Differences

This shows you the differences between two versions of the page.

--- intern:construction_har [2022/08/11 02:18] – kenzjk
+++ intern:construction_har [2022/08/13 10:28] (current) – kenzjk
@@ Line 5: / Line 5: @@
 ==== 2. Reading (can make subsections for notes) ====
 // Human Activity Recognition using Wearable Sensors by Deep Convolutional Neural Networks //
-> SVM method has a relatively high accuracy but a relatively high computational cost.
+  * SVM method has a relatively high accuracy but a relatively high computational cost.
-> Feature Selection method has a relatively low computational cost but a significantly low accuracy.
+  * Feature Selection method has a relatively low computational cost but a significantly low accuracy.
-> A trade-off exists between accuracy and computational cost.
+  * A trade-off exists between accuracy and computational cost.
-> DCNN+ and DCNN methods achieve a good balance between the two (accuracy and efficiency).
+  * DCNN+ and DCNN methods achieve a good balance between the two (accuracy and efficiency).
 ====3. Coding (w/ description) ====
 ==== 4. Research Progress (update every other day) ====
-> Aug/02/22
+__Aug/02/22__
   * Learned about the basics of Human Activity Recognition and gathered that 3D motion sensors often utilized accelerometers and gyroscopes send motion signals to a laptop with code ready to interpret the signals to classify the motion.
   * Research on Xsens and the different modules of sensors it offered was done.
 __Aug/04/22__
-  * Looked into the modules MTw Awinda and Xsens DOT.
+  * Looked into the modules MTw Awinda (https://www.xsens.com/products/mtw-awinda) and Xsens DOT (https://www.xsens.com/xsens-dot).
-  * Created a document detailing the differences between the two to aid my professor in deciding which one to purchase.\\
+  * Created a document detailing the differences between the two to aid my professor in deciding which one to purchase.
-> Aug/06/22
+__Aug/06/22__
   * Attended weekly online meeting with UC Berkeley students and professors and exchanged our findings regarding the most suitable motion sensors for our project.
   * Also discussed common actions by construction workers (i.e. excavating, laying bricks, climbing, soldering, etc.), which our final algorithm would have to be able to distinguish.
 __Aug/08/22__
   * Studied // Human Activity Recognition using Wearable Sensors by Deep Convolutional Neural Networks // and discussed this paper with my professor.
+  * Did research and watched lecture videos on common libraries implemented by machine learning programs, such as SciKit-Learn (https://www.youtube.com/watch?v=0Lt9w-BxKFQ&t=904s), NumPy (https://www.youtube.com/watch?v=zoLHuhefMNw), MatPlotLib (https://www.youtube.com/watch?v=3Xc3CA655Y4), etc.
 __Aug/10/22__
-  * Did research on some common libraries implemented by machine learning programs, such as SciKit-Learn, NumPy, MatPlotLib, etc.
   * Coded an iris flower-distinguishing program using Sklearn.
   * Began to code and train a standing-or-walking-distinguishing program as a first step to our final program to be implemented at construction sites.
 __Aug/12/22__
+  * Successfully read in all the data from the UCI HAR Dataset (https://archive.ics.uci.edu/ml/datasets/human+activity+recognition+using+smartphones) into my machine learning program.
+  * Studied an open source HAR algorithm on GitHub (https://github.com/ma-shamshiri/Human-Activity-Recognition/tree/main/code) to familiarize myself with typical HAR programs.
 __Aug/14/22__
+  * Successfully trained my program using the UCI HAR Dataset, but prediction accuracy was not ideal.
+  * Experimented with common HAR classifiers seen on GitHub to try to find better prediction accuracy.
-==== 4.1 Final Report PPT ====
+==== 5. Final Report PPT ====