| def load_ascii_bytes(model): | |
| graph_def = tf.GraphDef() | |
| graph_def.ParseFromString(model) | |
| with tf.Graph().as_default() as graph: | |
| tf.import_graph_def( | |
| graph_def, | |
| input_map=None, | |
| return_elements=None, | |
| name="", |
| # This file is useful for reading the contents of the ops generated by ruby. | |
| # You can read any graph defination in pb/pbtxt format generated by ruby | |
| # or by python and then convert it back and forth from human readable to binary format. | |
| import tensorflow as tf | |
| from google.protobuf import text_format | |
| from tensorflow.python.platform import gfile | |
| def pbtxt_to_graphdef(filename): | |
| with open(filename, 'r') as f: |
| import os | |
| os.environ["PYTHONPATH"]="/Users/vtangirala/spark-2.4.0/python/lib/py4j-0.10.7-src.zip:"+os.environ["PYTHONPATH"] | |
| os.environ["PYTHONPATH"] | |
| from pyspark import SparkConf | |
| from pyspark import SparkContext | |
| conf = SparkConf() | |
| conf.setMaster("spark://vtangirala-mac.adaptiveplanning.com:7077").setAppName("venu") | |
| conf.set("spark.cores.max", "1") |
Classic Networks
LeNet-5: LeCun et al, "Gradient-Based Learning Applied to Document Recognition": http://yann.lecun.com/exdb/publis/pdf/lecun-01a.pdf
AlexNet: Krizhevsky et al, "ImageNet Classification with Deep Convolutional Neural Networks": https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf
VGG-16: Simonyan et al, "Very Deep Convolutional Networks for Large-Scale Image Recognition": https://arxiv.org/pdf/1409.1556.pdf
Resnets
| def minArea(x, y, k): | |
| # def eucledian_dist(a,b): | |
| # return sqrt(sum( (a - b)**2 for a, b in zip(a, b))) | |
| # d = eucledian_dist(x,y) | |
| def avg(l): | |
| return sum (l) / float (len (l)) | |
| def stats(x, y): | |
| x_min = min (x) | |
| x_max = max (x) |
| 1. Create empty file: | |
| This file will contain virtual memory contents so make file big enough for your needs. This one will create 1Gb file which means +1Gb swap space for your system: | |
| dd if=/dev/zero of=/media/fasthdd/swapfile.img bs=1024 count=1M | |
| If you want to make 3Gb file then change count value to count=3M. See man dd for more information. | |
| 2. Bake swap file: | |
| Following command is going to make "swap filesystem" inside your fresh swap file. | |
| mkswap /media/fasthdd/swapfile.img |
| import tensorflow as tf | |
| from keras import backend as K | |
| num_cores = 4 | |
| if GPU: | |
| num_GPU = 1 | |
| num_CPU = 1 | |
| if CPU: | |
| num_CPU = 1 |
| import requests | |
| import ast | |
| url="https://gist.githubusercontent.com/yrevar/942d3a0ac09ec9e5eb3a/raw/c2c91c8e767d04621020c30ed31192724b863041/imagenet1000_clsid_to_human.txt" | |
| data=requests.get(url).text | |
| class_num_label = ast.literal_eval(data) |
| SELECT | |
| table_schema as `Database`, | |
| table_name AS `Table`, | |
| round(((data_length + index_length) / 1024 / 1024), 2) `Size in MB` | |
| FROM information_schema.TABLES | |
| ORDER BY (data_length + index_length) DESC; |
| find -type d -readable -exec sh -c 'printf "%s " "$1"; ls -1UA "$1" | wc -l' sh {} ';' |