Medium Blog Recommendation Automation Using Telegram

##### [Project Github repository](https://github.com/meysamraz/alzheimers_diagnosis_using_brain_MRI)

In this project, I made a model for Zamir diagnosis by using brain MRI data. The data I used for this project contained four classes :

1.  Non Demented
2.  Moderate Demented
3.  Mild Demented
4.  Very Mild Demented

Project overview :

### 1 - Load Images

convert images into batches, split validation data and resize images to 224 (w,h)

### 2- Import and train model

![](/images/loss-02ee9b14.png)

### 4 - Save model

### 5 - Create a GUI

### Run The Project

### Install Libraries

`pip install -r requirements.txt`

### Run

`streamlit run app.py`

## Libraries used in the project

*   [streamlit](https://streamlit.io/)
*   [fastai](https://www.fast.ai/)



Alzheimer's Diagnosis: Brain MRI Image Analysis for Early Detection

layout

By default, the <title> tag for this page is determined by the title field (in the Content group). You can override the tag value here.

metaTitle

media

bottomSections

By default, the description tag for this post is taken from the Excerpt field. You can override the tag value here.

metaDescription

By default, the "og:image" meta tag set for social sharing this post points to the Featured Image field (in the Content group). You can override that image here.

socialImage

markdown_content

If enabled, the title suffix defined in the site configuration is appended to the title tag of this page.

addTitleSuffix

To add or override any meta tag for this page, add entries to this list. Entries defined here take precedence over any other defaults.

metaTags

title

colors

date

author

excerpt

featuredImage

##### [Project Github repository](https://github.com/meysamraz/analyzing-amazon-books-related-to-data-science)

Books can be a very good source for learning data science. In this project, I tried to analyze the data of 1000 books about data science collected from Amazon

## What is the purpose of this analysis ?

*   Get to know the books related to data science that are sold on Amazon
*   Find the best books based on user reviews related to data science and related fields

## Project overview :

![](/images/4.png)![](/images/8.png)![](/images/9.png)![](/images/5.png)

**This project uses the Data science books dataset available on Kaggle https://www.kaggle.com/datasets/die9origephit/amazon-data-science-books.**

### The dataset contains 946 books obtained from scraping Amazon books related to data science, statistics, data analysis, Python, deep learning, and machine learning.

#### There are 18 columns :

*   title: title of the book

*   author: author (or the authors) of the book

*   price: price (in dollars)

*   pages: number of pages

*   avg_reviews: average reviews (out of 5)

*   n_reviews: reviews done for each book

*   star5: percentage of 5 star reviews

*   star4: percentage of 4 star reviews

*   star3: percentage of 3 star reviews

*   star2: percentage of 2 star reviews

*   star1: percentage of 1 star reviews

*   dimensions: size of the book (in inches)

*   weight: weight (in pounds or ounces)

*   language: language of the book

*   publisher: publisher

*   ISBN-13: ISBN_13 code

*   link: link of the Amazon book

*   complete_link: complete link of the Amazon book (including the domain https://amazon.com)

### Loading the data

### Understanding the data

*   shape of the data
*   check column dtypes
*   check is there any null values

### Data preprocessing and feature engineering

*   Split authors
*   Convert weight from pounds and ounces to gr
*   Extract publisher year from publisher
*   Extract inch year from dimensions
*   Remove punct from title

### Exploratory data analysis

*   Authors with more than one books
*   Correlation between features
*   Price distrubtion
*   Whats the min avg max price
*   Number of pages distrubtion
*   Does number of pages of a book affect the price ?
*   What's the most giving avg review ?
*   Price vs avg_reviews
*   In what year have the most books been published ?
*   Price vs publish_year
*   Price vs weight (in gr)

### Clustring Books by Title

*   Apply TFIDF to books title
*   Find the best k for k means using Elbow method
*   Apply kmeans clustring
*   Labling clusters
*   Tag books by titls
*   Best books in each cluster
*   Books price in each cluster

### Scrape reviews from amazon

### Summarizing reviews

### Interactive dashboard

![](/images/demo2.gif)

I used cleaned data and Streamlit in order to create a interactive dashboard . By choosing the title of a desire book , we can see the book details and reviews
In this dashboard, we can summarize long reviews

![](/images/11.png)

### Libraries and Frameworks used in the project

*   [streamlit](https://streamlit.io/)
*   [pandas](https://pandas.pydata.org/)
*   [numpy](https://numpy.org/)
*   [matplotlib](https://matplotlib.org/)
*   [seaborn](https://seaborn.pydata.org/)
*   [termcolor](https://pypi.org/project/termcolor/)
*   [wordcloud](https://pypi.org/project/wordcloud/)
*   [yellowbrick](https://pypi.org/project/yellowbrick/)
*   [spotipy](https://spotipy.readthedocs.io/)
*   [scikit-learn](https://scikit-learn.org/)



Collecting, Analyzing, and Summarizing Amazon Books on Data Science

##### [Project Github repository](https://github.com/meysamraz/weeds_detection_yolo_custom_object_detection)

Weeds can cause a lot of damage to the agricultural industry and water control, and weed detection can be a tedious task. In this project, I tried to detect weeds using the YOLO algorithm.

