# Pytorch: Different Input within Stacking Ensemble & Apex library! **URL:** \ **Category:** vision\ **Created:** [December 23, 2021, 3:09pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106 "2021-12-23T15:09:14Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![Sang\_Dinh](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/sang_dinh/32/45006_2.png) [@Sang\_Dinh](https://discuss.pytorch.org/u/Sang_Dinh)\ **Post date:** [December 23, 2021, 3:09pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/1 "2021-12-23T15:09:14Z")
I shortly will describe my project: +2 input for 2 networks: image(3x256x256) & 20 features(1 tensor of 20 number) +Stacking ensemble. [more](https://discuss.pytorch.org/t/custom-ensemble-approach/52024/4) This is my model: ```auto class MetaMelanoma(nn.Module): def __init__ (self,out_dim=9,n_meta_features=0,n_meta_dim=[512, 128]): super(MetaMelanoma,self). __init__ () self.enet = timm.create_model('tf_efficientnet_b0_ns',pretrained=True) self.n_meta_features = n_meta_features self.dropouts = nn.ModuleList([ nn.Dropout(0.5) for _ in range(5)]) in_ch = self.enet.classifier.in_features self.meta = nn.Sequential( nn.Linear(n_meta_features, n_meta_dim[0]), nn.BatchNorm1d(n_meta_dim[0]), Swish_Module(), nn.Dropout(p=0.3), nn.Linear(n_meta_dim[0], n_meta_dim[1]), nn.BatchNorm1d(n_meta_dim[1]), Swish_Module(), nn.Linear(n_meta_dim[1],9) ) self.myfc = nn.Linear(in_ch, out_dim) self.enet.classifier = nn.Identity() def extract(self, x): x = self.enet(x) return x def forward(self, x,x_meta): x = self.extract(x).squeeze(-1).squeeze(-1) if self.n_meta_features > 0: x_meta = self.meta(x_meta) x = torch.cat((x, x_meta), dim=1) for i, dropout in enumerate(self.dropouts): if i == 0: out = self.myfc(dropout(x)) else: out += self.myfc(dropout(x)) out /= len(self.dropouts) return out ``` In order to train the network, I need Apex Nvidia library for optimizer and loss function [more](https://nvidia.github.io/apex/) Problem: To install Apex, we need old cuda version satisfying pytorch version. However, I am using a sever from my school, I can’t change cuda version. So anyone have any solution for this, maybe the same library or protocol? ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [December 23, 2021, 8:53pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/2 "2021-12-23T20:53:51Z")
`apex.amp` is deprecated so use the native mixed-precision training utility via `torch.cuda.amp`. You can find the examples [here](https://pytorch.org/docs/stable/notes/amp_examples.html#amp-examples). ---
**Author:** ![Diaz65](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/d/e274bd/32.png) [@Diaz65](https://discuss.pytorch.org/u/Diaz65)\ **Post date:** [December 29, 2021, 9:35am UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/3 "2021-12-29T09:35:06Z")
Thanks for sharing this info this is useful keep it up[.](https://www.skylightpaycard.me/) ---
**Author:** ![Sang\_Dinh](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/sang_dinh/32/45006_2.png) [@Sang\_Dinh](https://discuss.pytorch.org/u/Sang_Dinh)\ **Post date:** [January 9, 2022, 6:46am UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/4 "2022-01-09T06:46:12Z")
Thank you very much, that was the solution for me. I searched for someone who had the same model, but unfortunately, I am having some problems there with torch.cuda.amp. Could you suggest some projects using different inputs in same the model? ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [January 9, 2022, 8:45pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/5 "2022-01-09T20:45:42Z")
What kind of issues are you seeing using `torch.cuda.amp`? Since this util. is built into PyTorch, you don’t have to build any extensions manually using a locally installed CUDA toolkit as described in your first post. ---
**Author:** ![MyCenturaHealth](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/mycenturahealth/32/46345_2.png) [@MyCenturaHealth](https://discuss.pytorch.org/u/MyCenturaHealth)\ **Post date:** [October 15, 2022, 10:49am UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/7 "2022-10-15T10:49:47Z")
torch.amp provides convenience methods for mixed precision, where some operations use the torch.float32 (float) datatype and other operations use lower precision floating point datatype (lower\_precision\_fp)[:](https://mycenturahealth.site/) torch.float16 (half) or torch.bfloat16. ---
**Author:** ![GrantElliot](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/grantelliot/32/45450_2.png) [@GrantElliot](https://discuss.pytorch.org/u/GrantElliot)\ **Post date:** [November 7, 2022, 8:17pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/8 "2022-11-07T20:17:59Z")
Thanks a lot[.](https://paybyplatema.site/) I will try and let here know how it went ---
**Author:** ![SamuelChukweze](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/samuelchukweze/32/62918_2.png) [@SamuelChukweze](https://discuss.pytorch.org/u/SamuelChukweze)\ **Post date:** [August 5, 2023, 6:18am UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/10 "2023-08-05T06:18:42Z")
Did it worked for you? If yes, can you also help me in this regard[.](https://ehallpass.us/login/) Thanks in advance. ---
**Author:** ![HarryKane](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/harrykane/32/61642_2.png) [@HarryKane](https://discuss.pytorch.org/u/HarryKane)\ **Post date:** [August 11, 2023, 4:51pm UTC](https://discuss.pytorch.org/t/pytorch-different-input-within-stacking-ensemble-apex-library/140106/11 "2023-08-11T16:51:08Z")
I can help you in this regard[,](https://mostlyboating.com/do-canoes-tip-over-easily/) let me know if you need my assistance. I would be more than happy to help you out.