Identifies the faulty wafer before it can be used for the fabrication of integrated circuits and, in photovoltaics, to manufacture solar cells.

Overview

Retrainable-Faulty-Wafer-Detector

Aim of the project:

In electronics, a wafer (also called a slice or substrate) is a thin slice of semiconductor, such as crystalline silicon (c-Si), used for the fabrication of integrated circuits and, in photovoltaics, to manufacture solar cells. The wafer serves as the substrate for microelectronic devices built in and upon the wafer. The project aims to successfully identify the state of the provided wafer by classifying it between one of the two-class +1 (good, can be used as a substrate) or -1 (bad, the substrate need to be replaced) and then train the model on this data so that it can continuously update itself with the environment and become more generalized with time. In this regard, a training and prediction dataset is provided to build a machine learning classification model, which can predict the wafer quality.

Data Description:

The columns of provided data can be classified into 3 parts: wafer name, sensor values and label. The wafer name contains the batch number of the wafer, whereas the sensor values obtained from the measurement carried out on the wafer. The label column contains two unique values +1 and -1 that identifies if the wafer is good or need to be replaced. Additionally, we also require a schema file, which contains all the relevant information about the training files such as file names, length of date value in the file name, length of time value in the file name, number of columns, name of the columns, and their datatype.

Directory creation:

All the necessary folders were created to effectively separate the files so that the end-user can get easy access to them.

Data Validation:

In this step, we matched our dataset with the provided schema file to match the file names, the number of columns it should contain, their names as well as their datatype. If the files matched with the schema values then they are considered good files on which we can train or predict our model, however, if it didn't match then they are moved to the bad folder. Moreover, we also identify the columns with null values. If all the data in a column is missing then the file is also moved to the bad folder. On the contrary, if only a fraction of data in a column is missing then we initially fill it with NaN and considered it as good data.

Data Insertion in Database:

First, open a connection to the database if it exists otherwise create a new one. A table with the name train_good_raw_dt or pred_good_raw_dt is created in the database, based on the training or prediction process, for inserting the good data files obtained from the data validation step. If the table is already present then new files are inserted in that table as we want training to be done on new as well as old training files. In the end, the data in a stored database is exported as a CSV file, to be used for the model training.

Data Pre-processing and Model Training:

In the training section, first, the data is checked for the NaN values in the columns. If present, impute the NaN values using the KNN imputer. The columns with zero standard deviation were also identified and removed as they don't give any information during model training. A prediction schema was created based on the remained dataset columns. Afterwards, the KMeans algorithm is used to create clusters in the pre-processed data. The optimum number of clusters is selected by plotting the elbow plot, and for the dynamic selection of the number of clusters, we are using the "KneeLocator" function. The idea behind clustering is to implement different algorithms to train data in different clusters. The Kmeans model is trained over pre-processed data and the model is saved for further use in prediction. After clusters are created, we find the best model for each cluster. We are using four algorithms, Random Forest, K-Neighbours, Logistic Regression and XGBoost. For each cluster, both the algorithms are passed with the best parameters derived from GridSearch. We calculate the AUC scores for all the models and select the one with the best score. Similarly, the best model is selected for each cluster. For every cluster, the models are saved so that they can be used in future predictions. In the end, the confusion matrix of the model associated with every cluster is also saved to give the a glance over the performance of the models.

Prediction:

In data prediction, first, the essential directories are created. The data validation, data insertion and data processing steps are similar to the training section. The KMeans model created during training is loaded, and clusters for the pre-processed prediction data is predicted. Based on the cluster number, the respective model is loaded and is used to predict the data for that cluster. Once the prediction is made for all the clusters, the predictions along with the Wafer names are saved in a CSV file at a given location.

Retraining:

After the prediction, the prediction data is merged with the previous training dataset and then the models were retrained on this data using the hyperparameter values obtained from the GridSearch. The cycle repeats with every prediction it does and learns from the newly acquired data, making it more robust.

Deployment:

We will be deploying the model to Heroku Cloud.

Owner
Arun Singh Babal
Engineer | Data Science Enthusiasts | Machine Learning | Deep Learning | Advanced Computer Vision.
Arun Singh Babal
Create VSCode Extensions with python

About Create vscode extensions with python. Installation Stable version: pip install vscode-ext Why use this? Why should you use this for building VSc

Swas.py 134 Jan 07, 2023
Randomly distribute members by groups making sure that every sector is represented

Generate Groups Randomly distribute members by groups making sure that every sector is represented The Scenario Imagine that you have a large group of

Jorge Gomes 1 Oct 22, 2021
🐍 A Python lib for (de)serializing Python objects to/from JSON

Turn Python objects into dicts or (json)strings and back No changes required to your objects Easily customizable and extendable Works with dataclasses

Ramon Hagenaars 253 Dec 14, 2022
A data engineering project with Kafka, Spark Streaming, dbt, Docker, Airflow, Terraform, GCP and much more!

