Today we’re ready to add more data train the machine 😀 we have 3 different case and here’s how we’re doing 😀
Adding more data
start the training
testing 🙂
we have finished about 80 – 90% of the work 🙂 the next thing to do is adding more data, do more training and make documentation 🙂 Wish these guys luck so they can be graduated very soon 😉
*Photos and videos are also available here
Day #4 🙂
We finally figured out all errors and managed to make Tensorflow learn our own datasets 😉
Lots of things is going on, so many files to be modified
we’re going to make a proper notes and tutorial for this long and tiring process 😀
We’re going to add more datasets and categories next week 🙂
We have installed Tensorflow and the dependencies to do object recognition and object counting. Today we started to collect our own data and do labelling so we can start do training with our own data set
it requires lots of steps and lots of time 😀
we have to resize all the images and label the image before feeding it to the learning system
We started with very few data set and going to how the result is. And then we will add more data and do classifications.
Great progress for today 🙂 We still have 2 more sessions and we’re going to finish the target by then 🙂
Day #2 of Machine Learning Coding Camp 🙂 We are exploring more difficult projects and start to think about projects to be created tomorrow. The projects should be original and can be done in teams 😉
Kinan is working hard on several projects, one of them is to determine the best way to go to school and also a project to differ two different objects
Sandra is working really hard today
The first project is Chatbot that works so well
The second one is an original project that can tell you One Direction personnel when we type in things related to them 🙂
Sandra did a lot of research and entered lots of data sets 🙂
It worked perfectly
and had been tested by a true Directioners 😀
The data sets ;D
Sarah is also very productive today
The first project is Sorting Hat
Worked perfectly 😉
nice drawing, btw 😉
The second project is Pacman
It also worked nicely!
And the third project is Tamagotchi project 🙂
super cute project and super cute drawings 🙂
Kiran
But before that, she created this Tourist Recommendation Project
The Pacman project
Kinan’s Pacman can avoid the Ghost for 12 moves 😉
The report
Bebe’s first project today is defining books category from the cover of the book
the second project is the Pacman
the third one is the Tourist Recommendation Project
Bebe is teaching Hafiz the Pacman Project because it is very difficult 😉
Kimi’s first project is Rock, Scissor, Paper Project. Worked well!
The second is recognising sounds for places
The report 🙂
Hafiz
Hafiz’s first project today is Tourist Recommendation Project
and then the Pacman
The report 😉
Ariel had just joined us today 🙂
He started with the Titanic project
so he had to install Python
Ariel’s first Python Machine Learning project 🙂 *yes, Jack did not survived 🙂
the next project is to recognize pictures
and the third is to differ nice and bad words 🙂
The report 🙂
Renald joined the class today like Ariel
Ren gone through mostly the same thing as Ariel
The report
Capt. Faris on duty
Capt. Faris is responsible to help Bima and Kresna exploring Tensorflow and Keras 🙂
Bita
The first project is about the best way to get to school 🙂
The report
Cinta
the first project is Chatbot
The next is classifying books
The report
Bima is exploring Keras and Tensorflow today
we’re training the system to read numbers
observing the result
trying different codes and configurations
learning about CNN
Kresna did some exploration in AWS Machine Learning
and then back to Tensorflow to Keras offline
Kresna’s first object detection with Tensorflow 🙂
Lunch break
The class
*Additional session: Fadhlan consulting his robot. We have to debug this obstacle avoiding-mapping robot for a competition in Russia, October 2019 😀
An interesting class, indeed 🙂 These guys know each other and been in the same robotic competitions 😀
Photos and videos are also available here
We finished the #3 and #4 session today 😉
We explored both darknet and darkflow object detection today. This is the example of darkflow object video with camera
We tried very hard to make either darknet or darkflow works on each computer
Some of the results today
The project today is to create our own dataset of 3 kind of instant noodle: Indomie, Supermie and MiSedap. We had to download the picture, put box around it and label it, make the system learn and then see if it can classify or detect the object
At the end of the class we were still figuring out about how to get the new weights to train our dataset. The process took lots of time, we had to continue the rest at home 🙂
*Photos and videos are also available here
Most of the photos are courtesy of Telkom Bandung Digital Valley
Code - and rule the world 