Loading...
main logo

ScalaAnalytics

ScalaAnalytics Weather Prediction

Data Model

  • 10 year long weather data of Seoul (Sep 7th, 2007~ Sep 7th, 2017)
  • 7 year long data is separated for training and the rest 3 year data is used for testing

Network Model

  • Keras regression model (Tensorflow backend)
  • Training epochs : 200
  • batch size : 32
  • optimizer : adam.

      • Comments : The input dimension is 43.
      We eliminated blank data from the dataset we downloaded.
      We still found blank data in some of its training set and filled them with mean values.
      I guess the prediction errors on highest temperature prediction are caused by those blank data.
      Except those errors, mean accuracy is almost +- 0.5 norm. Celcius.

ScalaAnalytics Breast Cancer Prediction

Data Model : Dr. William H. Wolberg of the University of Wisconsin Hospitals, Madison, U.S.A (UCI Lab.)

Related Studies:

  1. Wolberg,W.H., & Mangasarian,O.L. (1990). Multisurface method of pattern separation for medical diagnosis applied to breast cytology. In Proceedings of the National Academy of Sciences, 87, 9193-9196.
  2. Zhang,J. (1992). Selecting typical instances in instance-based learning. In Proceedings of the Ninth International Machine Learning Conference (pp. 470-479). Aberdeen, Scotland: Morgan Kaufmann.

Network Model

  • Keras regression model (Tensorflow backend)
  • Training epochs : 300
  • batch size : 16
  • optimizer : adam.
# Attribute Domain
  1. Sample code number id number
  2. Clump Thickness 1 - 10
  3. Uniformity of Cell Size 1 - 10
  4. Uniformity of Cell Shape 1 - 10
  5. Marginal Adhesion 1 - 10
  6. Single Epithelial Cell Size 1 - 10
  7. Bare Nuclei 1 - 10
  8. Bland Chromatin 1 - 10
  9. Normal Nucleoli 1 - 10
  10. Mitoses 1 - 10
  11. Class: (2 for benign, 4 for malignant)

ScalaAnalytics Stock Market Prediction

Data Model : Apple and Amazon original 5 year long daily data from Yahoo https://finance.yahoo.com/

Network Model : LSTM RNN

Goal : Predicts next 7 days of stock market trends And error rates were like these:

  1. AAPL Apple RSME: 0.008078 MSE: 0.000065
  2. AMZN Amazon RSME: 0.012324 MSE: 0.000152

ScalaAnalytics Epileptic Seizure Prediction

Data Model : Qiuyi Wu (qw9477'@'rit.edu) and Ernest Fokoue (epfeqa'@'rit.edu) of School of Mathematical Sciences, Rochester Institute of Technology, NY, U.S.A. (Date Donated: May 24th, 2017).

Network Model

  • Keras regression model (Tensorflow backend)
  • Training epochs : 120
  • batch size : 64
  • Results The accuracy is almost 97.7% and loss is 0.0768
This dataset has 5 classes:
  1. Having an epileptic seizure
  2. They recorder the EEG from the area where the tumor was located
  3. They identify where the region of the tumor was in the brain and recording the EEG activity from the healthy brain area
  4. eyes closed, means when they were recording the EEG signal the patient had their eyes closed
  5. eyes open, means when they were recording the EEG signal of the brain the patient had their eyes open
From Wikipedia Electroencephalography
https://en.wikipedia.org/wiki/Electroencephalography

ScalaAnalytics Face Image Reconstructions

부분 촬영된 영상을 이용하여 정면 얼굴 이미지를 생성해 내는 프로젝트

다음과 같이 얼굴이 가려진 경우에도 개략적으로 추정하여 얼굴을 생성할 수 있음

  1. the person is wearing sunglasses,
  2. shot angle of the CCTV toward the person
  3. CCTV sensors types are color or BW
  4. CCTV records are noisy or not clean images

Samples 1

Samples 2

ScalaAnalytics Similar Image Searching

입력된 이미지와 가장 유사한 이미지를 찾아내는 Image Search 프로젝트

  • 204,721 images MS COCO dataset images into one library
  • CCTV 등에서 유사한 이미지 탐색
  • Object Tracking

Samples 1

Samples 2

Contact (연락처)

Call Us
+82 2 6956 3946
Fax
+82 2 6956 3947
Email
scalawox@scalawox.com
Address
서울시 금천구 가산디지털2로 70, 13층 1310호(가산동, 대륭테크노타운 19차)