DEV Community: Vidya

The Composer's Quill: Weaving Melodies using text prompts with MusicGen

Vidya — Fri, 16 Jun 2023 17:54:11 +0000

Step by Step Guide to generate music in Google Colab

1. Go to https://github.com/camenduru/MusicGen-colab

Do star his repo !!!

2. Scroll down to reach the colab notebooks

3. Choose any colab notebook and click on "Open in Colab".

(You will need gmail ID to use colab.)

4. Loading ... This may take awhile.

5. The magical colab notebook with just 5 lines !!! Click on the arrow.

6. You will get a warning message - Click "Run anyway".

7. The commands in the cell will get executed - the git repo will be cloned and the required libraries will get installed This step could take a while, even with the notebook having a setting of GPU by default.

8. Ignore the error in red and click on the gradio live link.

9. The Music Generation playground !!!

No amount of exclamations can tell how exciting it is to be able to create music with text especially for a plebian !

10. Add your music prompt. This was my prompt.

Create a binaural beat composition that promotes concentration and focus for studying. The beats should be calming and soothing, without any loud or jarring noises. Aim for a frequency range that encourages an alert mental state without inducing sleepiness.

Add other details are per your liking

duration of the composition
model - melody (works with both text and sample music prompt)
medium, small and large models (these work only with text prompts)
an increase in temperature gives more randomness in the output

11. Click on generate to process your text prompt to generate music.

12. The model gives the time it will take to create the music along with the number of steps and %age of completion.

13. Some of the other options to check out -

Melody Condition (optional) - To remix mostly
Music Sample - Works only with the melody model, To create music similar to a piece of music you like
Seed gives the seed of the created music and if you want to create music similar to one you have already created before, the seed number can be added

The author of the repo says -

These parameters, such as top-k, top-p, temperature, and classifier-free guidance, provide different ways to influence the output of a music generation model and strike a balance between creativity, diversity, coherence, and control. The specific values for these parameters can be tuned based on the desired outcome and user preferences.

14. And finally - The generated music result ! You can enjoy listening to your creation in the browser as well as download your creation to your drive as a mp4 file.

15. Trying the melody option to get a different version of the beats I like.

16. The result is not bad, although the original is definitely better. Lots of steps and processing when asked 40 second composition. The option with change in temperature came out pretty good.

"So finally I can compose music", without knowing how to write musical notes !!

"What a time to be alive !!!"

References :

Tutorial by the author of the colab notebook - https://www.youtube.com/watch?v=EGfxuTy9Eeo
One little coder - https://www.youtube.com/@1littlecoder
In the words of Dr Károly Zsolnai-Fehér of the youtube channel "Two Minute Papers" - what a time to be alive !!!!
The surreal images have been created using Bing Image Creator - https://www.bing.com/create

Business at the Speed of AI

Vidya — Sun, 05 Feb 2023 13:54:39 +0000

Today, Artificial Intelligence(AI) plays a crucial role in shaping our online experiences, often driving them.

When we publish posts on LinkedIn, share blog articles, or tweet, AI algorithms review them along with the hashtags and may reshare, retweet, or add them to relevant newsfeeds.
Financial trading bots may act on news published by reputable financial outlets regarding a company's results.
Perhaps the financial article may have been written, assisted by a language model like ChatGPT, just like business proposals, formal and informal emails.
AI algorithms generate product recommendations, AI-powered recruitment bots screen job applications, and advertising bots target our social media usage.
Chatbots offer AI-based customer support, Dall-E creates product images, and AI is winning art competitions.

The reality of an AI-based neuralink is already here, and it is conceivably not incorrect to say; business is conducted at the speed of AI, not speed of human interactions.

Key points of Consistent Hashing

Vidya — Tue, 10 May 2022 03:52:13 +0000

Key points of Consistent Hashing

The key to understand Consistent Hashing is to understand the following:

We not only hash the key IDs but also the server/node IDs. Modulo hashing works well for a fixed number of servers. However in reality servers can go down and become unavailable; so a fixed number of servers cannot be assumed.
Server IDs are hashed to a fixed range. The key IDs are also hashed to the same fixed range. Can imagine this range to be like a circle.
The key IDs are then routed to the closest server in the clockwise direction (It can be anticlockwise as well, as long as the same direction is used for all the keys).
Finally, only a single hashing algorithm need not be used. We can use multiple hashing algorithms to make the servers spread out across the circle. Thus the load can be distributed in a better manner.

