This repository has been archived by the owner on Aug 8, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
5 changed files
with
9 additions
and
9 deletions.
There are no files selected for viewing
Binary file modified
BIN
+41 Bytes
(100%)
...SKIN LESION CLASSIFICATION ON IMBALANCED DATA USING DEEP LEARNING WITH SOFT ATTENTION.pdf
Binary file not shown.
Binary file modified
BIN
+14 Bytes
(100%)
...SKIN LESION CLASSIFICATION ON IMBALANCED DATA USING DEEP LEARNING WITH SOFT ATTENTION.rar
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -98,9 +98,9 @@ | |
% Affiliations / Addresses (Add [1] after \address if there is only one affiliation.) | ||
|
||
\address{ | ||
$^{1}$ \quad School of Electrical and Electronic Engineering, Hanoi university of Science and Technology, Vietnam; [email protected]\\ | ||
$^{1}$ \quad School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, Vietnam; [email protected]\\ | ||
$^{2}$ \quad Faculty of Information Technology, University of Transport and Communications, Vietnam; [email protected]\\ | ||
$^{3}$ \quad School of Electrical and Electronic Engineering, Hanoi university of Science and Technology, Vietnam; [email protected]} | ||
$^{3}$ \quad School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, Vietnam; [email protected]} | ||
|
||
% Contact information of the corresponding author | ||
\corres{Correspondence: [email protected]; Tel.: +84-9834-443-22 (N.V.D.)} | ||
|
@@ -389,7 +389,7 @@ \subsubsection{Loss Function} | |
\section{Results} | ||
\subsection{Experimental Setup} | ||
\subsubsection{Training} | ||
Before training, the data set is split into two sub set for training (90 percent) and validation (10 percent). Test set, otherwise is provided by the HAM10000 data set, contains 857 images. To analyze the effect of augmented data on the model, during the training the image data is augmented by the following technique:\\ | ||
Before training, the data set is split into two sub set for training (90 percent) and validation (10 percent). Test set, otherwise is provided by the HAM10000 data set, contains 857 images. To analyze the effect of augmented data on the model, before the training the image data is augmented to 53573 images by the following technique:\\ | ||
- Rotation Range: rotate the image in a fixed angle. \\ | ||
- Width and height shift range: Shift the image horizontally and vertically, respectively. \\ | ||
- Zoom Range: Zoom in or zoom out the image to create new image. \\ | ||
|
@@ -459,19 +459,19 @@ \subsection{Discussion} | |
\hline | ||
DenseNet201 & 0.84 & \textbf{0.89}\\ | ||
\hline | ||
ResNet50 & 0.76 & 0.70 \\ | ||
ResNet50 & 0.76 & 0.70\\ | ||
\hline | ||
ResNet152 & - & 0.57\\ | ||
ResNet152 & 0.81 & 0.57\\ | ||
\hline | ||
NasNetLarge & - & 0.84\\ | ||
\hline | ||
MobileNetV2 & - & 0.81\\ | ||
\hline | ||
MobileNetV3Small & - & 0.78\\ | ||
\hline | ||
MobileNetV3Large & - & \textbf{0.86}\\ | ||
MobileNetV3Large & 0.85 & \textbf{0.86}\\ | ||
\hline | ||
NasNetMobile & - & \textbf{0.86}\\ | ||
NasNetMobile & 0.84 & \textbf{0.86}\\ | ||
\hline | ||
\end{tabular} | ||
\caption{Accuracy of all models. ACC stands for accuracy. AD stands for augmented data, this indicate that the model is trained with augmented data. MD stands for Metadata which indicate that the model is trained with Metadata} | ||
|