stephenhardy

-- Remove the history from 
rm -rf .git

-- recreate the repos from the current content only
git init
git add .
git commit -m "Initial commit"

-- push to the github remote repos ensuring you overwrite history
git remote add origin git@github.com:<YOUR ACCOUNT>/<YOUR REPOS>.git

craffel

import matplotlib.pyplot as plt

def draw_neural_net(ax, left, right, bottom, top, layer_sizes):
    '''
    Draw a neural network cartoon using matplotilb.
    
    :usage:
        >>> fig = plt.figure(figsize=(12, 12))
        >>> draw_neural_net(fig.gca(), .1, .9, .1, .9, [4, 7, 2])
    

harpone

import tensorflow as tf
from tensorflow.python.framework import ops
import numpy as np

# Define custom py_func which takes also a grad op as argument:
def py_func(func, inp, Tout, stateful=True, name=None, grad=None):
    
    # Need to generate a unique name to avoid duplicates:
    rnd_name = 'PyFuncGrad' + str(np.random.randint(0, 1E+8))
    

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

rtqichen

## Weight norm is now added to pytorch as a pre-hook, so use that instead :)

import torch
import torch.nn as nn
from torch.nn import Parameter
from functools import wraps

class WeightNorm(nn.Module):
    append_g = '_g'
    append_v = '_v'

MInner

import datetime
import linecache
import os

import pynvml3
import torch

print_tensor_sizes = True
last_tensor_sizes = set()
gpu_profile_fn = f'{datetime.datetime.now():%d-%b-%y-%H:%M:%S}-gpu_mem_prof.txt'

peteflorence

Here's a simple implementation of bilinear interpolation on tensors using PyTorch.

I wrote this up since I ended up learning a lot about options for interpolation in both the numpy and PyTorch ecosystems. More generally than just interpolation, too, it's also a nice case study in how PyTorch magically can put very numpy-like code on the GPU (and by the way, do autodiff for you too).

For interpolation in PyTorch, this open issue calls for more interpolation features. There is now a nn.functional.grid_sample() feature but at least at first this didn't look like what I needed (but we'll come back to this later).

In particular I wanted to take an image, W x H x C, and sample it many times at different random locations. Note also that this is different than upsampling which exhaustively samples and also doesn't give us fle

apaszke

import sys
from collections import OrderedDict

PY2 = sys.version_info[0] == 2
_internal_attrs = {'_backend', '_parameters', '_buffers', '_backward_hooks', '_forward_hooks', '_forward_pre_hooks', '_modules'}


class Scope(object):
    def __init__(self):
        self._modules = OrderedDict()

Miladiouss

import torchvision
import torchvision.transforms as transforms
from torchvision.datasets import CIFAR10
from torch.utils.data import Dataset, DataLoader
import numpy as np

# Transformations
RC = transforms.RandomCrop(32, padding=4)
RHF = transforms.RandomHorizontalFlip()
RVF = transforms.RandomVerticalFlip()

sajadn

import torch.nn as nn
from torch import _weight_norm
from torch.nn.utils import weight_norm


def data_dependent_init(model, data_batch):
    def init_hook_(module, input, output):
        std, mean = torch.std_mean(output, dim=[0, 2, 3])
        g = getattr(module, 'weight_g')
        g.data = g.data/std.reshape((len(std), 1, 1, 1))