wonglkd

$ ../llvm/Build/bin/llc -filetype=obj -o fetch-soaap.o fetch.soaap                                                                                $ ../llvm/Build/bin/clang -o fetch-soaap fetch-soaap.o 
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 information.
fetch-soaap.o: In function `DigestCalcHA1':
fetch.soaap:(.text+0xbcfc): undefined reference to `MD5Init'
fetch.soaap:(.text+0xbd19): undefined reference to `MD5Update'

wonglkd

$ ../soaap/Build/Release/bin/soaap -o fetch.soaap fetch_c.ll --soaap-list-sandboxed-funcs --soaap-infer-fp-targets --soaap-dump-rpc-graph --soaap-sandbox-platform=capsicum --soaap-emulate-performance
* Running Soaap Pass
* Processing command-line options
* Finding class hierarchy (if there is one)
* Finding sandboxes
* Building basic callgraph

WARNING: Function "fprintf" is unresolved. Its behaviour will not be analysed.
WARNING: Function "socketpair" is unresolved. Its behaviour will not be analysed.
WARNING: Function "perror" is unresolved. Its behaviour will not be analysed.

wonglkd

Keybase proof

I hereby claim:

• I am wonglkd on github.
• I am wonglkd (https://keybase.io/wonglkd) on keybase.
• I have a public key whose fingerprint is 7DCA 8B5A DE0B F45B 4CC5 B013 B8FE 1C9E 26B0 A6F3

To claim this, I am signing this object:

wonglkd

# on server
# tips: -u for UDP,
nc -l <port>
# on client
nc <ip> <port>

wonglkd

        # Implement REINFORCE rule by using crossentropy
        # + (reward - baseline) as sample_weight.
        self.model.compile(loss='categorical_crossentropy', optimizer=Adam(lr=self.lr))

        # Adapted from model.compile: redo the computation of loss.
        loss_functions = [objectives.get('categorical_crossentropy')]

        total_loss = None

        for i in range(len(self.model.outputs)):

wonglkd

• Git Bash + MinGW: https://nuwen.net/files/mingw/mingw-14.1.exe from https://nuwen.net/mingw.html

Python

• Sublime Text
• conda
  • https://conda.io/miniconda.html

Things to check

• Make sure MinGW is in your path
• Checking your path: Windows+Pause (or right-click Computer -> Properties) -> Environment variables

wonglkd

import pdb, sys, traceback
def info(type, value, tb):
    traceback.print_exception(type, value, tb)
    pdb.pm()
sys.excepthook = info

wonglkd

import pdb, sys, traceback
def info(type, value, tb):
    traceback.print_exception(type, value, tb)
    pdb.pm()
sys.excepthook = info

wonglkd

scikit-learn==0.18.1
pandas==0.20.1
Keras>=2.0.4
numpy>=1.12.1
scipy>=0.19.0
tensorflow>=1.1.0
spacy>=1.8.2
# sacred==0.7.0
git+https://github.com/IDSIA/sacred.git@5a823657d824103cbbe09177a3aaf706c557ce55#egg=sacred
pymongo>=3.4.0

wonglkd

#!/bin/bash
# usage: bash ../utils/spell-check.sh
# setup:
# brew install hunspell
# wget https://raw.githubusercontent.com/wooorm/dictionaries/master/dictionaries/en-US/index.dic -O ~/Library/Spelling/en_US.dic
# wget https://raw.githubusercontent.com/wooorm/dictionaries/master/dictionaries/en-US/index.aff -O ~/Library/Spelling/en_US.aff
#
hunspell -l -t *.tex | sort | uniq > ../utils/misspelled.tmp
sort ../utils/dictionary -o ../utils/dictionary
comm -23 ../utils/misspelled.tmp ../utils/dictionary