Skip to content

[experiment] mfos add training paths hardstop #106

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 20 commits into
base: main
Choose a base branch
from
Open

[experiment] mfos add training paths hardstop #106

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
20 commits
Select commit Hold shift + click to select a range
916c149
simplify learning rate scheduling logic
newtonkwan Sep 27, 2022
f2ac47d
tidy up
newtonkwan Sep 27, 2022
52bb7a7
add a bunch of shit to try to anneal correctly
newtonkwan Sep 27, 2022
3915315
tidy up
newtonkwan Sep 27, 2022
e25d4b9
tidy up
newtonkwan Sep 27, 2022
92c9504
Merge branch 'main' of github.com:akbir/pax into halfway_learners
akbir Sep 28, 2022
22ffc2d
Merge branch 'main' of github.com:akbir/pax into halfway_learners
akbir Sep 28, 2022
716d950
first push
akbir Sep 28, 2022
018d3f9
first push
akbir Sep 28, 2022
1b06264
first push
akbir Sep 28, 2022
08f9913
add ipd models to earl, gs, and mfos
newtonkwan Sep 28, 2022
52a1908
add a bunch of shit to eval branch
newtonkwan Sep 28, 2022
f99d870
added mfos support
akbir Sep 28, 2022
d785889
add mfos support
akbir Sep 28, 2022
44c1da0
add training paths for mfos
newtonkwan Sep 28, 2022
89b0654
add mfos support
akbir Sep 29, 2022
2e5e25e
Merge branch 'hardstop_learner' of github.com:akbir/pax into mfos_add…
akbir Sep 29, 2022
cb581cf
added hard stop code
akbir Sep 29, 2022
fdf1e0d
added hard stop code
akbir Sep 29, 2022
ec7cbaf
add hard stop paths
newtonkwan Sep 29, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 23 additions & 2 deletions pax/conf/experiment/cg/pre_train.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,8 +61,29 @@ num_seeds: 10
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0/2022-09-19_10.25.28.808811/generation_2900

# {NEW} EARL pretrained against PPO RNN
run_path: ucl-dark/cg/v348mp4r
model_path: exp/EARL-PPO_memory-vs-PPO_memory/run-seed-0/2022-09-25_23.12.56.999833/generation_400
# run_path: ucl-dark/cg/v348mp4r
# model_path: exp/EARL-PPO_memory-vs-PPO_memory/run-seed-0/2022-09-25_23.12.56.999833/generation_400

# FINAL EARL pre-trained against PPO RNN
# run_path: ucl-dark/cg/z0ckvwtf
# model_path: exp/EARL-PPO_memory-vs-PPO_memory/run-seed-0/2022-09-28_01.58.33.843138/generation_1200

# FINAL EARL pre-trained against Tabular
# run_path: ucl-dark/cg/2dxz9fup
# model_path: exp/EARL-PPO_memory-vs-Tabular/run-seed-0/2022-09-28_01.59.11.312943/generation_50

# FINAL MFOS pre-trained against PPO RNN
# run_path: ucl-dark/cg/34iaxeps
# model_path: exp/MFOS-vs-PPO_memory/run-seed-0/2022-09-28_14.23.22.065961/generation_570

# FINAL GS trained against PPO RNN
# run_path: ucl-dark/cg/2qrono7s
# model_path: exp/GS-PPO-vs-PPO_memory/run-seed-0/2022-09-28_01.51.37.913074/generation_1350

# FINAL GS traind against Tab
run_path: ucl-dark/cg/2xc30qob
model_path: exp/GS-PPO-vs-Tabular/run-seed-0/2022-09-28_02.01.03.381847/generation_50


# PPO agent parameters
ppo:
Expand Down
19 changes: 0 additions & 19 deletions pax/conf/experiment/ipd/earl_v_ppo_mem.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@ payoff: [[-1, -1], [-3, 0], [0, -3], [-2, -2]]
evo: True
num_devices: 1


# Training
top_k: 5
popsize: 1000
Expand All @@ -27,24 +26,6 @@ num_generations: 5000
total_timesteps: 1e11
eval_every: 1e11

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
8 changes: 6 additions & 2 deletions pax/conf/experiment/ipd/earl_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,8 +41,12 @@ num_seeds: 20
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# run_path: ucl-dark/ipd/1ui7wfop
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# hardstop
run_path: ucl-dark/ipd/1ow3zit1
model_path: exp/EARL-PPO_memory-vs-Tabular/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-28_14.32.56.208897/generation_4900

# PPO agent parameters
ppo:
Expand Down
18 changes: 0 additions & 18 deletions pax/conf/experiment/ipd/gs_v_ppo.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,24 +27,6 @@ num_generations: 5000
total_timesteps: 1e11
num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
18 changes: 0 additions & 18 deletions pax/conf/experiment/ipd/gs_v_ppo_mem.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,24 +27,6 @@ num_generations: 5000
total_timesteps: 1e11
num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
55 changes: 40 additions & 15 deletions pax/conf/experiment/ipd/gs_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,21 +29,46 @@ num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# GS vs. Tabular trained on seed=0, where Naive Learners have their learning rate annealed halfway through the trial
# run_path: ucl-dark/ipd/1gg0p92x
# model_path: exp/GS-PPO-vs-Tabular/run-seed-0-pop-size-1000/2022-09-28_01.57.34.854198/generation_4900

