Old dependencies can't be installed

[nessie7purple]

Is there any solution to bypass the incompatible thing with old dependencies？

(vctree) ai@ai:/data/disk2/dy/VCTree-Scene-Graph-Generation-master$ conda install pytorch=0.3 torchvision cuda90 -c pytorch Channels:

• pytorch
• defaults Platform: linux-64 Collecting package metadata (repodata.json): done Solving environment: failed

LibMambaUnsatisfiableError: Encountered problems while solving:

• nothing provides cudnn 6.0* needed by pytorch-0.3.0-py27cuda7.5cudnn6.0_0
• cannot install both pin-1-1 and pin-1-1

Could not solve for environment specs The following packages are incompatible ├─ pin-1 is installable with the potential options │ ├─ pin-1 1, which can be installed; │ └─ pin-1 1 conflicts with any installable versions previously reported; └─ pytorch 0.3** is not installable because there are no viable options ├─ pytorch 0.3.0 would require │ └─ cudnn 6.0* , which does not exist (perhaps a missing channel); ├─ pytorch [0.3.0|0.3.1] would require │ └─ cudatoolkit 8.0* , which does not exist (perhaps a missing channel); └─ pytorch 0.3.1 would require └─ cudatoolkit 8.0.* , which does not exist (perhaps a missing channel).

Pins seem to be involved in the conflict. Currently pinned specs:

• python 3.6.* (labeled as 'pin-1')