Rishith P

22 y/o SWE@Nokia BLR | Learning ML | Football fan | Interested in geopolitics & stock market | Sharing my journey & daily learnings.
🇮🇳

[Rishith P @rishiiiiiiii06]

@likithsh18 Garmin has changed you

[Rishith P @rishiiiiiiii06]

@ThenNowForeve @_classof92_ India has an 80% literacy rate and you aree doing your part to keep that 20% alive

[Rishith P @rishiiiiiiii06]

@Devvops Let's gooo

[jitesh💙 @Jitesh_117]

someone should make a vscode extension which plays "FAAAAH" sound whenever a test fails

[susa @prinkasusa]

You woke up at 7 AM & it's raining outside... Mom smiles and says,“No school today”

You stay under the blanket a little longer. Cold weather.. warm bed.. zero worries.

You turn on the TV.. Watching Doraemon or Shinchan.. No pressure.. No responsibilities.

Back then, Happiness didn't need any reasons. We didn't chase peace.. We lived in it.

Back then, life WAS good.

[Rishith P @rishiiiiiiii06]

Day 67 ( feb 16)
- NLP introduction
- word embeddings
-visualisation using t sne
- parameter reduction
- problems with one hot
- transfer learning
- properties of word embedding
- embedding matrix
#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

Day 66 (feb 13)

- bidirectional rnn
- deep rnn
#MachineLearning
#DeepLearning
#BuildInPublic

[neural nets. @cneuralnetwork]

I'm not even a political person but WTF
YOUR COUNTRY LITERALLY MADE A HUGE JUMP IN AI TODAY
WE HAD A DEEPSEEK MOMENT

But yes pls highlight the one moment out of 100000 moments where we failed

I don't think this party is worthy to even be called an opposition, because the opposition should think for bettering the country

never voting for them.

[Rishith P @rishiiiiiiii06]

Day 65
- gradient clipping
- gated recurrent unit (gru)
- How gru solves the problem of vanishing gradient
-candidate vs final memory
-long short term memory
-intuition of each gate is lstm (forget , update and output gate)
#MachineLearning
#DeepLearning
##BuildInPublic

[Rishith P @rishiiiiiiii06]

Day 64
Couldn't study much as the whole day went in traveling and office work.

- vanishing gradients in rnn
-long term dependencies problem

#DeepLearning
#MachineLearning
#BuildInPublic

[Ahmad @TheAhmadOsman]

https://t.co/YoXbHmTEjg

[Rishith P @rishiiiiiiii06]

Day 63
- defining vocabulary, token generation , initialize parameters, forward pass , loss computation, backpropagation through time , update weight, generation of sentences based on probability.
-sampling of sequences
#MachineLearning
#DeepLearning
#BuildingInPublic

[Rishith P @rishiiiiiiii06]

@Mann_Bhatt88 Sure buddy 🙌

[Rishith P @rishiiiiiiii06]

Day 62 of ML
- Weight update in rnn
- different types of rnn
-language models and sequence generation
-chain rule of probability
-how rnn language model is structured

#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

Day 61 (feb 6)
-RNN model and problem with regualar NN
-forward propagation in rnn , backpropagation through time.
-gradient flow in hidden state

#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

Day 60
-Intersection over union
-non max suppression and yolo algo
Started with sequence models
- sequence model vs neural network
-notations for sequence model
-one hot encoding

#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

Day 59
- Dived deep into replacing FC layer with 1×1 convolution layer
- Classic CNN and CNN with 1×1 convolution pipeline (from raw data to output)
- Convolution implementation of sliding window
- Bounding box prediction and YOLO
#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

First marathon done. New PRs in the 5K and 10K along the way. First of many.

[Rishith P @rishiiiiiiii06]

Day 57
-Inception networks
-How is number of filters decided
-MobileNet
-Transfer learning
-Freezing layers in transfer learning
#MachineLearning
#DeepLearning
#BuildInPublic

[Rishith P @rishiiiiiiii06]

@bryanalsiregar To transition from software engineer to ML engineer

[Rishith P @rishiiiiiiii06]

Day 56
-Identity mapping in resnet.
-how and why exactly resnet solves the limitation of a simple NN
-Network inside a Network; how a 1×1 convolution solves the problem of better understanding of data
#MachineLearning
#DeepLearning
#BuildInPublic