Skip to content

Instantly share code, notes, and snippets.

* Converts images to GGB (grayscale)
* Creates subsets for training and validation
* Adds columns to indicate training or validation (useful for analysis of the deployed model)
* Adds rotated images to the dataset
* Creates comma-separated CSV file
* Creates zip archive
import pandas as pd
hussius /
Created June 14, 2018 12:17
Preprocess yeast DNA csv file from Genome Research paper
from pathlib import Path
import os
import sys
from fire import Fire
import numpy as np
import pandas as pd
from tqdm import tqdm
hussius /
Created April 26, 2018 11:02 — forked from hannes-brt/
Function to decode a COSSMO training example in tfrecord format
def read_single_cossmo_example(serialized_example, n_tissues=1, coord_sys='rna1'):
"""Decode a single COSSMO example
coord_sys must be one of 'rna1' or 'dna0', if 'dna0' then an extra 'strand' field
must exist in the tfrecord and is extracted.
assert coord_sys in ['dna0', 'rna1']
context_features = {
1. Install appropriate version of the Tensorflow (Python) framework from
In my case (Mac OS X 10.11), I did:
- Get the .whl file (this is more likely to work than a direct pip install)
- Install using non-Anaconda pip
/usr/local/bin/pip3 install tensorflow-0.11.0-py3-none-any.whl_
x_data <- runif(100, min=0, max=1)
y_data <- x_data * 0.1 + 0.3 + rnorm(n, mean=0, sd=0.025)
W <- tf$Variable(tf$random_uniform(shape(1L), -1.0, 1.0))
b <- tf$Variable(tf$zeros(shape(1L)))
y <- W * x_data + b
hussius /
Created November 17, 2016 09:39
Sum transcript TPMs by gene using the FASTA file used for a Kallisto index
import sys
import gzip
if len(sys.argv)<3:
sys.exit("python <cDNA FASTA file> <TPM table>")
ensg = {}
mapf =[1])
ctr = 0
hussius / merge_kallisto_TPM.R
Last active November 17, 2016 09:38
R script for merging Kallisto TPMs from output directories below path given as command-line argument
args = commandArgs(trailingOnly=TRUE)
merge_two <- function(x,y){
if ("tpm" %in% colnames(x)){
x_ <- x[,c(1,5)]
import sys
f = open(sys.argv[1])
wts = [] # This will be a list of list of lists (=list of matrices) with weights
bias_in = [] # List of lists of bias values in the input part of each sublayer
bias_out = [] # List of lists of bias values in the output part of each sublayer
prefs = [] # Prefixes for file names
state = None # state can be None, 'weight', 'bias_in' or 'bias_out'
import sys
for line in open(sys.argv[1]):
if 'layer' in line:
fname = '_'.join(line.strip().split()) + '_activities.txt'
outf = open(fname,'w')
hussius /
Last active June 28, 2019 17:12
Toy example of single-layer autoencoder in TensorFlow
import tensorflow as tf
import numpy as np
import math
#import pandas as pd
#import sys
input = np.array([[2.0, 1.0, 1.0, 2.0],
[-2.0, 1.0, -1.0, 2.0],
[0.0, 1.0, 0.0, 2.0],
[0.0, -1.0, 0.0, -2.0],