amjames

import math
import torch
import triton
import triton.language as tl
from typing import Optional, Tuple



def _has_triton():
    if not torch.cuda.is_available():

amjames

import torch
import torch._dynamo as torchdynamo

torchdynamo.config.verbose = True
torchdynamo.config.suppress_errors = True

class TestCasePlaceholder:
    def assertTrue(self, v):
        assert v, "Expected True"

amjames

"""
Prior to adding the checks which error out when to_dense is called with hybrid (having dense dims) tensors
all of these examples triggered a segfault

The commented out lines trigger the error now, previously segfault
"""
import torch
dense = torch.randn(3, 4, dtype=torch.float32)
dense = dense * dense.relu().bool()
csr = dense.to_sparse_csr()

amjames

def _array_from_fchk(fchk_text, array_name):
    # matches a line that looks like:
    # {array_name}          R        N=     {number of elements in array}
    matcher = re.compile(r'\A(?P<title>{})\s+R\s+N=\s+(?P<nele>\d+)\Z'.format(array_name))
    fchk_lines = fchk_text.split('\n')
    start_line = 0
    nline = 0
    found_match = False
    for i, line in enumerate(fchk_lines):
        # see if the line matches the regex

amjames

Memory used after allocating lg_array: 800
Memory used after creating sm_vector: 840
sm_vector = [ 0, 1, 2, 3, 4,]
sm_vector (after modify) = [ 100, 101, 102, 103, 104,]
First 5 elements of the original lg_array = [ 0, 1, 2, 3, 4,]
size of allocation lg_array:          800
size of allocation sm_vector.data():  40

amjames

// I ran though this quick and expanded my useless comments with everything I could remember about this
// it was 2 years ago when I first wrote this.
// Some of it might be obvious to you, or just wrong, and I apologize but these ~200 lines are just about all of the js
// I have written in my life. Hope it helps!
google.maps.event.addDomListener(window, 'load', init);

var map;
var infowindow = null;
function init() {
  //googlemaps api docs can explain what each option does

amjames