Skip to main content
ariq913
Projects
hyperparam-test
Reports
Hyper-Parameter tuning
Log in
Sign up
Share
Comment
1 star
Share
Comment
1 star
Hyper-Parameter tuning
Hyper-parameter tuning on Roberta model on climate-tweet dataset
ARIQ RAHMAN
Created on December 11
|
Last edited on December 11
Comment
Section 1
Parameter importance with respect to
accuracy
Parameters
1-4
of 4
Config parameter
Importance
Correlation
freezelayer
freezelayer
num_freeze_layers
num_freeze_layers
units
units
learning_rate
learning_rate
Loading...
null
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
num...
null
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
fre...
null
0.000000
0.000002
0.000004
0.000006
0.000008
0.000010
0.000012
0.000014
0.000016
0.000018
0.000020
0.000022
0.000024
0.000026
0.000028
0.000030
lea...
null
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
units
null
0.22
0.24
0.26
0.28
0.30
0.32
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
acc...
val_accuracy
val_accuracy
0
200m
400m
600m
800m
1
Step
0.25
0.3
0.35
0.4
accuracy
accuracy
0
200m
400m
600m
800m
1
Step
0.24
0.26
0.28
0.3
0.32
0.34
0.36
0.38
loss
loss
0
200m
400m
600m
800m
1
Step
1.35
1.4
1.45
1.5
val_loss
val_loss
0
200m
400m
600m
800m
1
Step
1.25
1.3
1.35
1.4
1.45
Run set
4
Add a comment