# GS vs. Tabular trained on seed=1, where Naive Learners have their learning rate annealed halfway through the trial
# run_path: ucl-dark/ipd/scffrmfv
# model_path: exp/GS-PPO-vs-Tabular/run-seed-1-pop-size-1000/2022-09-28_05.00.56.131987/generation_4900

# GS vs. Tabular trained on seed=2, where Naive Learners have their learning rate annealed halfway through the trial
# run_path: ucl-dark/ipd/2858x8sa
# model_path: exp/GS-PPO-vs-Tabular/run-seed-2-pop-size-1000/2022-09-28_07.38.37.221049/generation_4900

# GS vs. Tabular trained on seed=3, where Naive Learners have their learning rate annealed halfway through the trial
# run_path: ucl-dark/ipd/1y9tefvj
# model_path: exp/GS-PPO-vs-Tabular/run-seed-3-pop-size-1000/2022-09-28_01.57.40.696321/generation_4900

# GS vs. Tabular trained on seed=4, where Naive Learners have their learning rate annealed halfway through the trial
# run_path: ucl-dark/ipd/8j6zmb6h
# model_path: exp/GS-PPO-vs-Tabular/run-seed-4-pop-size-1000/2022-09-28_05.11.49.206169/generation_4900

# GS vs. Tabular trained on seed = 0
# run_path: ucl-dark/ipd/tywwxijw
# model_path: exp/GS-PPO-vs-Tabular/run-seed-0-pop-size-1000/2022-09-25_16.06.55.715665/generation_4900
# GS vs. Tabular trained on seed = 1
# run_path: ucl-dark/ipd/2lyn9n10
# model_path: exp/GS-PPO-vs-Tabular/run-seed-1-pop-size-1000/2022-09-25_16.07.48.978281/generation_4900
# GS vs. Tabular trained on seed = 2
# run_path: ucl-dark/ipd/f2xhuhcz
# model_path: exp/GS-PPO-vs-Tabular/run-seed-2-pop-size-1000/2022-09-25_16.08.35.015944/generation_4900
# GS vs. Tabular trained on seed = 3
# run_path: ucl-dark/ipd/16wzxeb6
# model_path: exp/GS-PPO-vs-Tabular/run-seed-3-pop-size-1000/2022-09-25_16.09.01.274669/generation_4900
# GS vs. Tabular trained on seed = 4
# run_path: ucl-dark/ipd/3dzkof3f
# model_path: exp/GS-PPO-vs-Tabular/run-seed-4-pop-size-1000/2022-09-25_16.41.50.643263/generation_4900


# hard stop
run_path: ucl-dark/ipd/2kyx0680
model_path: exp/EARL-PPO-vs-Tabular/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-28_14.35.25.775807/generation_4900

# PPO agent parameters
ppo:
Expand Down
18 changes: 0 additions & 18 deletions pax/conf/experiment/ipd/mfos_v_ppo.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,24 +26,6 @@ num_generations: 5000
total_timesteps: 1e11
num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
18 changes: 0 additions & 18 deletions pax/conf/experiment/ipd/mfos_v_ppo_mem.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,24 +26,6 @@ num_generations: 5000
total_timesteps: 1e11
num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900

# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
33 changes: 18 additions & 15 deletions pax/conf/experiment/ipd/mfos_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,22 +28,25 @@ num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# MFOS vs. Tabular trained on seed = 0
# run_path: ucl-dark/ipd/xyq4feoj
# model_path: exp/MFOS-vs-Tabular/run-seed-0-pop-size-1000/2022-09-28_22.45.26.138403/generation_4900
# MFOS vs. Tabular trained on seed = 1
# run_path: ucl-dark/ipd/13srlkhp
# model_path: exp/GS-MFOS-vs-Tabular/run-seed-1-pop-size-1000/2022-09-25_20.33.11.352762/generation_4300
# MFOS vs. Tabular trained on seed = 2
# run_path: ucl-dark/ipd/3pfmqrpw
# model_path: exp/GS-MFOS-vs-Tabular/run-seed-2-pop-size-1000/2022-09-25_20.34.04.832865/generation_4400
# MFOS vs. Tabular trained on seed = 3
# run_path: ucl-dark/ipd/groh4iwx
# model_path: exp/GS-MFOS-vs-Tabular/run-seed-3-pop-size-1000/2022-09-25_20.36.02.555928/generation_4400
# MFOS vs. Tabular trained on seed = 4
# run_path: ucl-dark/ipd/26cqaqyc
# model_path: exp/GS-MFOS-vs-Tabular/run-seed-4-pop-size-1000/2022-09-25_20.38.01.382774/generation_4400

