syed-ahmed

import torch
import ctypes
 
def print_binary_val(tensor):
    print(bin(ctypes.c_uint.from_buffer(ctypes.c_float(tensor.item())).value))
 
# Error 1: Large round-off incurred when
# representing the same number in a lower precision.
A = torch.tensor(1.0001, dtype=torch.float32).cuda()
B = A.half()

syed-ahmed

pld_dir?=/media/lilbirb/research/code/pld/
src=$(pld_dir)/do_blink/symbiflow_benchmark/fig2_at200/1600/src
base=$(abspath .)

symb_dir?=$(pld_dir)/third_party/symbiflow-arch-defs
do_blink_dir?=$(pld_dir)/build/do_blink/rrgraphs
YOSYS = $(symb_dir)/env/conda/envs/symbiflow_arch_def_base/bin/yosys
VPR?=$(pld_dir)/third_party/vtr-verilog-to-routing/vpr/vpr
GENFASM = $(symb_dir)/env/conda/envs/symbiflow_arch_def_base/bin/genfasm
XCFASM = $(symb_dir)/env/conda/envs/symbiflow_arch_def_base/bin/xcfasm

syed-ahmed

set_property DONT_TOUCH true [get_cells design_1_i/page_0/inst]
set_property HD.RECONFIGURABLE true [get_cells design_1_i/page_0/inst]
set_property DONT_TOUCH true [get_cells design_1_i/leaf_dummy_2/inst]
set_property HD.RECONFIGURABLE true [get_cells design_1_i/leaf_dummy_2/inst]
set_property DONT_TOUCH true [get_cells design_1_i/leaf_dummy_3/inst]
set_property HD.RECONFIGURABLE true [get_cells design_1_i/leaf_dummy_3/inst]
set_property DONT_TOUCH true [get_cells design_1_i/leaf_dummy_4/inst]
set_property HD.RECONFIGURABLE true [get_cells design_1_i/leaf_dummy_4/inst]
set_property DONT_TOUCH true [get_cells design_1_i/leaf_dummy_5/inst]

syed-ahmed

#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <sys/socket.h>
#include <linux/if_alg.h>
#include <linux/socket.h>

#define SHA384_DIGEST_SZ 48

int main(void) {

syed-ahmed

#ifndef _APINT
#define _APINT
#define NULL __null


template <int Bits>
using ap_uint = __attribute__((__ap_int(Bits))) unsigned ;

template <int Bits>
using ap_int = __attribute__((__ap_int(Bits))) int ;

syed-ahmed

\documentclass{article}
\usepackage[utf8]{inputenc}
\usepackage{enumitem}

\title{ESE-532 Homework 1}
\author{James Bond \\ Partner: Donald Duck}
\date{\today}

\begin{document}
\maketitle

syed-ahmed

FROM ubuntu:18.04

ARG PYTHON_VERSION=3.6
RUN apt-get update && apt-get install -y --no-install-recommends \
         ca-certificates wget \
         gnupg2 \
         build-essential \
         cmake \
         git \
         curl \

syed-ahmed

diff --git a/test/common_quantization.py b/test/common_quantization.py
index 188fbf200..13dbc6358 100644
--- a/test/common_quantization.py
+++ b/test/common_quantization.py
@@ -55,6 +55,10 @@ class QuantizationTestCase(TestCase):
         self.train_data = [(torch.rand(20, 5, dtype=torch.float), torch.randint(0, 1, (20,), dtype=torch.long)) for _ in range(20)]
         self.img_data = [(torch.rand(20, 3, 10, 10, dtype=torch.float), torch.randint(0, 1, (20,), dtype=torch.long))
                          for _ in range(20)]
+        fullname = self.id().lower()
+        is_cuda_test = 'gpu' in fullname or 'cuda' in fullname or 'cuda' in torch.tensor([]).device.type

syed-ahmed

Summary: Time (seconds) to get 134217728 randoms with varying number of threads

Best and Worst Average Times per Thread

syed-ahmed

diff --git a/aten/src/ATen/CPUGenerator.cpp b/aten/src/ATen/CPUGenerator.cpp
index c0b6e30..97cbb4d 100644
--- a/aten/src/ATen/CPUGenerator.cpp
+++ b/aten/src/ATen/CPUGenerator.cpp
@@ -50,6 +50,7 @@ inline uint64_t make64BitsFrom32Bits(uint32_t hi, uint32_t lo) {
 CPUGenerator::CPUGenerator(uint64_t seed_in)
   : Generator{Device(DeviceType::CPU)},
     engine_{seed_in},
+    engine2{seed_in, 0, 0},
     next_float_normal_sample_{c10::optional<float>()},