Ref:

Book Review - Learning React - 2nd Edition by Alex Banks and Eve Porcello

Vidya — Thu, 23 Dec 2021 10:14:52 +0000

Book Review - Learning React - 2nd Edition by Alex Banks and Eve Porcello

I enjoyed learning from this book but overall I got the feeling that the authors lost steam towards the end and were trying to complete the book in a hurry before the deadline.

The 2nd chapter in this book that deals Javascript basics one needs to know for learning React is extremely good.

Pros

Based on latest React framework
Very well explained Javascript basics chapter
Covers all major parts of learning React
Hooks and State management explained very well
Suspense which is still experimental has been addressed
Code examples are very good

Cons

Would have preferred incremental single project approach. To some extent this has been addressed but not completely
Many files are missing in the github repo - Socket.io example, GraphQL files, Chapter 10 files, Parallel requests example etc.
Code examples should always mention which file they are referring to in github

Overall

A very good book to learn fundamentals of React, State management and Hooks. It can be a supplement to other streams of learning not a single source of learning.

#100DaysOfCode Challenge

Vidya — Tue, 11 May 2021 16:39:59 +0000

100 Days Of Code Github Wall

My experience with the #100DaysOfCode Challenge
Motivation

I decided to try the #100DaysOfCode Challenge for a couple of reasons.
First, to see how I would manage this and secondly to complete items that were languishing due to low motivation.

I did not tweet daily as is generally recommended. For me the public motivation / goal does not appeal.
I also decided to finish my incomplete items first. And only then start anything new.

Code Complete

Started and Completed - 'Data Cleaning Cookbook' by Michael Walker
Completed - 'Automate the boring stuff' by AI Sweigart
Python course in Udemy
100+ Python Exercises

Other

Strictly does not fall into the coding challenge but I thought it relevant all the same.

Completed Google Cloud Sprint course in Coursera
LinkedIn Assessments - 4

Lessons Learnt

Pros

Definitely works. Nothing like not wanting to miss the streak in the github profile as motivation !
Was able to complete every item that had just been pending for no reason other than low motivation
A very good experience, gives a confidence boost along with a sense of accomplishment
If do-not-break-the-streak or chain works for you, this challenge is highly recommended

Cons

I believe in complete digital detoxification for one day during the week / fortnight. Obviously, this could not happen with the #100DaysOfCode challenge
Very easy to just go through the motions if fatigued

Mandelbrot and Julia - Mathematical visualization and beauty

Vidya — Wed, 20 Jan 2021 08:32:53 +0000

I have been fascinated by the idea of chaos theory since I read about the butterfly effect. I read some more about it in the book 'Chaos' by James Gleick including about the Mandelbrot. But until I read 'Make your own Mandelbrot' by Tariq Rashid I did not realize just how simple the equation to draw the Mandelbrot was.

This is the innocuous looking equation that generates the awesome Mandelbrot Set.

A slight change in the input to the same equation will give the Julia set. It gives rise to a complicated boundary that reveals a recursive aspect; there are many smaller sized Mandelbrot one can see as one zoom's in.

To learn more - look at this video - "What's so special about the Mandelbrot Set?"

So I used the code from the book (python) to get the Mandelbrot set and the Julia set. I also got to see the 3D surface close ups (after smoothing).

MandelBrot 3D

Julia 3D

But the fun of the Mandelbrot is when one can zoom in and see the beautiful patterns interactively.

After a bit of search (and multiple tries), I was able to finalize on JavaScript code works reasonably without needing specific libraries or frameworks or threads. I also improved on the color palette, to get brightly colored areas to observe the beauty of the mathematical patterns.

Following is the core of the Mandelbrot code. This calculates the value of the function for a maximum number of iterations.

Results

Using mousedown to zoom in, ctrl to zoom out and shift to pan in the Mandelbrot and Julia results, the images (converted to gifs) is shown below.

Links to the live demo are also mentioned at the end of the post.

