Last active
October 26, 2016 14:24
-
-
Save vabh/3db7580172fb9253ff23232944c180fb to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| local sys = require 'sys' | |
| local image = require 'image' | |
| local paths = require 'paths' | |
| local M = {} | |
| local function loadAndConvertImages(imageList, baseDir, numExamples) | |
| local data = torch.Tensor(numExamples,3,256,256) | |
| local labels = torch.Tensor(numExamples) | |
| local image_dimensions | |
| idx = 0 | |
| local f = io.open(imageList) | |
| assert(f ~= nil, ' File does not exist: ' .. imageList) | |
| while true do | |
| print(idx) | |
| idx = idx + 1 | |
| local line = f:read('*line') | |
| if not line then break end | |
| -- split line into filename and label | |
| local image_data = {} | |
| for word in line:gmatch("%S+") do | |
| table.insert(image_data, word) | |
| end | |
| local image_file = paths.concat(baseDir, image_data[1]) | |
| -- add 1 to label to have non-zero labels for each example | |
| local label = tonumber(image_data[2]) + 1 | |
| local ok, input_image = pcall(function() | |
| return image.load(image_file, 3, 'byte') | |
| end) | |
| if ok then | |
| data[idx] = torch.Tensor(3, 256, 256):copy(input_image) | |
| labels[idx] = label | |
| else | |
| print('Cannot load: ' .. image_file) | |
| end | |
| end | |
| f:close() | |
| return { | |
| data = data, | |
| labels = torch.Tensor(labels):float() | |
| } | |
| end | |
| function M.exec(opt, cacheFile) | |
| -- file containing path to image and associated label | |
| local trainList = opt.trainList | |
| local valList = opt.valList | |
| local trainData = loadAndConvertImages(trainList, opt.trainBaseDir, NUM_TRAIN_SAMPLES) | |
| local testData = loadAndConvertImages(valList, opt.valBaseDir, NUM_TEST_SAMPLES) | |
| print("-- Saving dataset to " .. cacheFile) | |
| torch.save(cacheFile, { | |
| train = trainData, | |
| val = testData, | |
| }) | |
| end | |
| return M |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment