aakbar5

To visualize freezed Tensorflow model file use Tensorboard:
 - Download import_pb_to_tensorboard.py (https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/tools/import_pb_to_tensorboard.py)
 - Create a directory (/path/to/tmp/folder) where intermediate files will be created to visualize model in Tensorboard.
 - Execute
    - python import_pb_to_tensorboard.py --model_dir="/path/to/folder/tensorflow_model_frozen.pb" --log_dir="/path/to/tmp/folder"

        NOTE: Wrapped values to param # model_dir and param # log_dir in "". Otherwise it might not work.
    
    - tensorboard --logdir="/path/to/tmp/folder"

aakbar5

# Ref: https://www.gnu.org/software/make/manual/html_node/Setting.html#Setting

########
# Setting variables:
# ?= Set variable if it is empty
# := Expand value variable before assignment, 
#	NOTE: (1) The behaviour of ::= is same as :=
#         (2) Expansion of value variable will be only once at the of value assignment.
# = Simple variable assignment, value will evaluated/expanded on its usage
# += Append value to variable

aakbar5

GLOBAL_VAR=GOLBAL_VAR_VALUE     # This value will be set as it is global
export GLOBAL_VAR_EXPORT=GLOBAL_VAR_EXPORT_VALUE  # This value will be set as it is global

# @$(foreach var, $(filter GB_ENV_%, $(.VARIABLES)), $(eval export $(subst GB_ENV_,,$(var))=$($(var))))

all:
#	Following two vars remain unset as Makefile doesn't set var under phony target
	@LOCAL_VAR=LOCAL_VAR_VALUE
	@export LOCAL_VAR_EXPORT=LOCAL_VAR_EXPORT_VALUE

aakbar5

# ~/.bashrc: executed by bash(1) for non-login shells.
# see /usr/share/doc/bash/examples/startup-files (in the package bash-doc)
# for examples

# If not running interactively, don't do anything
case $- in
    *i*) ;;
      *) return;;
esac

aakbar5

# ~/.bashrc: executed by bash(1) for non-login shells.
# see /usr/share/doc/bash/examples/startup-files (in the package bash-doc)
# for examples

# If not running interactively, don't do anything
case $- in
    *i*) ;;
      *) return;;
esac

aakbar5

#
# Commands are tested on Ubuntu 14.04 
#

# Install the python3 as Ubuntu 14.04 comes with Python2
sudo apt-get update -y && \
sudo apt-get install python-pip python3 python3-pip

# Let's manage two versions

aakbar5

#
# Commands are tested on Ubuntu 14.04 
#

# Issue following command to verify which GCC version is installed
gcc --version
 --- OR --- 
g++ --version

gcc (Ubuntu 4.8.5-4ubuntu8~14.04.2) 4.8.5

aakbar5

# Incase you are interested to find out docker short, full id, hostname and ip address 
# of the running container, execute following commands from inside of your container

# To find out docker container id (short form)
DOCKER_ID=`cat /etc/hostname`
echo $DOCKER_ID

# To find out docker container host name using /etc/hosts
HOST_NAME=`grep $DOCKER_ID /etc/hosts | sed 's/\s/\n/g' | tail -1`
# - grep: search for the line having docker id

aakbar5

# To convert dts file into dtb 
<kernel-source>/scripts/dtc/dtc -I dts -O dtb <dts_file> -o <dtb_file>

# To convert dtb to dts
<kernel-source>/scripts/dtc/dtc -I dtb -O dts <dtb_file> -o <dts_file>

# Both of above commands are using dtc built with linux kernel source. 
# However you can also use dtc binary comes with ubuntu packages. For this you
# need to install following package:

aakbar5

In case you are having ldconfig error in Ubuntu 14.04.5 LTS related to Nvidia EGL use follow fixes

Error
---
/sbin/ldconfig.real: /usr/lib/nvidia-384/libEGL.so.1 is not a symbolic link
/sbin/ldconfig.real: /usr/lib32/nvidia-384/libEGL.so.1 is not a symbolic link

Fix
---
sudo mv /usr/lib/nvidia-384/libEGL.so.1 /usr/lib/nvidia-384/libEGL.so.1.org