- Видео 29
- Просмотров 244 457
Ahmed Besbes
Франция
Добавлен 27 янв 2013
Hello there,
My name is Ahmed, I'm a data scientist and my goal with this channel is to share with you programming tips, machine learning tutorials, and everything data-related to make your life easier.
I've always found that online content was a precious gift to learning so I decided to make my own channel.
Consider subscribing to keep posted for future content.
-Ahmed ;)
My name is Ahmed, I'm a data scientist and my goal with this channel is to share with you programming tips, machine learning tutorials, and everything data-related to make your life easier.
I've always found that online content was a precious gift to learning so I decided to make my own channel.
Consider subscribing to keep posted for future content.
-Ahmed ;)
How to extract semantic and meaningful keywords with BERT
👉 Subscribe to my channel on this link bit.ly/31ZOVFD
In this video, I'll review the KeyBert package to extract meaningful and semantic keywords from text documents.
Here's what we'll cover:
✅ We'll understand how KeyBert works and how it adds diversity in the results
✅ We'll try it out on some examples
✅ We'll embed it in a Streamlit application
All the experimentation + Streamlit code is available on my repo: github.com/ahmedbesbes/keywords-extractor-with-bert
KeyBert (repo): github.com/MaartenGr/KeyBERT
KeyBert (blog post): towardsdatascience.com/keyword-extraction-with-bert-724efca412ea
👉 Follow me at:
- Medium: bit.ly/3xSR6cK
- My blog: ahmedbesbes.com
- Twitter: bit.ly/3ttihaI
- Github: github....
In this video, I'll review the KeyBert package to extract meaningful and semantic keywords from text documents.
Here's what we'll cover:
✅ We'll understand how KeyBert works and how it adds diversity in the results
✅ We'll try it out on some examples
✅ We'll embed it in a Streamlit application
All the experimentation + Streamlit code is available on my repo: github.com/ahmedbesbes/keywords-extractor-with-bert
KeyBert (repo): github.com/MaartenGr/KeyBERT
KeyBert (blog post): towardsdatascience.com/keyword-extraction-with-bert-724efca412ea
👉 Follow me at:
- Medium: bit.ly/3xSR6cK
- My blog: ahmedbesbes.com
- Twitter: bit.ly/3ttihaI
- Github: github....
Просмотров: 21 722
Видео
How to Build and Deploy a Document Anonymizer with Streamlit and SpaCy
Просмотров 2,6 тыс.3 года назад
#streamlit #spacy #webapp 👉 Subscribe to my channel on this link bit.ly/31ZOVFD Anonymizing data is important to be compliant with GPRD regulations: this is a very common use-case where data science can help. In this video, we'll see how to build and deploy a text anonymizer with Streamlit, SpaCy and a Streamlit component called st-annotated-text. This app will: ✅ process text input from text a...
How to Deploy a Streamlit app on Heroku - Part 4
Просмотров 1,1 тыс.3 года назад
#steramlit #python #webapp 👉 Subscribe to my channel on this link bit.ly/31ZOVFD Streamlit is a great tool to build interactive web apps for your data science projects without HTML, CSS or JavaScript knowledge. In this video, we'll see how to easily deploy a Streamlit app on Heroku for free. ✅ Learn more about Streamlit: streamlit.io/ ✅ Learn more about Heroku: www.heroku.com/ ✅ Want to see a m...
How to organize the layout of your Streamlit app - Part 3
Просмотров 5 тыс.3 года назад
#steramlit #python #webapp 👉 Subscribe to my channel on this link bit.ly/31ZOVFD Streamlit is a great tool to build interactive web apps for your data science projects without HTML, CSS or JavaScript knowledge. In this video, we'll see how to organize the layout of a Streamlit app using a sidebar, columns, containers and expanders. ✅ Learn more about Streamlit: streamlit.io/ ✅ Want to see a mac...
How to Collect user inputs with Streamlit - Part 2
Просмотров 14 тыс.3 года назад
#steramlit #python #webapp 👉 Subscribe to my channel on this link bit.ly/31ZOVFD Streamlit is a great tool to build interactive web apps for your data science projects without HTML, CSS or JavaScript knowledge. In this video, we'll see how to collect user inputs with Streamlit using a variety of widgets such as text inputs, sliders, select boxes, number inputs, etc. ✅ Learn more about Streamlit...
