Get your Github student pack for unlimited private repositories
Please inform the onwers/research group members if the provided links are invalid.
-
Sheffield HPC:
Please contact Haiping for requesting access.
-
Departmental Resources:
Please contact Twin Karmakharm ([email protected]) or RSE group ([email protected]) for requesting access.
-
TALE Private Node on ShARC:
-
Starting an interactive session
Type
qrshx -P tale
to get access to the node (node156) with default 2GB real memory allocation. Usetop
to check how busy it is (currently idle, so please use it).If more memory e.g., 16GB is needed for your job, type
qrshx -P tale -q tale.q -l rmem=16G
. Be sure to request enough resource for your job in the interactive session, otherwise, the job will be killed without any warning or error message and you will be dropped out of the TALE node. -
Submitting batch jobs
Batch Jobs can be submitted to the TALE private node by adding the
-P tale -q tale.q
parameters, for example:#!/bin/bash #$ -P tale #$ -q tale.q echo "Hello world"
More about the command for configuring batch jobs can be found on https://www.sheffield.ac.uk/cics/research/hpc/sharc/batch
-
-
Data Storage:
TALE group shared storage folder: "tale2". Only TALE group members have the access.
Access research storage: https://www.sheffield.ac.uk/cics/research-storage/using-research-storage
Access "tale2" from Sheffield HPC: Shared folders are not visible but accessible from Sheffield HPC (ShARC or iceberg) by using this command:
cd /shared/tale2/Shared
- Learning transferable features with deep adaptation networks (DAN) Long et al., in ICML, 2015. [Code]
- Transfer Feature Learning with Joint Distribution Adaptation (JDA) Long et al., in ICCV, 2013. [Code]
- Domain adaptation via transfer component analysis (TCA) Pan et al.., in IJCA, 2009.
- Boosting for transfer learning (TrAdaBoost) Dai et al., in ICML, 2007.