sai-prasanna

import csv
from typing import Dict, Optional
import logging
import torch
import random
from collections import Counter
import numpy as np

from overrides import overrides
from torch.utils.data import Dataset, IterableDataset, DataLoader, DistributedSampler

kylemcdonald

killeent

A Tour of PyTorch Internals (Part I)

The fundamental unit in PyTorch is the Tensor. This post will serve as an overview for how we implement Tensors in PyTorch, such that the user can interact with it from the Python shell. In particular, we want to answer four main questions:

1. How does PyTorch extend the Python interpreter to define a Tensor type that can be manipulated from Python code?
2. How does PyTorch wrap the C libraries that actually define the Tensor's properties and methods?
3. How does PyTorch cwrap work to generate code for Tensor methods?
4. How does PyTorch's build system take all of these components to compile and generate a workable application?

Extending the Python Interpreter

PyTorch defines a new package torch. In this post we will consider the ._C module. This module is known as an "extension module" - a Python module written in C. Such modules allow us to define new built-in object types (e.g. the Tensor) and to call C/C++ functions.

chrischoy

No more manual logging

The template run.ch provides a wrapper for running and managing experiments. The file will

• Log all stdout/stderr into a log file in a respective folder
• Maintail the log of all experiments with the details in manifest.md file.
• Update the experiments if you kill the job prematurely

Usage

keon

# coding: utf-8
"""
Seq2Seq (Encoder-Decoder) Model
this model is the basic encoder decoder model without attention mechanism.
author: Keon Kim
"""

import numpy as np
import torch as th
import torch.nn as nn

jihunchoi

def _sequence_mask(sequence_length, max_len=None):
    if max_len is None:
        max_len = sequence_length.data.max()
    batch_size = sequence_length.size(0)
    seq_range = torch.range(0, max_len - 1).long()
    seq_range_expand = seq_range.unsqueeze(0).expand(batch_size, max_len)
    seq_range_expand = Variable(seq_range_expand)
    if sequence_length.is_cuda:
        seq_range_expand = seq_range_expand.cuda()
    seq_length_expand = (sequence_length.unsqueeze(1)

MaximumEntropy

import torch
import torch.nn as nn
from torch.autograd import Variable
from torch.nn.utils.rnn import pad_packed_sequence, pack_padded_sequence

x = Variable(torch.randn(10, 20, 30)).cuda()
lens = range(10)

x = pack_padded_sequence(x, lens[::-1], batch_first=True)

genekogan

from bs4 import BeautifulSoup
import requests
import re
import urllib2
import os
import argparse
import sys
import json

# adapted from http://stackoverflow.com/questions/20716842/python-download-images-from-google-image-search

jfsantos

def train_fn(model, optimizer, criterion, batch):
    x, y, lengths = batch

    x = Variable(x.cuda())
    y = Variable(y.cuda(), requires_grad=False)

    mask = Variable(torch.ByteTensor(x.size()).fill_(1).cuda(),
            requires_grad=False)
    for k, l in enumerate(lengths):
        mask[:l, k, :] = 0

iamaaditya

# author: Aaditya Prakash

# NVIDIA-SMI does not show the full command, and when it was launched and its RAM usage.
# PS does but it does but you need PIDs for that
# lsof /dev/nvidia gives PIDs but only for the user invoking it

# usage:
#        python programs_on_gpu.py

# Sample Output