How to Display data with Streamlit - Part 1
Просмотров 6 тыс.3 года назад
#streamlit #python #webapp 👉 Subscribe to my channel at this link bit.ly/31ZOVFD Streamlit is a great tool to build interactive web apps for your data science projects without HTML, CSS, or JavaScript knowledge. In this video, we'll get started using Streamlit to see how we can display data (text, markdown, images, dataframes, etc) on the screen. ✅ Learn more about Streamlit: streamlit.io/ ✅ Wa...
Detecting Pneumonia from Chest X-Ray images with PyTorch
Просмотров 10 тыс.3 года назад
#pneumonia #pytorch #deeplearning # Subscribe to my channel on this link bit.ly/31ZOVFD In this video, I'll show you how you can build a deep learning model to detect Pneumonia directly from chest X-Ray Images. I'll walk you through the end-to-end process of building, evaluating and testing a convolutional neural network to accomplish this classification task. If you have little to no experienc...
People of Data Science #1 - Kaggle Grandmaster Théo Viel
Просмотров 7933 года назад
This is the first video in a series where I interview and discuss with people with experience in data science and machine learning. In this first video, I'll be discussing with Théo Viel, Kaggle Grandmaster. We'll learn more about Théo's journey, his current work as a deep learning researcher as well as some tips on boosting your machine learning models. Théo's Kaggle profile: www.kaggle.com/th...
End to end tutorial to Build and Deploy a Streamlit Application on Heroku
Просмотров 2,4 тыс.3 года назад
#streamlit #python #machinelearning # Subscribe to my channel on this link bit.ly/31ZOVFD If you're a data science practitioner you should definitely check this Streamlit app I made: it's called Playground. Playground is a machine learning application that allows you to tinker with machine learning models from the browser and understand many things about models (overfitting, underfitting, the i...
How to use MLflow on AWS to Better Track your Machine Learning Experiments
Просмотров 12 тыс.3 года назад
# Subscribe to my channel on this link bit.ly/31ZOVFD In this video, I experiment with MLflow to: - track machine learning experiments based on: * metrics * hyper-parameters * source scripts executing the run * code version * notes & comments - compare different runs between each other - set up a tracking server locally and on AWS EC2 and S3 All the code is available on Github: github.com/ahmed...
9 Common Mistakes You Shouldn't Do as a Data Scientist !
Просмотров 1,5 тыс.4 года назад
#datascience #mistakes # Subscribe to my channel on this link bit.ly/31ZOVFD In this video, I will cover 9 common mistakes data scientists make when they start their career. I personally made some of them a few years ago and my hope with this video is that you learn from my experience and stop making them yourself. This is by no means an exhaustive list, it's more of a personal opinion built up...
Scrape Tweets with No Limitation and No API Key with Twint
Просмотров 50 тыс.4 года назад
# Subscribe to my channel on this link bit.ly/31ZOVFD Twint is a great package to scrape tweets. In this video, I'll show you how: - It can fetch almost all Tweets (Twitter API limits to last 3200 Tweets only); - Easy it is to set up - It can be used anonymously and without Twitter sign up - It has no rate limitations Repo: github.com/twintproject/twint Learn more about Twint in my article: med...
Deploy a React app to Netlify in less than 60 seconds
Просмотров 1,4 тыс.4 года назад
#cartoonify #netlify #deployment # Subscribe to my channel on this link bit.ly/31ZOVFD How to deploy the React app we built in the previous video so that it goes live on the internet and anyone can have access to it? In this video, we will look at Netlify: an excellent platform that allows us to easily build and deploy React applications for free in less than 30 seconds. To be able to use it: c...
How to build a React App to interact with a machine learning model
Просмотров 19 тыс.4 года назад
#cartoonify #reactjs #webapp # Subscribe to my channel on this link bit.ly/31ZOVFD How to build a custom interface to interact with the model? In this video, we will look into React.js and how to use it to build a simple and yet user-friendly interface to communicate with our deployed model. By the end of this video, you should be able to understand how the code is structured, how to improve it...
