Our Prize Winning AWS Hackathon Entry: Difference between revisions

From gpu
Jump to navigation Jump to search
Created page with "<br>This text is written to explain my project that I submitted to the AWS Hackathon on Devpost. Me and my pal had initial started the challenge as a studying step to construct and deploy a machine learning venture. We ultimately ended up profitable the runner up for [https://canadasimple.com/index.php/User:CrystalOldaker BloodVitals tracker] Computer Vision for our efforts. With speedy urbanization and industrialization Air pollution has develop into an alarming challe..."
 
(No difference)

Latest revision as of 03:48, 6 October 2025


This text is written to explain my project that I submitted to the AWS Hackathon on Devpost. Me and my pal had initial started the challenge as a studying step to construct and deploy a machine learning venture. We ultimately ended up profitable the runner up for BloodVitals tracker Computer Vision for our efforts. With speedy urbanization and industrialization Air pollution has develop into an alarming challenge globally. Among different air pollutants particles , matter in (PM2.5) dimension range are in a position traverse deeply into our respiratory tract , transmitting hazardous chemicals into the human lungs and blood inflicting respiratory and cardiovascular well being points . We purpose to develop picture-based mostly air quality evaluation application , particularly a one that estimates the concentration of particulate matter with diameters in range of 2.5 micrometers. It could possibly let you know what your respiratory from photos generated utilizing client grade hardware. Convolutional Neural Network (CNN) with the high computational power of Amazon EC2 DL1 instance to unravel a pc imaginative and prescient downside of classifying pure images into three completely different categories ;healthy , reasonable and hazard based on their PM2.5 concentrations.



We created our personal dataset by combining information from 2 sources . We learned that, some reasons for overfitting could be high imbalance in information , we had 2232 healthy pictures where because the moderate and danger categories had solely 1480 and 591 photos respectively . 2232. Then we trained our model following the same method as before however this time we used ImageNet weights un-freezed the last a hundred layers for training , then advantageous-tuned the model with a lower learning rate ,we also used early stopping callback with a endurance of 3 . Now, there are some extra things now we have accomplished to improve our mannequin. We migrated our mannequin onto the dl1 instance and initially trained using 1 gaudi processor. This significantly improved our efficiency and effectivity .We were able to quickly take a look at , BloodVitals test experiment and tweak changes . 76 %. We then converted our mannequin right into a tensorflow lite model and used a huggingface demo software called gradio to shortly reveal our utility.



Initially as we tried coaching our model we noticed a number of the picture information in the dataset have been corrupted .We automated the means of scanning by means of and deleting these files after which proceeded to training mannequin . We used EfficientNet structure since its renown for its performance in low spec machines, BloodVitals test this is named switch studying . 30 layers and set the weights parameter to false. 60 % but our training accuracy was high, the model may very well be overfitting. As we have seen we had been in a position an accuracy of about 76 % which is fairly good contemplating the data we had. Now , why do we need this application and how is it totally different from the existing air monitoring system programs, we already know Exposure to fine particles could cause long run health results comparable to lung and BloodVitals health heart downside but additionally they cause brief time period health results similar to eye, BloodVitals test nose, Blood Vitals throat irritation.



The prevailing air high quality monitoring methods are extremely dependent on monitoring stations, which are again situated far away due to the high setup price and expensive tools. For a selected location, the air quality will depend on how air strikes by way of the world and how people are influencing the air , so we cannot depend on something that was predicted a while again or in some place that is not our site of interest .This software helps you analyse the air quality around you with a single image and extra importantly is portable. We are able to carry it anyplace we wish to and BloodVitals test know what your being uncovered to . Honestly I don’t know, I believe there's potential for BloodVitals monitor this app to combine into our routine. People would possibly need features to help plan their day out, BloodVitals test integrate with their phone’s calender and suggest the very best time to plan an outside activity. Right now the Artificial intelligence that runs the machine runs on the azure cloud. Sooner or later we would like to have the ability to deliver it into the app and run it natively.



Certain constituents within the blood affect the absorption of light at various wavelengths by the blood. Oxyhemoglobin absorbs mild extra strongly within the infrared region than in the crimson region, BloodVitals home monitor whereas hemoglobin exhibits the reverse conduct. Therefore, extremely oxygenated blood with a high focus of oxyhemoglobin and a low concentration of hemoglobin will are inclined to have a high ratio of optical transmissivity within the pink region to optical transmissivity within the infrared region. These alternating parts are amplified after which segregated by sampling gadgets operating in synchronism with the purple/infrared switching, so as to offer separate indicators on separate channels representing the crimson and infrared gentle transmission of the physique structure. After low-pass filtering to take away sign parts at or above the switching frequency, BloodVitals test every of the separate signals represents a plot of optical transmissivity of the body construction at a particular wavelength versus time. AC element precipitated solely by optical absorption by the blood and varying on the pulse frequency or coronary heart rate of the organism.