Activity

r.nick0510

a hell lot of more grinding. I want to hit 10 hours today!

Attachment
0
r.nick0510

grinded half of the dp section in cses

Attachment
0
r.nick0510

I grinded a bit on segment trees. saved beats for later

Attachment
0
r.nick0510

I have been grinding for a while. I have finished a huge portion of the stuff!

Attachment
0
r.nick0510

So today I worked more on the Representation learning part. I create embeddings by using an AutoEncoder on the music. The embeddings are the output of the bottleneck layer. I also clustered the embeddings using DBSCAN (I did not use K-Means because I did not want to assume anything about the data distribution).
The plan for tomorrow is to work on the part where we improve the clustering. Right now I am thinking of doing deep clustering. So deep clustering is basically first getting the clusters of all the embeddings and then training the model not only to reconstruct the input but also to classify the cluster in a multi-task learning wise fashion. How does this help? Well doing this will decrease the distance between elements of the same cluster and increase the distance between different clusters (hopefully). We then recluster and run the cycle multiple times, in the end we get a neural network that gives embeddings that have clusterable patterns. After getting the patterns, I have some interesting plans! (follow the project for more updates)

yours,
Raj

Attachment
Attachment
0
r.nick0510

Here is what I did today:
I got the audio encoding done part. I first split the audio into small segments, after this Constant-Q transform is applied to these segments to make them into a format that a neural network can process. The images are the visualization of 2 songs (one of them is never gonna give you up, so you got rick rolled).

What is this project about you might ask, well ill try to explain it. What im trying to do is to take the patterns that can be found in music, and then use these patterns to create tabular data, The end goal is to crank out as many datasets as possible so that I could become a grandmaster in kaggle!

Attachment
Attachment
Attachment
Attachment
0
r.nick0510

Been a while guys, today I did fenwick tree. ill do problems tomorrow!

Attachment
0
r.nick0510

Baseline Le Nouveau is getting ready guys!

Attachment
0
r.nick0510

worked a bit on this, though not as much as before. holidays are starting so I should be able to increase my grind time hopefully!

Attachment
0
r.nick0510

closing up for today, finished the baselines, ready for tomorrow’s sprint round!

Attachment
0
r.nick0510

I got the baseline player to work (finally). now im making some changes to make it perform better!

Attachment
0
r.nick0510

This is it for today, I mostly finished the baseline robot for my team to work on. This will provide a starting pipeline so that further changes could be made. I only could not get some parts to work on which I will work tomorrow.

Attachment
0
r.nick0510

building the baseline lets get cooking!

Attachment
0
r.nick0510

Now that I have completed about 20 problems in floyd-warshall. I think I’m ready to go for special graphs, Ill grid till 10 pm then Ill go for battle code grind

Attachment
0
r.nick0510

lets go, a new day and lets start cooking, I need to finish floyd-warshall by today!

Attachment
2

Comments

aneezakiran07
aneezakiran07 2 months ago

make sure u understand the dynamic programming first TT

r.nick0510
r.nick0510 2 months ago

thanks for the support @aneezakiran07