Anyone else's programme getting stuck at " Use `tf.cast` instead "?

can this code be adapted to not only identify a coin but to recognise the date and images from a dual camera setup? I want to identify collectible coins as they pass by the cameras then either accept or reject them.....

So cool, a bit beyond me (at the moment!) but it gives me a good idea for coin collecting

Try getting the fit clone, but it asks for a username and password. Try my GitHub credentials but no luck. Any idea the problem?

HI Speed HQ cam in glases, wireless transporting to a box on the ancle sending to a computer in a van parked outside. Can you implement Basic Strategy in your system, to tell you to Hit, Split, Stand og DD ?

what raspberry pi os are you using?

my computer automaticlly when downloading the .xml files convert them to "Microsot Edge Html Documents" even though it still is a .xml file. Can this ause in the conversion from .xml files to .csv?

when I Install the Object Detection API it shows in my files the file "cuda-repo-ubuntu1804-11-0-local_11.0.2-450.51.05-1_amd64.deb". why didn't you get it?

when training the AI it shows an error that "TFA has entered a minimal maintenance and release mode until a planned end of life in May 2024". How do I solve this?

If i were to use my own custome .tflite model, would I still need the labelmap.txt file? How would I get one from my tflite file?

Goodjob

tysm for the tutorial! For those who want to save the state of training to Drive (so it's easier to continue with all the disconnects), I created a small script: EDIT: I'm writing this at 4am, so PLEASE check everything before using it and be careful with your drive 1. create a folder in your drive root directory called "obj_detection" 2. Create a new code block directly below the... ("If you want to stop training early, just click Stop a couple times or right-click on the code block and select "Interrupt Execution". Otherwise, training will stop by itself once it reaches the specified number of training steps."), ...and paste this (it saves the whole content folder just in case): ### from google.colab import drive drive.flush_and_unmount() !zip -r /backup_obj_detection.zip /content/ drive.mount('/content/gdrive') !cp /todownloadfile.zip /content/gdrive/MyDrive/obj_detection/ ### It will save the whole content folder to drive, so you need quite a bit of free space. If you want to load the progress back, follow the first steps and run this code instead of running the 3rd step when you get there: ### from google.colab import drive drive.mount('/content/gdrive') !cp /content/gdrive/MyDrive/obj_detection/backup_obj_detection.zip / !unzip -q /backup_obj_detection.zip -d / ### After that follow all the steps that don't create or modify files (as those are loaded already) and run the train script.

I am in need of deployment of this model on android

Hey im running the bullseye os on a raspberry pi 4 B. I can't seem to get across the problem regarding running the .sh script

Hi, could you please make a video how to recognize sounds with a RP? Thanks in advance!

my notebook is crashing at training session. how much ram should I be having for this? \

Can someone help me? I have a problem with the following command step "sudo pip3 install virtualenv", when I execute this command the following error "externally-managed-environment" appears, I performed all the previous steps but I was unable to resolve it. I use Raspberry Pi 4 64 bits

Many thanks for the good video and the excellent Google Colab script. But when I want to use the T100 CPU (Batch 64) in Colab, problems occur. It then takes a very long time before the training starts and the output in the tensor board is relatively senseless. (Only flat graphs). Maybe you would like to check the code, because a lot has changed in the packages. (Unfortunately, I can't, because my detailed knowledge is not yet good enough). Thank you very much for your efforts. Best regards, Michael

@EdjeElectronics I'm trying to develop the identification of objects through their videos, but in the Raspberry Pi 4 interface, the Enable option does not appear in the interface tab, nor does the camera option appear, do I have to install something? Raspberry is already updated

Helpful video. can i peform segmentation instead of object detection in tensorflow lite?

Superb..It is hard to install tensorflow api because the version incompatibilities..you handle those superbly

I would pay for this. Can you make one for me?

This is transfer learning right?

I have a question. What were the specs of your pc? Do you need a high powered pc to make the neural network or can I make it on my raspberry pi 4? Also if I made it on my pc, can I make my pc give a physical output if it sees a “correct” card like turn on a led or move a servo? The raspberry pi has pins so I feel it would be easier on it.

the cat want to escape 😼

the TensorFlow model is not working for Android application. if any suggestion please welcome

issue: when I train the model for grocery shops like (bourbon, wafers, maaza, frooti, maggy) it detects only bourbon and sometimes other products but it detects every brown, black, and gray item as bourbon so is there any suggestions please advice thank you

im having trouble when training my model, for some reason it takes wayy longer then 2-6 hours to train it to completion (around 22 hours). I have re tried from the beginning a few times to see if thats what was wrong but it didnt change anything?

Thank you for the video. But does anyone face the same problem as me? On 7.1, I cannot test the model. It shows "ValueError: Sample larger than population or is negative."

you look like jessy pinkman from "breaking bad" webseries

no more free gpu option wah wahhhhh..... :(

Great video

does this still work?

i think can use a RTX2050 for this job it haves TENSOR CORES inside by CUDA CORES and WAY MORE CHEAP(gives 2x more performance than same priced tpu) AND WAY MORE EFECTIVE WAY(when nvidia drivers available to linux arm there is hardest part)

i have a question:what is performance in pi5 i think this is fourteen percent more performant than pi 4 even 15 more fps in 1080p probably 720p recomended without an tpu

i have a problem that lablel map.text cannot be converted to tf files ? can someone help me

labelmap.pbtxt is not created

I need a bit of help label_map_pbtxt_fname = '/content/labelmap.pbtxt' is not making a labelmap.pbtxt file

i am getting insufficient disk error(^c) i want to train 1000 images which is of 7 gb size having 8 classes on efficient det model please tell me the possible solutions

i cant understand how i can add i description to the images, i want to do this with 10k roman ancient coins but i need to add precise infor about every coin, this is exactly what i need but im a bit confuse about how to create the dataset and assigning info to each photo. i would really appreciate a video or explanation about it

Sigh need to find this updated for raspberry pi 5 can't past the cv2 error.

Can we do this for helmet and number plate detection?

Hi, can I use Raspberry Pi for vehicle Detection and tracking?

Can i run your ipynb in lightning ai?

do you need to have all the objects in all the pictures or you could have 4 different folder for all diffrent type of objects?

Have you looked into updating this for the current version of Raspberry Pi os for the Raspberry Pi 5? I would LOVE to get that working...I have not been able to get it to work....

Do you need a powerful PC for this? Because I tried the Teachable Machine thing and can never get past the "preparing training data" part. It just stays like that for two hours, then freezes and crashes.

Hello, Thank you for your nice video. How can I print the text what is detected. If bottle is detected. I print bottle when it is over 70%. Can you help me?

Everything worked perfect except bounding box is not directly on the objects.placement of rectangle box is away from the object detected.can any1 know how to solve it?

great idea that ive only just thought of years later, taking customers?