Bosch Shared
- Top Results
See all results
- About mrdreambot
- About mrdreambot
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
About mrdreambot
since
08-16-2023
10-09-2023
mrdreambot
New Poster
5
Posts
0
Likes
0
Solutions
10-09-2023
04:31 AM
Dear BSTRobin, I did ask in this Forum regarding the issues I encountered back in 16-08-2023, topic "BME688 Mobile App Live-test always say it is coffee". A couple of other forum users said they encountered the same issues that I had. I posted more info here but there was no working solution proposed other than a list of actions to try out which I did. Therefore, I had to find my own way to resolve them. I did not see this as a negative thing and had no negative sentiment towards the company or the product. In fact, it provided me with a chance to experiment and learn more about your excellent BME688 sensors. I just wanted to share what I learned with my fellow Forum users here.
... View more
10-07-2023
11:43 PM
Dear fellow Forum users, As you all know, Bosch provides Arduino-based firmware examples and the BME AI-Studio for training AI models to identify different smells using the BME688. You can collect your training data and save it on a SD card by using the sample firmware. The BME AI-Studio is proprietary software which allows you to train your AI model based on the training data and generate a configuration file and a meta information file about the AI model in a JSON-like format which you put on a SD card and insert into the BME688 Evaluation Kit to run your model. Unfortunately, I was not getting the results I expected using these proprietary tools (likely due to my own faults and ignorance). I went on a journey to develop my own firmware by enhancing a BSEC2 (Bosch Senortech Ennvironmental Cluster 2) library example and train my own AI model using Jupyter notebook and scikit-learn. I also built a REST API for the trained AI model and an AI application using it to predict smells, containerised them and pushed the container images to the quay.io repository. You can now just run the container images on Podman, MicroShift or OpenShift. If it does not work for you due to different sensor drift and ageing characteristics between my sensor and yours, then you have to retrain the AI model using my Jupyter notebook. All documentation including the architecture, Jupyter notebook, training data, scripts, deployment instructions and source code can be found in my Github repository. An accompanying video can be found here. A link can also be found in my Github repository's README.md file. I hope the info can help you with your project or at least provide you with an alternative. I welcome your feedback and suggestions. MrDreamBot
... View more
08-21-2023
02:13 AM
Here is an update after spending more time with the Dev Kit: I was able to get further than before but still having some issues. I have summarised what I did, the results and my questions for the experts reading this post. If you have encountered these issues and resolved them, or have an explanation for them, please let me know. What I did Deleted all files on the SD card Generated a default .bmeconfig in AI Studio and copied it to the SD card Power up the Dev Kit and let it run for over a day Copied the example 'Coffee or Not' .aiconfig and .config files to the SD card and applied power (sampling natural air) Connected the mobile app and switched to Live-test and observed if the behaviour remained the same as before (always predicting coffee) The results Selected the 'Air' class. Initially, some sensors were still reporting Coffee but after some time (minutes), they reported 'Air' Put the Dev Kit in a container containing coffee and switched the mobile app 'Live-test' class to coffee. After sometime, the mobile 'Predictions' page was still predicting 'Air' while the 'Raw Data' page showed 'Prediction Likeliness' 100% for coffee. This continued until I disconnected and re-connected the mobile app. Then the Prediction showed 'Coffee' See screen shots attached. Took the Dev Kit out of the coffee container. The Prediction continued to be coffee until minutes later when 3 of the 4 sensors correctly predicted 'Air'. See screen shot attached. My Questions: Do I need to disconnect and reconnect for every environmental change before new environment can be detected? Is it normal for the Live-test Raw Data of the sensors contradict those on the Prediction page? How long does it take normally for the sensors to detect the change in environment (air to coffee and coffee to air)? For me, it took several minutes which, in my opinion, was too long. Was it normal for certain sensors to predict the environment incorrectly? For some of those incorrect predictions, looking at the Raw Data, it seemed that the 'Prediction Likeliness' were quite close for the 2 classes: coffee and air. What can I do to improve the detection accuracy and reduce the delay in detecting environment change? Screen Shots raw Data shows coffee but Prediction page still showing all air Took a long time to correctly prediction air after taking the dev Kit out of the coffee container Thanks in advance.
... View more
08-17-2023
08:01 AM
Thanks for the reminder, BSTRobin. I populated my SD card with the 2021_08_09_10_39_BoardConfiguration.bmeconfig which I download from one of the Community discussions. I inserted the SD card, baterry in the Dev Kit, plug the USB cable in the Adafruit Feaher and let it running for jsu under 24 hours. Does a few hours make such a difference. Should I let it run for another 24 hours? I've attched the bmeconfig here for reference. I uderstand that I could have use the AI Studio to create a bmeconfig. At that time, I did not know that yet. Since then, I have generated another .bmeconfig from AI Studio and repeated the sample collection, model training, deployment and testing cycle with the same outcome Yes, the SD card has been inserted and I've been recording many MByes of samples. And I could import files in AI Studio and train several modes but none of them worked. Attached the screen shot of one model which is supposed to destinguish tea and normal air. My question to you BSTRobin is what can I try? Run the dev kit for another 24 hours, change the .bmeconfig to certain values recommended by you? I welcome any suggestion.
... View more
08-16-2023
07:37 AM
Dear fellow users, I just started training the BME688 gas sensors (Dev Kit) using the AI Studio. The recorded samples (30 minutes) of coffee and (30 minutes) air were imported successfully into the AI Studio. I created a model (around 95% accuracte with few errors in the confusion matrix) using all default values and exported the .config and .aiconfig and saved them in the SD card. But when I ran the live-test from the mobile app, it did not matter which class I chose, it always reported it was coffee whether I put the sensors (Dev Kit) in a container containing coffee or not. I also created another model including all data channels instead of the default gas data channel only, trained the model and saved the exported config files to the SD card. The Live-test still reported coffee whether I put the sensors in the coffee jar or not. I even selected the Example Project from the AI Studio, selected the 'coffee or not' model, exported and saved the config files in the SD card, the result was the same: always coffee even it should have been air! Any suggestions? Note: I always deleted the .config and .aiconfig from the SD card before copying the new config files to it. If this issue has been solved before, please forgive me, I could not find it searching the topics.
... View more
Latest Tags
No tags yet
Still looking for something?
- Top Results
See all results