Deploy a Deep Learning model on AWS Lambda using Serverless framework
Просмотров 10 тыс.4 года назад
#cartoonify #aws #serverless # Subscribe to my channel on this link bit.ly/31ZOVFD How to deploy CartoonGAN, the model behind Cartoonify, on a serverless architecture using AWS? In this video, we will: - Look at serverless architectures and understand why they matter and why you should consider them - Use the Serverless framework to define the infrastructure as code - Deploy the model and test ...
Style Transfer with CartoonGAN - turn your pictures into cartoons!
Просмотров 8 тыс.4 года назад
Style Transfer with CartoonGAN - turn your pictures into cartoons!
4 Steps to Build and Scale a Machine Learning App - Part #0
Просмотров 9 тыс.4 года назад
4 Steps to Build and Scale a Machine Learning App - Part #0
Introduction to data scraping in Python using Requests and BeautifulSoup
Просмотров 1,6 тыс.4 года назад
Introduction to data scraping in Python using Requests and BeautifulSoup
Corona Papers: A Search Engine to Mine Covid19 research papers with artificial intelligence
Просмотров 1,5 тыс.4 года назад
Corona Papers: A Search Engine to Mine Covid19 research papers with artificial intelligence
How to build custom computer vision datasets for classification and object detection
Просмотров 1,8 тыс.5 лет назад
How to build custom computer vision datasets for classification and object detection
Introduction to Neural Networks and Deep Learning from Scratch
Просмотров 1,3 тыс.5 лет назад
Introduction to Neural Networks and Deep Learning from Scratch
Building a ConvNet for Knee injury classification from MRI exams and interpret its predictions
Просмотров 1,8 тыс.5 лет назад
Building a ConvNet for Knee injury classification from MRI exams and interpret its predictions
Comment analyser les contributions du Grand Débat National avec de l'intelligence artificielle
Просмотров 2 тыс.5 лет назад
Comment analyser les contributions du Grand Débat National avec de l'intelligence artificielle
Introduction to character level CNN in text classification with PyTorch Implementation
Просмотров 28 тыс.5 лет назад
Introduction to character level CNN in text classification with PyTorch Implementation
Image Captioning with Deep Learning and Attention Mechanism in PyTorch
Просмотров 6 тыс.6 лет назад
Image Captioning with Deep Learning and Attention Mechanism in PyTorch
Real-time object detection using TensorFlow (Yolo Algorithm v2)
Просмотров 1,9 тыс.6 лет назад
Real-time object detection using TensorFlow (Yolo Algorithm v2)
can we achieve this for short clip?
Hello, friend How to Determine Whether Two Different People's Tweet Reactions Persons Match or not
Hey, the colab link is broken but I'd love if you could send me the notebook
i want to confirm that i can safely collect data per month around 3200 tweets only right?
fantastic tutorial Ahmed, thank you!
looked at your other videos and wow, subscribed!
i got lost at training loop. so many unnecessary things there.
i got this error: CRITICAL:root:twint.run:Twint:Feed:noDataExpecting value: line 1 column 1 (char 0) sleeping for 1.0 secs do you know how can i fix it?
Hello,Would you mind if ask if you can demonstrate how to use a crawler to capture twitter space cations. If the audio file is very long (more than ten or twenty hours), how to quickly capture it?
Hey can i know where is code for it
Hii sir..I trained tflite model and apply mlflow..the whole exp is done in google colab...but models are not logging...(mlflow.tensorflow. autolog) is not working...what can I do for resolving this issue...and also autolog is not working..
thanks your great video, Can we output multiple images at once?
Dear Sir, I hope this message finds you well. I am writing to seek guidance and assistance regarding the acquisition of the MRnet dataset for a project focused on MRI diagnosis using Convolutional Neural Networks (CNN). As a recent graduate in computer science, I am highly motivated to embark on this project, and MRnet from Stanford University appears to be the most suitable dataset for its implementation. I have taken the initiative to register for access to the MRnet dataset as per the provided procedures, but it appears that I am encountering difficulties in the download process. Despite my efforts, I have not been able to successfully access the dataset. Having followed your work and profile closely, I am genuinely impressed and inspired by your expertise in the field of data science. As I aspire to develop my skills and knowledge in this domain, I am eager to seek your guidance on obtaining the MRnet dataset, which would be a crucial resource for my project. I sincerely appreciate your consideration of my request and would be immensely grateful for any insights or directions you can provide on how to access and download the MRnet dataset. Your mentorship and expertise would be invaluable as I pursue my goal of becoming a proficient data scientist. Thank you for your time and consideration. I eagerly anticipate the opportunity to learn and grow under your guidance. Warm regards
You know exactly how to make the video. Thanks
Now i got the right and required video...
have you any idea how to Group Keywords and Build Topic Clusters ?
have you done all this code by yourself? i thought that was imposible for any machine learning engineer
How would one go about extracting the sentiment for each of the topics?