Streamify A data pipeline with Kafka, Spark Streaming, dbt, Docker, Airflow, Terraform, GCP and much more! Description Objective The project will stre

Ankur Chavda 206 Dec 30, 2022
Modern API wrapper for Genshin Impact built on asyncio and pydantic.

genshin.py Modern API wrapper for Genshin Impact built on asyncio and pydantic.

sadru 212 Jan 06, 2023
Write Streamlit apps using Notion! (Prototype)

Streamlit + Notion test app Write Streamlit apps using Notion! ☠️ IMPORTANT: This is just a little prototype I made to play with some ideas. Not meant

Thiago Teixeira 22 Sep 08, 2022
DG - A(n) (unusual) programming language

DG - A(n) (unusual) programming language General structure There are no infix-operators (i.e. 1 + 1) Each operator takes 2 parameters When there are m

1 Mar 05, 2022
Python samples for Google Cloud Platform products.

Google Cloud Platform Python Samples Python samples for Google Cloud Platform products. Setup Install pip and virtualenv if you do not already have th

Google Cloud Platform 6k Jan 03, 2023
Data and analysis relating to the 5.8M Melbourne quake of 2021

quake2021 Data and analysis relating to the 5.8M Melbourne quake of 2021 Monash University Woodside Living Lab Building The building is located here T

Colin Caprani 6 May 16, 2022
Stop ask your soraka to ult you, just ult yourself

Lollo's auto-ultimate script Are you tired of your low elo friend who can't ult you with soraka when you ask for it? Use Useless Support and just ult

9 Oct 20, 2022
Design-by-contract in Python3 with informative violation messages and inheritance

icontract icontract provides design-by-contract to Python3 with informative violation messages and inheritance. It also gives a base for a flourishing

275 Jan 02, 2023
This Open-Source project is great for sensor capture and storage solutions.

Phase 1 This project helps developers in the creation of extended realities that communicate with Arduino and require the security of blockchain stora

Wolfberry, LLC 10 Dec 28, 2022
These are the scripts used for the project of ‘Assembly of a pan-genome for global cattle reveals missing sequence and novel structural variation, providing new insights into their diversity and evolution history’

script-SV-genotyping These are the scripts used for the project of ‘Assembly of a pan-genome for global cattle reveals missing sequence and novel stru

2 Aug 26, 2022
Comics/doujinshi reader application. Web-based, will work on desktop and tablet devices with swipe interface.

Yomiko Comics/doujinshi reader application. Web-based, will work on desktop and tablet devices with swipe interface. Scans one or more directories of

Kyubi Systems 26 Aug 10, 2022
Simple Crud Python vs MySQL

Simple Crud Python vs MySQL The idea came when I was studying MySQ... A desire to create a python program that can give access to a "localhost" databa

Lucas 1 Jan 21, 2022
Driving lessons made simpler. Custom scheduling API built with Python.

NOTE This is a mirror of a GitLab repository. Dryvo Dryvo is a unique solution for the driving lessons industry. Our aim is to save the teacher’s time

Adam Goldschmidt 595 Dec 05, 2022
Convert three types of color in your clipboard and paste it to the color property (gamma correct)

ColorPaster [Blender Addon] Convert three types of color in your clipboard and paste it to the color property (gamma correct) How to Use Hover your mo

13 Oct 31, 2022
Load, explore and analyse data from Scotland and rest of the world related to Covid19.

Streamlit Examples This is my first attempt with Streamlit. It is an open-source framework, free, Python-based and easy to use tool to build and deplo

Eyad Elyan 12 Mar 01, 2021
E5 自动续期

请选择跳转 新版本系统 (2021-2-9采用): 以后更新都在AutoApi,采用v0.0版本号覆盖式更新 AutoApi : 最新版 保留1到2个稳定的简易版,防止萌新大范围报错 AutoApi'X' : 稳定版1 ( 即本版AutpApiP ) AutoApiP ( 即v5.0,稳定版 ) —

95 Feb 15, 2021
Flask html response minifier

Flask-HTMLmin Minify flask text/html mime type responses. Just add MINIFY_HTML = True to your deployment config to minify HTML and text responses of y

Hamid Feizabadi 85 Dec 07, 2022