![](/images/weeds_detection_demo.gif)

# Metrices :

![](/images/PR_curve.png)

![](/images/F1\_curve.png)![](/images/confusion_matrix.png)

# Result :

![](/images/test2-160e21a4.jpg)

![](/images/test.png)

## Libraries used in the project :

*   [matplotlib](https://matplotlib.org/)
*   [opencv](https://opencv.org/)
*   [pytorch](https://pytorch.org/)
*   [numpy](https://numpy.org/)



Weed Detection on Farms Using YOLO Object Detection Algorithm

## Project Description

Have you ever wanted to get real-time data from a website that didn't offer an API? I faced this exact challenge and found a solution using [ScrapyRT](https://scrapyrt.readthedocs.io/en/stable/).

In my case, I wanted to access up-to-the-minute cryptocurrency data from [Coingecko](https://www.coingecko.com/) , a popular cryptocurrency information website. The data I needed included currency titles, icon images, prices, 1-hour price changes, 24-hour price changes, and 7-day price changes. These are crucial metrics for cryptocurrency enthusiasts and traders, as they provide insights into market trends.

To accomplish this, I employed ScrapyRT to crawl Coingecko's website, sending requests and parsing the HTML code to extract the desired data. With this data in hand, I used Flask to create an API, decoding the information scraped by ScrapyRT.

But I didn't stop there. I went a step further and built a simple web application using React.js to test and showcase the real-time data. Now, users can explore the latest cryptocurrency information through an intuitive and interactive interface.

Whether you're a cryptocurrency enthusiast, trader, or simply curious about real-time data extraction, this project offers a solution to a common challenge in the world of web scraping.

![](/images/Screenshot%20(1035).png)

## Problem Statement

Cryptocurrency market data is highly volatile and can change rapidly. Traditional websites often fail to provide real-time data, leaving users with outdated information. This project addresses this critical issue by transforming a static cryptocurrency website into a real-time API. By doing so, it offers users live updates on cryptocurrency prices, market capitalizations, and other essential data points. These real-time insights are invaluable for analysis, decision-making, and training machine learning models in the fast-paced world of cryptocurrency trading and investment.

## Tips for Using Scrapyrt

1.  Adapt to Website Changes:
    Scrapyrt is a powerful tool for web scraping, but it has limitations. One key consideration is that it works by decoding the HTML code you define. If the target website changes its HTML structure, your API may cease to function correctly. Regularly monitor and adapt your scraping code to accommodate website updates.

2.  Respect Website Policies:
    Keep in mind that web scraping exists in a gray area, with policies varying from website to website. While some websites don't explicitly prohibit scraping, others have strict anti-scraping policies and may block your requests. Exercise caution and always review a website's terms of service and policies before scraping its data.

## Setup and Local Development

### To run this project locally, follow these steps:

1.  Clone the repository to your local machine:

`git clone https://github.com/your-username/cryptocurrency-realtime-api.git`

2.  Open project realtimescraper (back-end) folder:

`cd yourdir/cryptorealtimescraper/realtimescraper`

3.  Install packages by running:

`pip install -r requirements.txt`

4.  Run realtimescraper (back side) by runinng:

`scrapyrt`

Than you can see raw result by opening :

<http://localhost:9080/crawl.json?start_requests=true&spider_name=cryptocurrecnydata>

5.  Run Flask application:

`python app.py`

Then you can see decoded json result by opening:

<http://127.0.0.1:5000/rest>

6.  Open reactapp (front-end):

`cd yourdir/cryptorealtimescraper/reactapp`

7.  Install packages by running:

`npm install -g node-modules`

8.  Run reactapp (front side) by runinng:

`npm  start`

Then you can see webapp by opening:

<http://localhost:3000/>

### Libraries used in the project

*   [Scrapy](https://scrapy.org/)
*   [Flask](https://flask.palletsprojects.com/en/2.1.x/installation/)
*   [flask-corse](https://flask-cors.readthedocs.io/en/latest/)
*   [requests](https://pypi.org/project/requests/)



My Projects

Medium Blog Recommendation Automation Using Telegram

Alzheimer's Diagnosis: Brain MRI Image Analysis for Early Detection

Collecting, Analyzing, and Summarizing Amazon Books on Data Science

Weed Detection on Farms Using YOLO Object Detection Algorithm

Convert Cryptocurrency Website into Realtime API