Weekly reports
Week 5
We were asked to create a team consisting of three members, so I decided to team up with Arva and Sam. We brainstormed some project ideas, initially we decided to make a google assistant related AI, some other ideas are to make AI to solve sudoku.

Week 6
This week, we started to do researching our project, which leads us to a lot of options, and at last, we changed our topic to text analysis.

Week 7
This week we started to implement some of the examples of sources that includes us implementing libraries and frameworks found in github, while we continue discussing more options of our project.

Week 8
This week, we talked about the things we need to install set up first and we also try to find how the algorithm flow works on this project before starting to tinker with the code from the reference.

Week 9
We started to discuss and decided to split up the work as it could be a lot more efficient. While discussing we also decided to use the library of words for embedding.

Week 10

We tried openCV for computer vision. My team worked with Ivan’s team for the computer vision task. For our final project, we found a suitable datasets. It was on Kaggle with the name of all the news. It contains 143,000 articles from 15 news publisher. The datasets also has the label that we want, therefore we decided to use the dataset.

Week 11

We started to work on our final project. One library is not enough, so we experimented with a couple libraries that could be used for Natural Language Processing. TensorFlow has their own library for NLP, and there is a thing called TensorHub by TensorFlow. It is an open source library from google, and is mostly used as the embedding layer for NLP.

Week 12
This week, we figured out that we could use cvlib that contains an object detection function which also we could refer to our own python code which takes the function and inputs its own parameter so that it could work perfectly with the function. We also started to use darknet which is a framework to train our models. We had a lot of difficulties during training our models due to lack of sources such as requires NVIDIA CUDA that are pretty hard to configure and different operating systems which leads us to an option that is working on a single operating system.

Week 13
This week, we presented our project including its demo including the training process, the code, how it works, and when tested with the test set to Mrs. Nunung.

And also, we presented about our difficulties and future works to Mr. Andreas