Data Liftoff

Python-Data-Science-Projects to recreate for everybody. Have a good day!

[Data Liftoff @Data_Liftoff]

For the quick training of neural networks with Google TPUs, I have published an introduction to data for self-driving cars with the associated code on Kaggle.
https://t.co/KQC9caRGDT

[Data Liftoff @Data_Liftoff]

Using Automated Evolutionary Algorithms to optimize a deep neural network on mnist.
#Kaggle #DeepLearning
https://t.co/3oW2WFEKRS

[Data Liftoff @Data_Liftoff]

Classify doodles with a 1D-CNN and a sequence of strokes:
#Kaggle
https://t.co/8KoYcwk1eJ

[Data Liftoff @Data_Liftoff]

How to train a deep network to predict the sentiment of a text. The network uses a sequence of letters as input and can be trained with any language without textual preprocessing: #kaggle
https://t.co/VUAkPAHvTr

[Data Liftoff @Data_Liftoff]

Comparing Redommender Engines and trying different libraries for the first time:
# Kaggle
https://t.co/oqkk9guU7s

[Data Liftoff @Data_Liftoff]

If you like to explore salaries of data scientists around the world, this notebook is for you: #Kaggle
https://t.co/eBFPlvrIgt

[Data Liftoff @Data_Liftoff]

Day 15: #100DaysOfMLCode
319 places up. Not unlike my electricity bill. #kaggle - https://t.co/5jOytxUsvn

[Data Liftoff @Data_Liftoff]

Jumped 134 spots on #kaggle. I'll sleep when I'm dead. https://t.co/5jOytxUsvn

[Data Liftoff @Data_Liftoff]

Day 14: #100DaysOfMLCode
I've been preparing a flow to compute gensim LDA on a corpus larger than your RAM. The code to optimize it is nearly complete.

[Data Liftoff @Data_Liftoff]

Day 13: #100DaysOfMLCode
Today I compared LDA in gensim and sklearn on different corpi from #Kaggle with and without multicore implementation. Interestingly sklearn needs more time with multiple cores and right now I can't explain why the single core gensim explodes.

[Data Liftoff @Data_Liftoff]

Day 12: #100DaysOfMLCode
Today I used #Kaggle Kernels to compare the LDA (topic modeling) implementation in sklearn and gensim.
Gensim seems to be "better" and faster.
https://t.co/6tSBn2mFO2

[Data Liftoff @Data_Liftoff]

Day 11: #100DaysOfMLCode
To compare different implementations of Latent Dirichlet Allocation (Topic Modelling) in the libraries (gensim, sklearn, lda) to each other I've been reading a lot of documentations today.

[Data Liftoff @Data_Liftoff]

Day 10: #100DaysOfMLCode
Finally it worked! We combined #Alexa with a custom Skill to tell you when to refuel your car and where to expect the best deal in your (german) town. #DoHack2018

[Data Liftoff @Data_Liftoff]

Day 9: #100DaysOfMLCode
Pulling an all-nighter to finish our #Raspberry #Alexa application at the #DoHack2018

[Data Liftoff @Data_Liftoff]

Day 8: #100DaysOfMLCode
Learning about stationary time series while participating at the #DoHack2018
https://t.co/aBho9bY2On

[Data Liftoff @Data_Liftoff]

Day 7: #100DaysOfMLCode
Started listening to the podcasts (machine learning, artificial intelligence, ...) of @DataSkeptic while traveling.
https://t.co/tEY5sjpu9T

[Data Liftoff @Data_Liftoff]

Day 6: #100DaysOfMLCode
I read tutorials for topic modeling with gensim and I am excited to experiment with some datasets from #Kaggle on my own.
https://t.co/uYXeZtErXT
https://t.co/mcmGbm7249

[Data Liftoff @Data_Liftoff]

Day 5: #100DaysOfMLCode
Started reading Ian Goodfellows book:
https://t.co/Kf5rBXx1PR

[Data Liftoff @Data_Liftoff]

This is a short introduction on IPython magic functions in #Kaggle notebooks.
Have a good day!
https://t.co/LpiA9b9xhg

[Data Liftoff @Data_Liftoff]

Day 4: #100DaysOfMLCode
Deepened my knowledge of IPython magic-functions to optimize execution-time and memory-usage with this tutorial.
https://t.co/1yvQCcWjg7