Browsing by Subject "Python"

Now showing 1 - 8 of 8
  • Thumbnail Image
    Item
    AI on the Edge with CondenseNeXt: An Efficient Deep Neural Network for Devices with Constrained Computational Resources
    (2021-08) Kalgaonkar, Priyank B.; El-Sharkawy, Mohamed A.; King, Brian S.; Rizkalla, Maher E.
    Research work presented within this thesis propose a neoteric variant of deep convolutional neural network architecture, CondenseNeXt, designed specifically for ARM-based embedded computing platforms with constrained computational resources. CondenseNeXt is an improved version of CondenseNet, the baseline architecture whose roots can be traced back to ResNet. CondeseNeXt replaces group convolutions in CondenseNet with depthwise separable convolutions and introduces group-wise pruning, a model compression technique, to prune (remove) redundant and insignificant elements that either are irrelevant or do not affect performance of the network upon disposition. Cardinality, a new dimension to the existing spatial dimensions, and class-balanced focal loss function, a weighting factor inversely proportional to the number of samples, has been incorporated in order to relieve the harsh effects of pruning, into the design of CondenseNeXt’s algorithm. Furthermore, extensive analyses of this novel CNN architecture was performed on three benchmarking image datasets: CIFAR-10, CIFAR-100 and ImageNet by deploying the trained weight on to an ARM-based embedded computing platform: NXP BlueBox 2.0, for real-time image classification. The outputs are observed in real-time in RTMaps Remote Studio’s console to verify the correctness of classes being predicted. CondenseNeXt achieves state-of-the-art image classification performance on three benchmark datasets including CIFAR-10 (4.79% top-1 error), CIFAR-100 (21.98% top-1 error) and ImageNet (7.91% single model, single crop top-5 error), and up to 59.98% reduction in forward FLOPs compared to CondenseNet. CondenseNeXt can also achieve a final trained model size of 2.9 MB, however at the cost of 2.26% in accuracy loss. Thus, performing image classification on ARM-Based computing platforms without requiring a CUDA enabled GPU support, with outstanding efficiency.
  • Thumbnail Image
    Item
    Asking Data Analysis Questions with PandasAI
    (2023-11-08) Dolan, Levi
    As easily accessible AI models have increased in visibility, one area of interest for those working with datasets programmatically is how AI might streamline common data analysis tasks. The recently-released PandasAI library is a Python library that connects to an OpenAI model (known for ChatGPT) and allows users to ask natural language-style questions about dataframes created in Pandas syntax. This lightning talk demonstrates how to start exploring this data analysis method using sample World Bank and World Happiness Report data. Potential limitations are also discussed.
  • Thumbnail Image
    Item
    A Finite Element Model for Investigation of Nuclear Stresses in Arterial Endothelial Cells
    (2022-12) Rumberger, Charles B.; Ji, Julie; Tovar, Andres; Yokota, Hiroki
    Cellular structural mechanics play a key role in homeostasis by transducing mechanical signals to regulate gene expression and by providing adaptive structural stability for the cell. The alteration of nuclear mechanics in various laminopathies and in natural aging can damage these key functions. Arterial endothelial cells appear to be especially vulnerable due to the importance of shear force mechanotransduction to structure and gene regulation as is made evident by the prominent role of atherosclerosis in Hutchinson-Gilford progeria syndrome (HGPS) and in natural aging. Computational models of cellular mechanics may provide a useful tool for exploring the structural hypothesis of laminopathy at the intracellular level. This thesis explores this topic by introducing the biological background of cellular mechanics and lamin proteins in arterial endothelial cells, investigating disease states related to aberrant lamin proteins, and exploring computational models of the cell structure. It then presents a finite element model designed specifically for investigation of nuclear shear forces in arterial endothelial cells. Model results demonstrate that changes in nuclear material properties consistent with those observed in progerin-expressing cells may result in substantial increases in stress concentrations on the nuclear membrane. This supports the hypothesis that progerin disrupts homeostatic regulation of gene expression in response to hemodynamic shear by altering the mechanical properties of the nucleus.
  • Thumbnail Image
    Item
    Houseplant Advisor
    (2021) Griswold, Andrew E.; Magnabosco, Nadia E.; Freije, Elizabeth; Cooney, Elizabeth
    For capstone in the ECET department, it was expected for this group to create a houseplant advisor device. The device includes an LCD touchscreen, temperature/humidity sensor, and a light sensor. The device is expected to take a 24-hour scan of a certain area of a home. Then based off the average light, humidity, and temperature, it would recommend a plant that would do well in that specific area and display it on the touchscreen. The customer also wanted the device to have the option of logging the owned plants and to send a reminder when to water them.