# hardstop
run_path: ucl-dark/ipd/32k6kn5v
model_path: exp/MFOS-vs-Tabular/run-seed-0-pop-size-1000/2022-09-28_14.54.58.821778/generation_4900
# PPO agent parameters
ppo:
num_minibatches: 4
Expand Down
30 changes: 15 additions & 15 deletions pax/conf/experiment/mp/earl_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,21 +28,21 @@ num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# seed = 0
# run_path: ucl-dark/mp/36w1tuju
# model_path: exp/GS-PPO_memory-vs-Tabular/run-seed-0-pop-size-1000/2022-09-25_15.52.33.707967/generation_4900
# seed = 1
# run_path: ucl-dark/mp/3074jksy
# model_path: exp/GS-PPO_memory-vs-Tabular/run-seed-1-pop-size-1000/2022-09-25_15.53.12.637138/generation_4900
# seed = 2
# run_path: ucl-dark/mp/1h18aq5c
# model_path: exp/GS-PPO_memory-vs-Tabular/run-seed-2-pop-size-1000/2022-09-25_15.53.49.913885/generation_4900
# seed = 3
# run_path: ucl-dark/mp/3ew2bidu
# model_path: exp/GS-PPO_memory-vs-Tabular/run-seed-3-pop-size-1000/2022-09-25_15.54.11.059838/generation_4900
# seed = 4
run_path: ucl-dark/mp/glpfg7zd
model_path: exp/GS-PPO_memory-vs-Tabular/run-seed-4-pop-size-1000/2022-09-25_15.55.16.444095/generation_4900

# PPO agent parameters
ppo:
Expand Down
32 changes: 17 additions & 15 deletions pax/conf/experiment/mp/gs_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,21 +28,23 @@ num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# seed = 0
# run_path: ucl-dark/mp/11v18zvw
# model_path: exp/GS-PPO-vs-Tabular/run-seed-0-pop-size-1000/2022-09-25_15.56.30.500721/generation_4900
# # seed = 1
# run_path: ucl-dark/mp/krvvohwg
# model_path: exp/GS-PPO-vs-Tabular/run-seed-1-pop-size-1000/2022-09-25_15.57.21.586407/generation_4900
# # seed = 2
# run_path: ucl-dark/mp/1yjlt34b
# model_path: exp/GS-PPO-vs-Tabular/run-seed-2-pop-size-1000/2022-09-25_15.58.28.437732/generation_4900
# # seed = 3
# run_path: ucl-dark/mp/20q2wjkp
# model_path: exp/GS-PPO-vs-Tabular/run-seed-3-pop-size-1000/2022-09-25_16.03.58.776658/generation_4900
# seed = 4
run_path: ucl-dark/mp/3gnmuhf6
model_path: exp/GS-PPO-vs-Tabular/run-seed-4-pop-size-1000/2022-09-25_16.04.41.243205/generation_4900



# PPO agent parameters
ppo:
Expand Down
30 changes: 15 additions & 15 deletions pax/conf/experiment/mp/mfos_v_tabular.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,21 +28,21 @@ num_devices: 1

# Evaluation
num_seeds: 20
# # EARL vs. PPO trained on seed=0
# run_path: ucl-dark/ipd/13o3v95p
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-0-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.31.908871/generation_2900
# EARL vs. PPO trained on seed=1
# run_path: ucl-dark/ipd/dopodr9n
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-1-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.15.58.912526/generation_2900
# EARL vs. PPO trained on seed=2
# run_path: ucl-dark/ipd/265ftn32
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-2-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.17.00.437954/generation_2900
# EARL vs. PPO trained on seed=3
# run_path: ucl-dark/ipd/1hffijy2
# model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-3-OpenES-pop-size-1000-num-opps-1/2022-09-15_00.19.01.387867/generation_2900
# EARL vs. PPO trained on seed=25
run_path: ucl-dark/ipd/1ui7wfop
model_path: exp/EARL-PPO_memory-vs-PPO/run-seed-25-OpenES-pop-size-1000-num-opps-1/2022-09-15_02.32.16.559924/generation_2900
# seed=0
run_path: ucl-dark/mp/2erj6940
model_path: exp/GS-MFOS-vs-Tabular/run-seed-0-pop-size-1000/2022-09-28_01.55.03.327061/generation_4900
# seed=1
run_path: ucl-dark/mp/fj2ximl5
model_path: exp/GS-MFOS-vs-Tabular/run-seed-1-pop-size-1000/2022-09-28_01.55.09.949508/generation_4900
# seed=2
run_path: ucl-dark/mp/vcf845u1
model_path: exp/GS-MFOS-vs-Tabular/run-seed-2-pop-size-1000/2022-09-28_01.55.17.814611/generation_4900
# seed=3
run_path: ucl-dark/mp/5c3buqf2
model_path: exp/GS-MFOS-vs-Tabular/run-seed-3-pop-size-1000/2022-09-28_01.55.31.913577/generation_4900
# seed=4
run_path: ucl-dark/mp/1vrh77zx
model_path: exp/GS-MFOS-vs-Tabular/run-seed-4-pop-size-1000/2022-09-28_06.32.22.210096/generation_4900

# PPO agent parameters
ppo:
Expand Down
Loading