Hello Did you happen to find a way to do that ?
@@mouryateja7848 Sorry, no I didn't.
I sir u have made this project in ur youtube channel
Hi Ahmed, thank you for the clear presentation of this code! I am wondering whether it is possible to only scrape tweets by location and for a specific time period, for example tweets about 'covid' in the United States from July, 2021 until August, 2021. Do you have a solution for this? Thanks in advance.
yes, it is possible , see the documentation of the twint
Can I deploy model on any other platform
Very Helpful Ahmed, Thanks for sharing
I am getting the following error [!] No more data! Scraping will stop now. Can you please suggest how to resolve this
donnt what u re doing at 1:47
Can u do one using Text-GAN also pls
Twint is so powerful... it almost feels like an illegal script😂😂
Hello, the deployment was succesfull, however I am getting application error when viewing the web app.
ImportError: cannot import name 'keyBERT' from 'keybert' (/usr/local/lib/python3.7/dist-packages/keybert/__init__.py) hi im getting error like this please help me to solve this
Can you do one about gradio ?
Thanks for the tutorial. Is it possible to gather all the tweets on a certain hashtag instead of setting n=1000 or a certain number? For example, If I want to collect all the historical tweets on a hashtag like ''climate change'', where should I make a change in the script?
Can I get ur email I'd?? Need to know something
Its not working👨💻
when installing twint should i use command prompt
Please please sir, I have task like cartoonify. I cant developed. I need your help please. How I developed. Please Please I am trying from one year. But I cant developed.
Hey..did u develope now?
@@Ritesh-v8o no
Wow. Amazing video. Explained so well. Thank You.
Could you please repeat the code of the program? Code link not working. thanks.
Thank you very much Ahmed. May I ask a question? How to identify the date or time frame that I want to scrape? maybe 5-10 years period back.
Anybody else have challenges installing keybert via Jupyter Notebook? It just seems to hang indefinitely for me...
Hi i would like to rebuild this project from zero to learn more about data science, i can run and this program on my Mac book or i should work with a virtual machine from AWS? Thanks
Hi Ahmed, I get an error every time I run the pip3 install . -r requirements.txt, please help
The google colab link is not working
Twint required python version 3.6
Sorry can you put the collab link? it doesn't work well
Can you also use this to extract keywords from a dutch text?
how can i use location and keywords togather ????
I am newbie, i would high appreciate if you help me . i got this type-error . i also create a folder named network and put 1. __init__.py 2. Transformer.py file in it but still dose not working TypeError Traceback (most recent call last) <ipython-input-21-5d7e76d58e01> in <module>() ----> 1 get_ipython().run_cell_magic('time', '', 'output450 = transform(models, "Hosoda", path, load_size=450)') 3 frames <decorator-gen-53> in time(self, line, cell, local_ns) <timed exec> in <module>() <ipython-input-16-cdf1a86207a2> in transform(models, style, input, load_size, gpu) 12 13 def transform(models, style, input, load_size=450, gpu=-1): ---> 14 model - models[style] 15 16 if (gpu > -1): TypeError: unsupported operand type(s) for -: 'Transformer' and 'Transformer'
Another useful video! One question:why don't you use AWS CDK to provide infustructure?
Very useful video!
Nice video! Does anyone know if there is a version of this model for different languages?
Nice video! Sir I'm getting unpickling error '\xe2'. How to rectify it.
Logging artifacts from a jupyter notebook didn’t work , someone has any suggestions? in other hand running from a simple python script ran successfully.
PD: for someone having this scenario just load s3 buckets keys into notebook using: pip install python-dotenv from a .env file