danoneata

"""Usage:

python aggregate_tree.py 500000
python aggregate_tree.py -3000

"""

from matplotlib import pyplot as plt
import numpy as np
import pdb

danoneata

% Read a set of vectors stored in the fvec format (int + n * float)
% The function returns a set of output vector (one vector per column)
%
% Syntax: 
%   v = fvecs_read (filename)     -> read all vectors
%   v = fvecs_read (filename, n)  -> read n vectors 
%   v = fvecs_read (filename, [a b]) -> read the vectors from a to b (indices starts from 1)
function v = fvecs_read (filename, bounds)

% open the file and count the number of descriptors

danoneata

class Expr(object):
    def accept(self, visitor):
        method_name = 'visit_{}'.format(self.__class__.__name__.lower())
        visit = getattr(visitor, method_name)
        return visit(self)


class Int(Expr):
    def __init__(self, value):

danoneata

def descriptors_to_sufficient_statistics(xx, gmm, **kwargs):
    # yael assumes that the data is in C-contiguous format.
    xx = np.ascontiguousarray(np.atleast_2d(xx))

    N = xx.shape[0]
    K = gmm.k
    D = gmm.d

    # Compute posterior probabilities using yael.
    Q = gmm_predict_proba(xx, gmm)  # NxK

danoneata

/**
  * Created by eurus on 28/02/2017.
  */
package components

import collection.immutable.Queue
import ActionType._

sealed trait SignalValue
object `0` extends SignalValue {

danoneata

from typing import (
    List,
    TypeVar,
)

# Type aliases
A = TypeVar('A')
Digit = int
Matrix = List[List[A]]
Grid = Matrix[Digit]

danoneata

import logging

from gensim.models.doc2vec import (
    Doc2Vec,
    TaggedDocument,
)

logging.basicConfig(
    format='%(asctime)s : %(threadName)s : %(levelname)s : %(message)s',
    level=logging.DEBUG,

danoneata

Course info

• Motivation: Deep learning in vision, speech, text, robotics.

• Deep learning courses:

  • Andrew Ng: Deep Learning
  • fast.ai: Practical Deep Learning For Coders
  • Stanford: CS231n Convolutional Neural Networks for Visual Recognition

• Oxford: Machine Learning

danoneata

from functools import singledispatch


class Document:
    pass


class Paragraph(Document):
    def __init__(self, text):
        self.text = text

danoneata

import numpy as np


def power_normalize(xx, alpha=0.5):
    """Computes a alpha-power normalization for the matrix xx."""
    return np.sign(xx) * np.abs(xx) ** alpha


def L2_normalize(xx):
    """L2-normalizes each row of the data xx."""