-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathcondor.submit_file
90 lines (65 loc) · 3.09 KB
/
condor.submit_file
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
##################
#
# Example Job for HTCondor
#
####################
#---------------------------------------------
# Name your batch so it's easy to distinguish in the q.
JobBatchName = "Adobe Multiview Train"
# --------------------------------------
# Executable and its arguments
executable = run.sh
# -----------------------------------
# Universe (vanilla, docker)
universe = docker
docker_image = registry.eps.surrey.ac.uk/pinakinathc:latest
# -------------------------------------
# Event, out and error logs
log = condor/c$(cluster).p$(process).log
output = condor/c$(cluster).p$(process).out
error = condor/c$(cluster).p$(process).error
# -----------------------------------
# File Transfer, Input, Output
should_transfer_files = YES
environment = "mount=/vol/research/sketchcaption,/vol/research/NOBACKUP/CVSSP/scratch_4weeks/pinakiR/,/vol/vssp/datasets/still/adobe-wang/"
# ------------------------------------
requirements = ( HasStornext == true ) && ( CUDACapability >= 6.2 )
# -------------------------------------
# Resources
request_GPUs = 1
+GPUMem = 10000
request_CPUs = 12
request_memory = 32G
#This job will complete in less than 1 hours
#NB interactive jobs are limited to 4 hours
+JobRunTime = 36
#This job can checkpoint
+CanCheckpoint = true
# -----------------------------------
# Queue commands
## Train Wang-18
arguments = python train.py --model_name=model_A --exp_name=wang18-model-A --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_AA --exp_name=wang18-model-AA --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_B --exp_name=wang18-model-B --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_BB --exp_name=wang18-model-BB --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_C --exp_name=wang18-model-C --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_D --exp_name=wang18-model-D --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_E --exp_name=wang18-model-E --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_F --exp_name=wang18-model-F --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_G --exp_name=wang18-model-G --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
arguments = python train.py --model_name=model_H --exp_name=wang18-model-H --data_dir=/vol/vssp/datasets/still/adobe-wang/
queue 1
## RENDER dataset
# arguments = python -m utils.render_data --input_dir=../adobe-dataset/shirt_dataset_rest/*/shirt_mesh_r_tmp.obj --output_dir=/vol/research/NOBACKUP/CVSSP/scratch_4weeks/pinakiR/tmp_dataset/training_data/wang18/
# queue 1
# arguments = python -m utils.render_data --input_dir=../adobe-dataset/duygu_dataset/sigasia15/*/*.obj --output_dir=/vol/research/NOBACKUP/CVSSP/scratch_4weeks/pinakiR/tmp_dataset/training_data/siga15
# queue 1