ºtºâªk½Òµ{_Machine Learning_01_DNN
ºtºâªk½Òµ{_Machine Learning_02_CNN
2 |
|
3
|
1
|
8
|
4
|
7
|
6
|
5
|
1
|
2
|
3
|
8 |
|
4
|
7 |
6
|
5
|
ÃD¥Ø: ²`«×¾Ç²ß¤¤¤å¤â¼g¼Æ¦r¿ëÃÑ
¿é¤J: ¤¤¤å¤â¼g¼Æ¦r¸ê®Æ¶°(dataset)
¿é¥X: ¤¤¤å¤â¼g¼Æ¦r¿ëÃѷǽT²v(accuracy)
¦û¤À: Á`¦¨ÁZ12%
¥Øªº: Åý¾Ç¥Í½m²ß¨Ï¥Î²`«×¾Ç²ß(deep learning)¼Ò«¬¸Ñ¨M¤¤¤å¤â¼g¼Æ¦r¿ëÃÑ°ÝÃD¡C²`«×¾Ç²ß(deep learning)¼Ò«¬¥i¬°²`«×¯«¸gºô¸ô(deep neural network, DNN)¡B±²¿n¯«¸gºô¸ô(convolutional neural network, CNN)¡Bªøµu´Á°O¾Ð(long short-term memory, LSTM)¯«¸gºô¸ô©Î¹h±±»¼°j³æ¤¸(gated recurrent unit, GRU)¯«¸gºô¸ô¡C
¸ê®Æ¶°»¡©ú: (¸ê®Æ¶°¤U¸ü³sµ²)
¸ê®Æ¶°ÀÉ®×handwrite_detect.zip¬°¤¤¤å¤â¼g¼Æ¦r¸ê®Æ¶°¡A¥]§t¤F°V½m¸ê®Æ¶°(training dataset)»P´ú¸Õ¸ê®Æ¶°(test dataset)
°V½m¸ê®Æ¦@¦³2450µ§
´ú¸Õ¸ê®Æ¦@¦³1700µ§
µû¤À¶µ¥Ø»P³W½d:
Source code(.py)
A short report (.pdf) (§t¦³)
¤¤¤å¤â¼g¿ëÃѷǽT²v(accuracy)¡A¥HºI¹Ï¤è¦¡§e²{
Source code¤§³v¦æ¸ÑÄÀ
ú¥æ¤è¦¡: ½Ð±N©Ò¦³ÀÉ®×À£ÁY¦¨ .zip «á¡A¦A¤W¶Ç¦Ü e-class §@·~ú¥æ°Ï
(¤W¶ÇÀɦW:Term_Project_¾Ç¸¹_©m¦W.zip)
ú¥æ®É¶¡: ´Á¥½¦Ò©P¤§¤U¤@©P¤G 20:00«eºI¤î
³Æµù : ¤Á¤Å¤¬¬Û§Ûŧ¡AYµo²{§Ûŧ¡A«h¥H0¤Àpºâ
ºtºâªk½Òµ{_Machine Learning_01_DNN
ºtºâªk½Òµ{_Machine Learning_02_CNN
Task A |
Task B |
Task C |
|
Tim |
$1 |
$2 |
$3 |
Bob |
$2 |
$3 |
$2 |
Alex |
$2 |
$2 |
$3 |