Skip to content

Instantly share code, notes, and snippets.

View suxue's full-sized avatar

FEI Hao suxue

12 followers · 5 following
  • CSIT of ANU
  • ACT, Australia
View GitHub Profile
@suxue
suxue / lightencnn.prototxt
Created February 16, 2017 02:00
name: "DeepFace_set003_net"
input: "data"
input_dim: 1
input_dim: 1
input_dim: 128
input_dim: 128
layer{
name: "conv1"
type: "Convolution"
@suxue
suxue / det4.prototxt
Created February 16, 2017 03:30
name: "LNet"
input: "data"
input_dim: 1
input_dim: 15
input_dim: 24
input_dim: 24
layer {
name: "slicer_data"
type: "Slice"
@suxue
suxue / multi_index_example.cpp
Created October 22, 2018 11:10
Boost multiple-index example
#include <boost/flyweight.hpp>
#include <boost/multi_index_container.hpp>
#include <boost/multi_index/member.hpp>
#include <string>
#include <cstdint>
#include <vector>
#include <iostream>
#include <tuple>
typedef std::tuple<short, std::uint8_t, std::uint8_t> Date;
@suxue
suxue / multi_index.cpp
Created October 22, 2018 11:11
multi_index
#include <boost/flyweight.hpp>
#include <boost/multi_index_container.hpp>
#include <boost/multi_index/member.hpp>
#include <string>
#include <cstdint>
#include <vector>
#include <iostream>
#include <tuple>
typedef std::tuple<short, std::uint8_t, std::uint8_t> Date;
@suxue
suxue / 1.cpp
Created October 22, 2018 11:12
multi_index
#include <boost/flyweight.hpp>
#include <boost/multi_index_container.hpp>
#include <boost/multi_index/member.hpp>
#include <string>
#include <cstdint>
#include <vector>
#include <iostream>
#include <tuple>
typedef std::tuple<short, std::uint8_t, std::uint8_t> Date;
@suxue
suxue / wrk.py
Created May 30, 2019 12:45
#!/usr/bin/env python
# coding: utf-8
# module for http benchmarking using wrk
# example command like
# ./wrk.py --connections '[1,2,3,4]' --filename ./benchmark_qa/query.json --url http://x.y.z.k/hello.php --output 1.csv --duration 600
import subprocess
import tempfile
import urlparse
from contextlib import contextmanager
@suxue
suxue / trt-bert.ipynb
Created March 11, 2020 12:06
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@suxue
suxue / llama_torchao_compile.py
Created August 29, 2024 09:01 — forked from SunMarc/llama_torchao_compile.py
`transformers` + `torchao` quantization + `torch.compile` on Llama3.1 8B
# REQUIRES torchao, torch nightly (or torch 2.5) and transformers
from transformers import AutoTokenizer, AutoModelForCausalLM, TorchAoConfig
from transformers import TextStreamer
import torch
from tqdm import tqdm
import os
os.environ["TOKENIZERS_PARALLELISM"] = "false" # To prevent long warnings :)
torch.set_float32_matmul_precision('high')