Timeseries:
1. Every 15 mins collect the views by using Python
2. Store them into any Timeseries DB by using Python
3. Give me a report which Song would get more views
Tools/Libs:
Use Anaconda/Miniconda and Python 3.7+ environment
Use VSCode
FS1210: RunLy
Collect song Lyrics from a running video (assuming the video has subtitle)
There are a lot of outcry on this topic like WSJ should not have diminsh Jill Biden like that. We have to measure topics like this and give a meter for these topics.
We have come up with TactML Score to identify researchers like Yoshua Bengio
Criteria:
Independent Research
Visionary Meter
FS1233: Tech or Not
Classify tech or non-tech video among 10,000 videos.
FS1234: AI Info Maker
AI Info maker
When we play this video, it shoud show
"Harry Shum might have graduated his school in 1980s"
"HS might have gone to USA for AI learning in 1983"
Everything must be based on the video content and we should not search things online.
FS1235: Video2Text
You need to convert this video to text with Amazon, Google, MS AI tools and do a benchmarking which one is more accurate.
You can start with a sample 2 mins audio as a sample work.
FS1236: Collect 200 Research Papers
Collect 200 research papers
Topics
1. ML
2. NLP
3. DL
All these papers should be in PDF format
Sources:
1. Neurips
2. Research gate
3. Google scholar
4. Arxiv
It's hard to view the Google Form without navigation. If you could improve the viewing option by Left, Right Nav it would be great. You will have to read the Excel sheet in Python and show the viewing page
FS1257: Open Source Contribution per capita
We need to find how many open source contributors in South India and Ontario and compare them regularly.
We need to increate the open source contributors count in South India. This is the ultimate goal of this project.
FS1258: California Exodus
Do some visual about this topic.
FS1259:
Make a visual as this image.
FS1260:
Do some ground work on TinkerBell and write an article about it.
Be careful out there: LinkedIn is infested with fake profiles. Out of thousands of invites I've received, hundreds have been fake. How can you recognize a fake profile?
Fakers use patterns and make mistakes that give them out. Here are some examples:
You can't find the same person elsewhere with a simple Google search.
Their work and education history is from large institutions with no unique details.
The name and professional summary are generic.
Participated groups and pages don't show a unique history.
Historical participation on posts doesn't exist or doesn't fit the background or location.
The friend list doesn't fit the background.
With Google image search the picture is elsewhere with another name. This is less useful today because it is so easy to generate deepfake images.
The image below shows a fake profile that has gathered 413 connections, many of them Finns. The name is generic, lacking surname. The photo looks like an image bank model or a deepfake. The study program she claims to participate doesn't exist.
Have you recognized fake profiles among your invites yet? Any other ways you have used to recognize a fake profile?
Assume your client website has some mp3 files in a web page. you need to download them by using simple python code. Automate the process by using BeautifulSoup.
1 x 10/100/1000 Mb/s Gigabit Ethernet (RJ45)
1 x 3.5 mm Headphone/Microphone Combo Jack
ClassicTesti.org
ClassicTesti.com
are available
The 3.0 release of JupyterLab brings many new features! Like a visual debugger
$ 𝚙𝚒𝚙 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 𝚓𝚞𝚙𝚢𝚝𝚎𝚛𝚕𝚊𝚋==3
Features • Stepping into Python code in JupyterLab with the visual debugger • The table of contents extension now ships with JupyterLab. This makes it easy to see and navigate the structure of a document • Support for multiple display languages
And many more check it here
https://lnkd.in/gk-Z4gg
⚡ Spread the Open Source love If you know an amazing project, paper or library drop me a message here on LinkedIn or Twitter @philipvollet
https://lnkd.in/gG3BgzG
