crcrpar

FROM nvidia/cuda:10.0-cudnn7-devel

RUN apt-get update -y && \
    apt-get install -y --no-install-recommends \
    python3-dev \
    python3-pip \
    python3-wheel \
    python3-setuptools \
    git \
    cmake \

crcrpar

import typing

import chainer
from chainer import functions
from chainer import links
from chainer import types


class AttentionAugmentedConvolution2D(chainer.Chain):

crcrpar

FROM swift:5.0.1-xenial

# Install git, process tools
RUN rm -rf /var/lib/apt/lists/* \
    && apt-get update -y \
    && apt-get -y --no-install-recommends install \
        git \
        procps \
        aria2 \
        libsqlite3-dev \

crcrpar

#include <algorithm>
#include <iostream>
#include <vector>


int * func1(int * x, int n) {
  // for (int i = 0; i < ++n; ++i) {
  for (int i = 0; i < n; ++i) {
    std::cout << "i = " << i << ", x = " << x << std::endl;
    ++(*(x++));

crcrpar

// /Library/Developer/Toolchains/swift-tensorflow-DEVELOPMENT-2019-06-17-a.xctoolchain
import TensorFlow


public struct SNConv2D<Scalar: TensorFlowFloatingPoint>: Layer {

    // Avoid below errors:
    // TensorFlow.Layer:2:20: note: protocol requires nested type 'Input'; do you want to add it?
    // TensorFlow.Layer:3:20: note: protocol requires nested type 'Output'; do you want to add it?
    public typealias Input = Tensor<Scalar>

crcrpar

/*
 * Introduction to Curiously Recurring Template Pattern
 *
 * Ref: https://theolizer.com/cpp-school2/cpp-school2-19/
 */
#include <iostream>
#include <iomanip>
#include <cstdint>
#include <tuple>

crcrpar

crcrpar

crcrpar

crcrpar

# data: fastai's DataBunch

def objective(trial: optuna.Trial):
    num_layers = trial.suggest_int('n_layers', 1, 5)  # `num_layers` is 1, 2, 3, 4, or 5.
    layers, ps = [], []  # define the number of unit of each layer / the ratio of dropout of each layer
    for i in range(n_layers - 1):  # `TabularModel` automatically adds the last layer.
        num_units = trial.suggest_categorical(f'num_units_layer_{i}', [800, 900, 1000, 1100, 1200])
        p = trial.suggest_discrete_uniform(f'dropout_p_layer_{i}', 0, 1, 0.05)
        layers.append(num_units); ps.append(p)