MNIST 데이터 셋을 시각화 해보기

딥러닝을 공부하면서 어떤 데이터를 이용하는지 파악하는 중에 데이터를 시각화 해봤습니다.

28*28 사이즈의 손글씨 이미지를 분석한다고해서 데이터셋 자체가 이미지일거란 생각을 했는데요 실제 예제를 보고 데이터를 출력해보니 0~1사이의 데이터였다는것을 확인하게 되었습니다.

제가 테스트한 예제는 이 블로그에서 참고하였고 이 블로그는 이 동영상강의를 참고하였다고 합니다.

---

일단 데이터를 출력해 봤습니다

총 28*28로 784개의 데이터 입니다

[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.41568631 0.89803928 1. 0.94117653 0.44705886 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.227451 0.38039219 0.90588242 0.98431379 0.99215692 0.99215692 0.99215692 0.99215692 0.65098041 0.227451 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.06666667 0.66274512 0.92941183 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.92941183 0.10980393 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.06666667 0.67843139 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.87450987 0.99215692 0.99215692 0.99215692 0.32156864 0.04705883 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.29803923 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.50196081 0.14901961 0.627451 0.99215692 0.99215692 0.99215692 0.57647061 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.28235295 0.98039222 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.69803923 0.2392157 0.02745098 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.4784314 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.88627458 0.30980393 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.1254902 0.61960787 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.65098041 0.56470591 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.01568628 0.13333334 0.64313728 0.64313728 0.64313728 0.20392159 0.10588236 0.07450981 0.71764708 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.10980393 0.99215692 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.21176472 0.99215692 0.99215692 0.99215692 0.99215692 0.74901962 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.18431373 0.8705883 0.99215692 0.99215692 0.99215692 0.99215692 0.40000004 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.17254902 0.87843144 0.99215692 0.99215692 0.99215692 0.99215692 0.79215693 0.09411766 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.17254902 0.89803928 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.13333334 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.56862748 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.82745105 0.07843138 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.16862746 0.85098046 0.98039222 0.99215692 0.99215692 0.99215692 0.99215692 0.83529419 0.10588236 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.04313726 0.89019614 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.99215692 0.50588238 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.04705883 0.91372555 0.99215692 0.99215692 0.99215692 0.99215692 0.97647065 0.81568635 0.15294118 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.19607845 0.9333334 0.99215692 0.99215692 0.99215692 0.83921576 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.20000002 0.64313728 0.99215692 0.59607846 0.16078432 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ]

그리고 이데이터는 아래의 값에 해당합니다 "9"

[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]

---

약간의 노가다를 통해 28*28의 구성을 한후 필드의 숫자를 기준으로 가중치를 줘봤습니다

[엑셀 링크]

어떤 숫자인지 눈으로 확인 되시나요?

---