Skip to content

Commit

Permalink
add scripts, tuning lawa
Browse files Browse the repository at this point in the history
  • Loading branch information
Niccolo-Ajroldi committed Nov 17, 2024
1 parent 6c7e69d commit 4b16329
Show file tree
Hide file tree
Showing 12 changed files with 111 additions and 444 deletions.
File renamed without changes.
File renamed without changes.
File renamed without changes.
45 changes: 45 additions & 0 deletions script/lawa/tuning/condor.sub
Original file line number Diff line number Diff line change
@@ -0,0 +1,45 @@
# Define variables
LOGS_DIR=/fast/najroldi/logs/algoperf
EXE=/home/najroldi/algorithmic-efficiency/script/lawa/tuning/auto_run.sh

num_jobs=8

# Job specific vars
workload_or_id=$(Process)
# workload_or_id=imagenet_resnet
framework=pytorch
submission=prize_qualification_baselines/external_tuning/pytorch_nadamw_full_budget.py
search_space=prize_qualification_baselines/external_tuning/tuning_search_space.json
name=a100_x4_yesTF32_10
study=1
num_tuning_trials=1
rng_seed=96
allow_tf_32=1
eval_num_workers=4

# Args
executable = $(EXE)
arguments = \
$(workload_or_id) \
$(framework) \
$(submission) \
$(search_space) \
$(name) \
$(study) \
$(num_tuning_trials) \
$(rng_seed) \
$(allow_tf_32) \
$(eval_num_workers)

# Logs
error = $(LOGS_DIR)/err/job.$(Cluster).$(Process).err
output = $(LOGS_DIR)/out/job.$(Cluster).$(Process).out
log = $(LOGS_DIR)/log/job.$(Cluster).$(Process).log

# Specs
request_memory = 700000
request_cpus = 36
request_gpus = 4
requirements = (TARGET.CUDADeviceName == "NVIDIA A100-SXM4-40GB")

queue $(num_jobs)
26 changes: 26 additions & 0 deletions script/lawa/tuning/nadamw_best.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
{
"dropout_rate": {
"feasible_points": [0.1]
},
"label_smoothing": {
"feasible_points": [0.0]
},
"learning_rate": {
"feasible_points": [0.00175]
},
"one_minus_beta1": {
"feasible_points": [0.06733926164]
},
"beta2": {
"feasible_points": [0.9955159689799007]
},
"weight_decay": {
"feasible_points": [0.08121616522670176]
},
"warmup_factor": {
"feasible_points": [0.02]
},
"seed": {
"feasible_points": [100, 200, 300]
}
}
35 changes: 35 additions & 0 deletions script/lawa/tuning/tuning.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
{
"dropout_rate": {
"feasible_points": [0.1]
},
"label_smoothing": {
"feasible_points": [0.0]
},
"learning_rate": {
"feasible_points": [0.00175]
},
"one_minus_beta1": {
"feasible_points": [0.06733926164]
},
"beta2": {
"feasible_points": [0.9955159689799007]
},
"weight_decay": {
"feasible_points": [0.08121616522670176]
},
"warmup_factor": {
"feasible_points": [0.02]
},
"lawa_burnin_pct": {
"feasible_points": [0.0, 0.02, 0.1]
},
"lawa_every_pct": {
"feasible_points": [0.001, 0.002, 0.005, 0.01]
},
"lawa_queue_len": {
"feasible_points": [5, 10, 20]
},
"seed": {
"feasible_points": [100, 200, 300]
}
}
117 changes: 0 additions & 117 deletions script/pytorch/2xa100/auto_run_array.sh

This file was deleted.

111 changes: 0 additions & 111 deletions script/pytorch/4xa100/auto_run_array.sh

This file was deleted.

Loading

0 comments on commit 4b16329

Please sign in to comment.