openface/batch-represent/batch-represent.lua

78 lines
2.0 KiB
Lua
Raw Normal View History

2015-09-24 06:19:30 +08:00
local ffi = require 'ffi'
local batchNumber, nImgs = 0
torch.setdefaulttensortype('torch.FloatTensor')
2015-09-24 06:19:30 +08:00
function batchRepresent()
local loadSize = {3, opt.imgDim, opt.imgDim}
2016-01-11 21:25:04 +08:00
print(opt.data)
local cacheFile = paths.concat(opt.data, 'cache.t7')
print('cache lotation: ', cacheFile)
2016-01-11 21:47:34 +08:00
local dumpLoader
2016-01-11 21:25:04 +08:00
if paths.filep(cacheFile) then
print('Loading metadata from cache.')
print('If your dataset has changed, delete the cache file.')
dumpLoader = torch.load(cacheFile)
else
print('Creating metadata for cache.')
dumpLoader = dataLoader{
paths = {opt.data},
loadSize = loadSize,
sampleSize = loadSize,
split = 0,
verbose = true
}
torch.save(cacheFile, dumpLoader)
end
collectgarbage()
2015-09-24 06:19:30 +08:00
nImgs = dumpLoader:sizeTest()
print('nImgs: ', nImgs)
assert(nImgs > 0, "Failed to get nImgs")
batchNumber = 0
for i=1,math.ceil(nImgs/opt.batchSize) do
local indexStart = (i-1) * opt.batchSize + 1
local indexEnd = math.min(nImgs, indexStart + opt.batchSize - 1)
local batchSz = indexEnd-indexStart+1
2015-09-24 06:19:30 +08:00
local inputs, labels = dumpLoader:get(indexStart, indexEnd)
local paths = {}
2015-12-27 22:09:14 +08:00
for j=indexStart,indexEnd do
2016-01-13 06:29:34 +08:00
table.insert(paths,
ffi.string(dumpLoader.imagePath[dumpLoader.testIndices[j]]:data()))
2015-09-24 06:19:30 +08:00
end
repBatch(paths, inputs, labels, batchSz)
2015-09-24 06:19:30 +08:00
if i % 5 == 0 then
collectgarbage()
end
end
if opt.cuda then
cutorch.synchronize()
end
end
function repBatch(paths, inputs, labels, batchSz)
batchNumber = batchNumber + batchSz
2015-09-24 06:19:30 +08:00
if opt.cuda then
inputs = inputs:cuda()
end
local embeddings = model:forward(inputs):float()
if opt.cuda then
cutorch.synchronize()
end
2015-09-24 06:19:30 +08:00
if batchSz == 1 then
embeddings = embeddings:reshape(1, embeddings:size(1))
end
for i=1,batchSz do
2015-09-24 06:19:30 +08:00
labelsCSV:write({labels[i], paths[i]})
repsCSV:write(embeddings[i]:totable())
end
print(('Represent: %d/%d'):format(batchNumber, nImgs))
end