D) File "/content/EmbLang-master/vis_imagine_static_voxels/models.py", line 252, in init input_ = inputs.GQNShapenet() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 592, in init self.child = MultiViewReconstructionInput2() File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 143, in init* self.train_data = self.make_data(const.train_file, True) File "/content/EmbLang-master/vis_imagine_static_voxels/inputs.py", line 176, in make_data fns = pickle.load(open("CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p","rb")) FileNotFoundError: [Errno 2] No such file or directory: 'CLEVR_64_36_AFTER_CORRECTION_NO_DEPTH/single_obj.p' Can you tell me where is the file "single_obj.p" or how to solve the error,please? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE4C5LBD5I4M26BKF6IKWHLTYJUXJANCNFSM5ARZSRFA .

thank you！

mihirp1998 commented 3 years ago

Hi stephen , I happened to have deleted the dataset. However you can generate the dataset by following these steps:

The dataset generation code is here: https://github.com/mihirp1998/blender_emblang/tree/master/image_generation. And the tfrecord generation code is here:https://github.com/mihirp1998/data_clevr_ashar_account/tree/master

You can run it to generate the data and then link the list of tfrecords to the fns variable at https://github.com/mihirp1998/EmbLang/blob/ba943b7a332be6a4789ec74280e64cecdcbd5c4a/vis_imagine_static_voxels/inputs.py#L176.

Also this paper/code specifically only works on 64x64 images. I'll highly recommend you to shift to the new code and paper as it works on high-resolution images i.e 256*256

Thanks