Moritzm00's group workspace
Group: Baseline
Name
9 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
architecture
augment
augment.active
augment.factors.random_brightness
augment.factors.random_flip
augment.factors.random_rotation
augment.factors.random_translation
augment_gpu
batch_size
callbacks.ckpt_filepath
callbacks.early_stopping_patience
callbacks.model_ckpt
callbacks.monitor
callbacks.use_wandb
callbacks.visualize_predictions
callbacks.with_wandb_ckpt
ckpt_dir
data_dir
dataset.channels
dataset.cutoff_age
dataset.img_size
dataset.name
dataset.path
decay_rate
decay_steps
early_stopping_mode
early_stopping_monitor
early_stopping_patience
epochs
finetune_whole_model
initial_learning_rate
log_dir
loss
lr_schedule
lr_schedule.stage_1._target_
lr_schedule.stage_1.decay_rate
lr_schedule.stage_1.decay_steps
lr_schedule.stage_1.initial_learning_rate
lr_schedule.stage_2._target_
lr_schedule.stage_2.decay_rate
lr_schedule.stage_2.decay_steps
lr_schedule.stage_2.initial_learning_rate
model.architecture
model.finetune_base
Finished
Baseline CNN Model with image augmentation (tf.data API)
The image augmentation with tf.data is much faster, leading to an improvement from over 200s per epoch to just 40s per epoch (this run)
moritzm00
Baseline
Image Augmentation
tf.data API
18m 56s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
Baseline CNN Model with image augmentation (tf.data API)
moritzm00
Baseline
Image Augmentation
tf.data API
11m 58s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
Baseline CNN Model with image augmentation (tf.data API)
moritzm00
Baseline
Image Augmentation
tf.data API
19m 33s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
Baseline CNN Model with image augmentation but no sample weighting
moritzm00
Baseline
Image Augmentation
1h 25m 59s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
no image augmentation
moritzm00
Baseline
51m 30s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
Baseline with sample weighting
moritzm00
8m 58s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
-
moritzm00
15m 41s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_mae
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
-
moritzm00
Baseline
12m 58s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_loss
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
Finished
-
moritzm00
Baseline
7m 33s
-
Simple CNN
-
-
-
-
-
-
-
32
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.96
100000
min
val_loss
5
50
-
0.001
-
mean_absolute_error
ExponentialDecay
-
-
-
-
-
-
-
-
-
-
1-9
of 9