Thanks to antville.org

discretize continuous features

shantanuo — Wed, 24 Apr 2024 09:33:38 GMT

You can "discretize" or "bin" continuous features into categorical features.

from sklearn.preprocessing import KBinsDiscretizer

kb = KBinsDiscretizer(n_bins=3, strategy='quantile', encode='ordinal')

kb.fit_transform(df['Fare'])

download food 11 dataset

shantanuo — Wed, 28 Aug 2019 10:31:00 GMT

The total file size of the Food-5K dataset is about 1.16 GB. Here are steps to download food 11 dataset.

mmspg.epfl.ch

Use these commands at linux prompt:

ftpopen tremplin.epfl.chFoodImage@grebvm2.epfl.chCahc1mooget Food-11.zip

audio visual modules for Machine Learning

shantanuo — Sat, 16 Mar 2019 11:08:11 GMT

Here is the list of some of the modules used for audio-visual processing using deep learning.

chroma: Feature Learning for Chord Recognition: The Deep Chroma ExtractorMFCC: Mel-frequency cepstral coefficients (MFCCs) are coefficients that collectively make up an MFCTCD-TIMIT: An Audio-Visual Corpus of Continuous SpeechlibrosaRBF: Radial Basis FunctionHMM: Hidden Markov ModelVGG: Resnet 50voxceleb-luigi: Luigi pipeline to download VoxCeleb audio from YouTube and extract speaker segmentsdeepfakesobamanetyolo

Extract words from Marathi wikipedia

shantanuo — Sun, 04 Nov 2018 08:06:09 GMT

Download the polyglot pickled marathi modelwget bit.lymv 12FI1xV polyglot-mt.pkl
Extract all the words from the embedding

from polyglot.mapping import Embeddingembeddings=Embedding.load("polyglot-mt.pkl")with open('myfile.txt', 'w') as fx:for i in embeddings.vocabulary:fx.write(i.encode('utf-8')+'\n')

Download the wordlist:s3.amazonaws.com