rename directories

docusaurus.config.js

@@ -188,6 +188,10 @@ export default {
             to: '/docs/tutorials/accessing-data',
             from: ['/data', '/docs/data']  // obscure fix for /data/ vs /data redirect(?)
           },
+          {
+            to: '/science/category/recent-projects',
+            from: ['/science']  // obscure fix for /data/ vs /data redirect(?)
+          },
         ],
       },
     ], [

news/2023-05-31-scalable-ci-in-aps.md

@@ -4,4 +4,4 @@ hide_reading_time: true
 tags: [edge computing, computer science, computational science, Argonne APS, Sage science]
 ---
 
-Earlier this year, Sage computer science researchers and computational scientists from Argonne's Advanced Photon Source (APS) collaborated to answer the question: Can edge computing be used in X-ray beamline experiments to process high-volume and fast data streams for real-time decision making?  Read more about the experiment [here](/science/scalable-ci-in-aps) under [Sage Science](/science).
+Earlier this year, Sage computer science researchers and computational scientists from Argonne's Advanced Photon Source (APS) collaborated to answer the question: Can edge computing be used in X-ray beamline experiments to process high-volume and fast data streams for real-time decision making?  Read more about the experiment [here](/science/recent/scalable-ci-in-aps) under [Sage Science](/science/category/recent-projects).

science/Other projects/_category_.yml → science/past/_category_.yml

@@ -1,3 +1,4 @@
 position: 2
+label: Past projects
 link:
   type: generated-index

science/Other projects/integrating-chameleon.md → science/past/integrating-chameleon.md

@@ -15,11 +15,11 @@ With this image, the Sage team created a customizable, innovative, and user frie
 That’s where the Chameleon platform comes in. Chameleon is a testbed of computational resources that are made available for the scientific community for running experiments. Chameleon has hardware available for Sage users at three different locations:CHI@UC, CHI@TACC, KVM@TACC. The CHI@UC and CHI@TACC site are both bare-metal sites, so only physical hardware is available. While the KVM@TACC site, is a virtual machine site. Users simply have to, create a reservation on the Chameleon, launch an instance, and attach an ip address to their instance. After doing these three steps, users are ready to start using there very own Chameleon machine.
 
 
-![creating chameleon instance](../imgs//integrating-chameleon-1.png)
-![creating chameleon instance 2](../imgs//integrating-chameleon-2.png)
+![creating chameleon instance](../imgs/integrating-chameleon-1.png)
+![creating chameleon instance 2](../imgs/integrating-chameleon-2.png)
 > Creating a reservation and instance taken from Chameleon.<sup>[[1](#references)]</sup>
 
-![allocating ips](../imgs//integrating-chameleon-3.png)
+![allocating ips](../imgs/integrating-chameleon-3.png)
 > Allocating floating ip.<sup>[[1](#references)]</sup>
 
 
@@ -56,7 +56,7 @@ Additionally, using GitHub Actions, I streamlined the the process of creating th
 
 Sage recognizes that users have all different sets of needs when running neural networks. That’s why users will have options to use different specialized versions/flavors of the Chameleon-Sage image. There are currently three different types of Chameleon-Sage images all available on the Chameleon catalog.
 
-![chameleon sage image](../imgs//integrating-chameleon-6.png)
+![chameleon sage image](../imgs/integrating-chameleon-6.png)
 > Snapshot of Sage image on Chameleon.<sup>[[1](#references)]</sup>
 
 Sage-VirtualWaggle-Ubuntu18.04 – This is the standard Chameleon-Sage image that is available at all three Chameleon sites. This image includes all of the previously stated features and runs with Ubuntu 18.04.

science/Recent Projects/_category_.yml → science/recent/_category_.yml

@@ -1,3 +1,4 @@
 position: 1
+label: Recent projects
 link:
   type: generated-index

science/Recent Projects/traffic-state-estimation.md → science/recent/traffic-state-estimation.md

@@ -1,11 +1,12 @@
+
 # Traffic State Estimation
 
 A real-time vehicle tracking application based on deep neural networks was utilized to calculate traffic state. The deep neural networks that were utilized for the estimation were introduced in the ‘Vehicle Tracking’ post. The traffic state has been calculated by ‘traffic flow = traffic speed x traffic density’ (traffic_char1996, edie1963). With the equation, we need to measure two of the parameters using instruments for each parameters to estimate the third state. However, on the other hand, the application introduced in this study based on deep neural networks can calculate traffic state independently using a video source.
 
 We tried to estimate traffic states directly from the information that can be extracted from videos. The traffic state, that are traffic volume, space occupancy, travel speed, travel time and travel delays can be categorized into three types of information: traffic speed, flow and density (traffic_char1996). We calculated the three traffic state, flow, speed and density, which can be represented as traffic volume, travel speed or time and space occupancy respectively. The image processing algorithm required videos that were recorded with frame rate higher than 12 fps. The calculation result reporting frequency can be determined based on the demand of user. The default reporting frequency is every 30 second.
 
 
-![traffic static space occupancy box](../imgs//traffic-state-estimation-1.png)
+![traffic static space occupancy box](../imgs/traffic-state-estimation-1.png)
 
 ## Traffic Flow
 Traffic flow was defined as the number of vehicles that passed a certain line on a road per unit time. Traffic flow was estimated by counting how many vehicles passes the bottom line or top line of the orange box in the figure above for outgoing and incoming traffic respectively for the total length of the videos.  The number of vehicles that passed through the lines of the orange box were counted for the total length of the video, and that was averaged into the number of passed vehicles per second. The raw counting number also can be reported along with the averaged traffic flow. Because the vehicle tracking algorithm was able to identify each vehicles with identification number, we only counted each vehicle once while we processing the videos.