Mandelbrot gif

Julia gif

Links

Live Demo - Mandelbrot

Live Demo - Julia

Full Code Mandelbrot in JS

Full Code Julia in JS

Python notebook with 3D images - Mandelbrot

Python notebook with 3D images - Julia

References

1. Make your own Mandelbrot - Tariq Rashid

If you don't know programming or mathematics, and want to start from scratch to understand the concept behind Mandelbrot in clear, simple and step-by-step manner, this book is great. Short (only ~130 pages), with plenty of illustrations, easy to grasp. Its only con is that the code is in Python 2.7.

2. HTML5 Canvas and Javascript - Main resource for the javascript code for zoom and pan function.

3. Color palette - Simplest color palette code for a wide range of colors

4. Various values of C for Julia sets

5. Color Inspiration

Book Review - Building Machine Learning Powered Applications

Vidya — Tue, 12 Jan 2021 10:50:49 +0000

Book Review - Building Machine Learning Powered Applications by Emmanuel Ameisen

I found this book a bit frustrating to follow.

Wherever the code is shown in the book and a link provided - it is to the main github repo and not to the individual jupyter notebook or python code. Nor was the filename mentioned beneath the code snippet.

Not a deal breaker but it is frustrating.

Pros

Covers a lot of theory on building a machine learning application
You build a ML Editor that gives recommendations
Important concepts are illustrated in specific notebooks
Incremental changes to the model for better results

Cons

More theory, less code
Light on deployment details
Book links point only to the main repo and not the specific file with the code
Not all tied together (filter models / tests).

My github repo - It contains the chapter numbers in the filenames.

Some images from the final built machine learning app with recommendations.

1. Main page of ML Editor

2. Model V1 - Sample Question

3. Model V1 - Readability score

4. Model V2 - Sample Question

5. Model V2 - Question Score

6. Model V3 - Sample Question

7. Model V3 - Detailed Recommendations

8. Model V3 - Another Recommendation example (Copied from writers.stackexchange.com)

Book Review — Grokking Algorithms by Aditya Bhargava

Vidya — Wed, 30 Dec 2020 07:55:41 +0000

Book Review — Grokking Algorithms by Aditya Bhargava

The word 'grokking' refers to 'understand (something) intuitively or by empathy' and it is the intuition part that this book attempts to impart via some really awesome illustrations.

First off - this is not a book for people who have completed Computer Science or have learnt algorithms. However it IS a wonderful book for beginners or as is mentioned on the cover 'for curious people'.

The chapter on dynamic programming alone makes this book highly worth reading.

Pros -

Very easy to understand and grasp the underlying intuition
Everyday relatable examples of the algorithm usage
The high quality illustrations
Written in a fun and engaging manner so easy to dive in and keep reading
Chapters are independent, so can pick without needing to read in sequence

Cons -

Minimal algorithms covered and in less depth
Not a substitute for the more formal books in this area
Will not help you master algorithms. This book is to start the journey into algorithms

Minimalist self-updating Github profile README

Vidya — Tue, 29 Sep 2020 06:54:06 +0000

Github's new profile readme is all the rage right now. However I wanted something simple, clean, minimalist and entirely within the Github ecosystem.

Creating a self-updating profile readme helped me -

Start a technical blog and TIL blog within Github itself
Retrieve relevant information from within Github ecosystem (no separate blog needed)

This self-updating readme has been inspired from the work at
Building a self-updating profile README for GitHub by Simon Willison.

Following are my steps towards creating the self-updating profile readme.

1. Create a technical blog within Github

I used the Github Learning Lab for GitHub Pages to learn how to create a blog with Github. The lab got me up to speed very quickly.

I chose the simple look of minima theme with the solarized-dark skin. How it looks

2. Create a TIL repo within Github

This blog was to retain snippets of 'Today-I-learned'. I used the same approach as for creating the technical blog. Used the basic Minimal theme for its clean design. How it looks

3. Self-Updating the readme for the TIL repo using Github actions.

This is to update the repo whenever a new TIL is added. The count of the total number of TILs is also updated.

The following files from the repo are needed to self-update the readme.

requirements.txt (No change required)
update_readme.py (No change required)
README.md (Personalize this)
build_database.py (Personalize the url shown below) ```python

url = "https://github.com/vidyabhandary/til/blob/master/{}".format(path)

- Create a manual workflow using Github actions
This will create a `./github/workflows` directory with a `main.yml` file.
Can rename this to `build.yml` or retain the existing name.
Update with code from [build.yml](https://github.com/vidyabhandary/TIL/blob/master/.github/workflows/build.yml).