  • Thumbnail Image
    Item
    Indianapolis Motor Speedway Display Project
    (2019-12-11) Shi, Charleston; Ibanez, Cristobal; Freije, Eliabeth; Kitchen, Jeff
    The Indianapolis Motor Speedway Museum and Xtrac want an interactive display of a racecar transmission and allows people of all ages to witness learn. Xtrac has commissioned IUPUI engineers to create a control box that correlates to other functions of a racecar and then correspond said box to a steering wheel, with additional features specified by the sponsor. Since a transmission has mostly a mechanical aspect, Mechanical Engineering Technology (MET) students are paired with a team of EET/CPET students. Any mounting and specification requirements are a part of the MET students’ project requirements. Details regarding electrical power, circuit design, and electromechanical integration will be generated by the EET students.
  • Thumbnail Image
    Item
    Plant Level IIoT Based Energy Management Framework
    (2023-05) Koshy, Liya Elizabeth; Chien, Stanley Yung-Ping; Chen, Jie; King, Brian
    The Energy Monitoring Framework, designed and developed by IAC, IUPUI, aims to provide a cloud-based solution that combines business analytics with sensors for real-time energy management at the plant level using wireless sensor network technology. The project provides a platform where users can analyze the functioning of a plant using sensor data. The data would also help users to explore the energy usage trends and identify any energy leaks due to malfunctions or other environmental factors in their plant. Additionally, the users could check the machinery status in their plant and have the capability to control the equipment remotely. The main objectives of the project include the following: • Set up a wireless network using sensors and smart implants with a base station/ controller. • Deploy and connect the smart implants and sensors with the equipment in the plant that needs to be analyzed or controlled to improve their energy efficiency. • Set up a generalized interface to collect and process the sensor data values and store the data in a database. • Design and develop a generic database compatible with various companies irrespective of the type and size. • Design and develop a web application with a generalized structure. Hence the database can be deployed at multiple companies with minimum customization. The web app should provide the users with a platform to interact with the data to analyze the sensor data and initiate commands to control the equipment. The General Structure of the project constitutes the following components: • A wireless sensor network with a base station. • An Edge PC, that interfaces with the sensor network to collect the sensor data and sends it out to the cloud server. The system also interfaces with the sensor network to send out command signals to control the switches/ actuators. • A cloud that hosts a database and an API to collect and store information. • A web application hosted in the cloud to provide an interactive platform for users to analyze the data. The project was demonstrated in: • Lecture Hall (https://iac-lecture-hall.engr.iupui.edu/LectureHallFlask/). • Test Bed (https://iac-testbed.engr.iupui.edu/testbedflask/). • A company in Indiana. The above examples used sensors such as current sensors, temperature sensors, carbon dioxide sensors, and pressure sensors to set up the sensor network. The equipment was controlled using compactable switch nodes with the chosen sensor network protocol. The energy consumption details of each piece of equipment were measured over a few days. The data was validated, and the system worked as expected and helped the user to monitor, analyze and control the connected equipment remotely.
  • Thumbnail Image
    Item
    Sustainable Waste Sorter
    (2019-05-02) Garza, Elisabeth; Smerdel, Aaron; Staton, Jordan; Weissbach, Robert; Cooney, Elaine; Freije, Elizabeth
    The purpose of this project is to help people eliminate the confusion on whether they should throw their trash away or dispose of it in a recycling bin. The sustainable waste sorter is an informative device that tells the user where to place their trash. Our customer and the origin of the idea came from an organization called Roche Diagnostics Operations. Roche Diagnostics Operations is a multinational healthcare organization, the Indianapolis location focuses more on creating and developing their diabetic test strips. The device is created of four main components which include a Raspberry Pi 2 Model B, a camera module, an LCD screen, and a casing/mount that holds all of these components together. All of these components are compatible with the Raspberry Pi 2 Model B. The software was programmed in Python and the database in MySQL. During the development of the device, the most challenging task was learning how to develop in the new language, Python. Once the device reached a stable state it was piloted at Roche Diagnostics Operations. The purpose of the first of three pilot sessions was to verify that the device worked in the environment and that the items entered in the database were recognized; as a result, the device passed that test. The second pilot session had the same purpose as the first pilot session but with more items in the database. The device received more interaction during the second pilot session, though the team decided to schedule a third pilot session once all the items were entered into the database and a revamped user interface was completed. The team entered about 800 entries into the database and created a new and interactive user interface for the device. The third pilot session was a success; the items that were scanned by testers were recognized and the new user interface was a success as well. Overall, the sustainable waste sorter project was successful and educational. We, as students, took all of our fundamental learnings from our previous courses and applied them to this project. This allowed us to enhance our problem solving and project management skills. As people use the device, we hope that it educates them on how to properly recycle therefore improving the environmental state of our planet.
  • Thumbnail Image
    Item
    Visualizing Tweet Activity for NIH Covid-19 Preprints
    (2021-10-11) Dolan, Levi
    As a result of the Covid-19 pandemic, the focus on preprint articles as a means of rapid research dissemination has sharply intensified. Produced as part of a larger project examining public comments on preprint articles, this poster breaks down an automated method which was used to gather geographic information via an Altmetric API for the 200k+ Tweets about 1,000 of the first articles in the NIH Preprint Pilot. The workflow in gathering the information for a final visualization is explained, with the results visualized as a global map overlaid with a scatterplot.