AI Code 101 offers cutting-edge educational tools that allow younger students to learn artificial intelligence (AI) and machine learning (ML) technologies using the block-based coding language Scratch for elementary students and Python for secondary school students, with robot interaction.
We offer project-based learning courses that adopt both computational thinking and design thinking frameworks.
When trying to flash my raspberry pi 4 sd card in my macbook pro, I keep getting "Source and destination checksums do not match". Would anyone happen to know a fix to this?
I did all that and get the ip address ,but when entering password this show Permission denied (publickey,password). note I enter the right password but permission denied
Hello. It may be an issue with the file '/etc/ssh/sshd_config' in your Raspberry Pi. Try changing it to say 'PasswordAuthentication yes'. Hope this helps.
I've seen some nice libraries (EloquentTinyML, Node-Red tensorflow js, AI Code 101) that seem to make it easier to use TensorFlowLite for edge / Microcontrollers. Do all the tools ( Neurona, TinyML, MicroML, etc) use supervised learning? i.e. you feed it data with labels/answers (correct input and output pairs) , so that the rules (function) can be generated and applied on non-training data. The output of a Supervised learning can be in the form of Non-Linear Multi-Regression or a classification Model. None of the current microcontroller ML approaches, use unsupervised learning? If so, how does ML compare to plain old Open-CV in the case of image classification? Also, if a microcontroller is connected to power, and has internet connection... how exactly does offloading to the cloud work? How do you setup a (AWS, linode , digitalocean) server to take your real-time microcontroller data and have it infer a prediction that is sent back to the esp32, pi (zero , pico), etc? And what are the different options to port from Tensorflow/Keras/etc output to the actual c or python code that runs on the controller? everywhereml, Edge Impulse, etc? I understand Arm has "Vela" that's used to compile the Tensorflow's tflite file, to something optimized for the microNPU (helium enabled Cortex-M55/Ethos-N or U65)?
Sorry for our late reply. When you connect to your Pi, the default username is “pi” and the password is “raspberry”. You will not see the password letters as you type, this is normal, so be careful you type it correctly. Hope this helps. Thanks
Make sure you are connected to the raspberry pi first. First use the ssh command "ssh pi@192.168.x.x" where the 192.168.x.x is the IP address that you retrieved from the "ping" command. If you did this correctly, there should be a prompt, type ’yes’ and press Enter. When asked for a password, type ’raspberry’. When you get into the device, now you can type the "vncserver" command. If you type "vncserver" into the mac terminal before properly connecting, bash will not understand the command and will give the "command not found" message. At time 11:01 in the video you can see what will happen in the terminal when successfully connected to the raspberry pi. Hope this helps.
@@aicode1019 hey thanks for taking the time to answer it. I have resolved it. In my case I was actually logged into my pi. vncserver indeed was missing. Had to install it to make it work. For some reason I didn't find it useful at the moment. Using pi as a server right now.
Hi Ben, so sorry that we missed your message. Our email associated to this account was crashed a couple of months ago. Have you solved your issues? All the best/AICODE101