## And finally ...

### 4. Create a repo with the same name as your Github ID
This is the special repository. Its README.md will appear on your public profile!

### 5. Self-updating read-me for the Github profile.
The following files from the [repo](https://github.com/vidyabhandary/vidyabhandary) are required to
create the self-updating read-me for Github.

- `requirements.txt` *(No change required)*
- `README.md` *(Personalize this)*
- `build_readme.py` *(Personalize the urls shown below)*

```python


til_readme = "https://raw.githubusercontent.com/vidyabhandary/TIL/master/README.md"
entries = feedparser.parse("https://vidyabhandary.github.io/blog/feed.xml")["entries"]

Create a manual workflow using Github actions This will create a ./github/workflows directory with a main.yml file. Update with code from main.yml.

And that's it !!!!

You now have a self-updating profile readme in Github that updates once a day with your blog posts and TILs.

DEV Community: Vidya

The Composer's Quill: Weaving Melodies using text prompts with MusicGen

Step by Step Guide to generate music in Google Colab

1. Go to https://github.com/camenduru/MusicGen-colab

Do star his repo !!!

2. Scroll down to reach the colab notebooks

3. Choose any colab notebook and click on "Open in Colab".

(You will need gmail ID to use colab.)

4. Loading ... This may take awhile.

5. The magical colab notebook with just 5 lines !!! Click on the arrow.

6. You will get a warning message - Click "Run anyway".

7. The commands in the cell will get executed - the git repo will be cloned and the required libraries will get installed This step could take a while, even with the notebook having a setting of GPU by default.

8. Ignore the error in red and click on the gradio live link.

9. The Music Generation playground !!!

No amount of exclamations can tell how exciting it is to be able to create music with text especially for a plebian !

10. Add your music prompt. This was my prompt.

11. Click on generate to process your text prompt to generate music.

12. The model gives the time it will take to create the music along with the number of steps and %age of completion.

13. Some of the other options to check out -

14. And finally - The generated music result ! You can enjoy listening to your creation in the browser as well as download your creation to your drive as a mp4 file.

15. Trying the melody option to get a different version of the beats I like.

16. The result is not bad, although the original is definitely better. Lots of steps and processing when asked 40 second composition. The option with change in temperature came out pretty good.

"So finally I can compose music", without knowing how to write musical notes !!

"What a time to be alive !!!"

Business at the Speed of AI

Key points of Consistent Hashing

Key points of Consistent Hashing

Book Review - Learning React - 2nd Edition by Alex Banks and Eve Porcello

Book Review - Learning React - 2nd Edition by Alex Banks and Eve Porcello

#100DaysOfCode Challenge

100 Days Of Code Github Wall

Mandelbrot and Julia - Mathematical visualization and beauty

MandelBrot 3D

Julia 3D

MandelBrot Code

Julia Code

Color Palette Code

Results

Mandelbrot gif

Julia gif

Links

References

1. Make your own Mandelbrot - Tariq Rashid

2. HTML5 Canvas and Javascript - Main resource for the javascript code for zoom and pan function.

3. Color palette - Simplest color palette code for a wide range of colors

4. Various values of C for Julia sets

5. Color Inspiration

Book Review - Building Machine Learning Powered Applications

Pros

Cons

Some images from the final built machine learning app with recommendations.

1. Main page of ML Editor

2. Model V1 - Sample Question

3. Model V1 - Readability score

4. Model V2 - Sample Question

5. Model V2 - Question Score

6. Model V3 - Sample Question

7. Model V3 - Detailed Recommendations

8. Model V3 - Another Recommendation example (Copied from writers.stackexchange.com)

Book Review — Grokking Algorithms by Aditya Bhargava

Book Review — Grokking Algorithms by Aditya Bhargava

Pros -

Cons -

Minimalist self-updating Github profile README

Creating a self-updating profile readme helped me -

1. Create a technical blog within Github

2. Create a TIL repo within Github

3. Self-Updating the readme for the TIL repo using Github actions.

And that's it !!!!

The final look - https://github.com/vidyabhandary