Alcazar90's workspace
Runs
17
State
Notes
User
Tags
Created
Runtime
Sweep
batch_size
clip_advantages
clip_ratio
ddpm_ckpt
eval_every_each_epoch
eval_rnd_seed
initial_lr
lr
num_batches
num_epochs
num_eval_samples
num_inference_steps
num_inner_epochs
num_samples_per_epoch
peak_lr
punishment
run_seed
save_model
task
threshold
warmup_pct
weight_decay
KL (current vs old policy)
average loss
batch
epoch
eval_mean_reward
inner_epoch
learning_rate
loss
max_reward
mean_reward
min_reward
pct_clipped_ratios
std_reward
Finished
Resuming training from W&B artifact: alcazar90/ddpo-over50-ddpm-celebahq256/Task.OVER50-flowing-puddle-16:v8
alcazar90
46m 44s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
9.4610e-7
-
10
10
64
40
1
100
9.4610e-7
-1
7136603
true
Task.OVER50
0.6
-
0.0001
9597.64844
0.2979
9
9
8.66386
0
9.4610e-7
0.28174
12.77805
9.26327
-12.95691
1
4.75499
Finished
Resuming training from W&B artifact: alcazar90/ddpo-over50-ddpm-celebahq256/Task.OVER50-glad-frog-15:v6
alcazar90
47m 1s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
9.4610e-7
-
10
10
64
40
1
100
9.4610e-7
-1
8139403
true
Task.OVER50
0.6
-
0.0001
11635.1084
0.32768
9
9
7.92053
0
9.4610e-7
0.37343
12.6593
8.87933
-9.5538
1
4.8885
Finished
Resuming training from W&B artifact: alcazar90/ddpo-over50-ddpm-celebahq256/Task.OVER50-peachy-water-12:v17
alcazar90
38m 9s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
9.4610e-7
-
10
8
64
40
1
100
9.4610e-7
-1
31239403
true
Task.OVER50
0.6
-
0.0001
12582.44336
0.32042
9
7
6.62528
0
9.4610e-7
0.36224
12.607
8.139
-10.48306
1
5.19129
Finished
Resuming training from W&B artifact: alcazar90/ddpo-over50-ddpm-celebahq256/Task.OVER50-dainty-fire-9:v18
alcazar90
1h 57m 45s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
6.4700e-7
-
10
25
64
40
1
100
0.00000345
-1
231239403
true
Task.OVER50
0.6
0.65
0.0001
2266.88696
0.35726
9
24
5.36473
0
6.5779e-8
0.40319
12.90304
5.91985
-12.68374
1
6.70227
Finished
Resuming training from W&B artifact: alcazar90/ddpo-over50-ddpm-celebahq256/Task.OVER50-twilight-glitter-8:latest
alcazar90
1h 57m 54s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
3.4500e-7
-
10
25
64
40
1
100
0.00000345
-1
91231239403
true
Task.OVER50
0.6
0.55
0.0001
1990.3363
0.37838
9
24
0.24858
0
3.5160e-8
0.41544
12.83589
1.23422
-12.47254
1
7.97319
Finished
-
alcazar90
1h 57m 46s
-
10
10
0.0001
google/ddpm-celebahq-256
1
650
9.0000e-8
-
10
25
64
40
1
100
0.00000345
-1
92013491249214130
true
Task.OVER50
0.6
0.45
0.0001
1695.172
0.32435
9
24
-5.1751
0
9.4458e-9
0.35734
12.06845
-5.16784
-12.8278
1
6.49851
Finished
-
alcazar90
12m 38s
-
5
1.5
0.0001
google/ddpm-celebahq-256
-
-
-
0.00001
2
15
-
40
1
10
-
0
-
-
Task.OVER50
0.8
-
0.0001
-
0
1
14
-
0
-
0
0
0
0
-
0
1-7
of 7