我正在使用一个名为NeuralTalk2的开源工具，它基本上需要几张图片并输出一个它认为描述图片的标题。那里有一个lua脚本，如下所示：
--[[
Same as DataLoader but only requires a folder of images.
Does not have an h5 dependency.
Only used at test time.
]]--
local utils = require 'misc.utils'
require 'lfs'
require 'image'
local DataLoaderRaw = torch.class('DataLoaderRaw')
function DataLoaderRaw:__init(opt)
local coco_json = utils.getopt(opt, 'coco_json', '')
-- load the json file which contains additional information about the dataset
print('DataLoaderRaw loading images from folder: ', opt.folder_path)
self.files = {}
self.ids = {}
if string.len(opt.coco_json) > 0 then
print('reading from ' .. opt.coco_json)
-- read in filenames from the coco-style json file
self.coco_annotation = utils.read_json(opt.coco_json)
for k,v in pairs(self.coco_annotation.images) do
local fullpath = path.join(opt.folder_path, v.file_name)
table.insert(self.files, fullpath)
table.insert(self.ids, v.id)
end
else
-- read in all the filenames from the folder
print('listing all images in directory ' .. opt.folder_path)
local function isImage(f)
local supportedExt = {'.jpg','.JPG','.jpeg','.JPEG','.png','.PNG','.ppm','.PPM'}
for _,ext in pairs(supportedExt) do
local _, end_idx = f:find(ext)
if end_idx and end_idx == f:len() then
return true
end
end
return false
end
local n = 1
for file in paths.files(opt.folder_path, isImage) do
local fullpath = path.join(opt.folder_path, file)
table.insert(self.files, fullpath)
table.insert(self.ids, tostring(n)) -- just order them sequentially
n=n+1
end
end
self.N = #self.files
print('DataLoaderRaw found ' .. self.N .. ' images')
self.iterator = 1
end
function DataLoaderRaw:resetIterator()
self.iterator = 1
end
--[[
Returns a batch of data:
- X (N,3,256,256) containing the images as uint8 ByteTensor
- info table of length N, containing additional information
The data is iterated linearly in order
--]]
function DataLoaderRaw:getBatch(opt)
local batch_size = utils.getopt(opt, 'batch_size', 5) -- how many images get returned at one time (to go through CNN)
-- pick an index of the datapoint to load next
local img_batch_raw = torch.ByteTensor(batch_size, 3, 256, 256)
local max_index = self.N
local wrapped = false
local infos = {}
for i=1,batch_size do
local ri = self.iterator
local ri_next = ri + 1 -- increment iterator
if ri_next > max_index then ri_next = 1; wrapped = true end -- wrap back around
self.iterator = ri_next
-- load the image
local img = image.load(self.files[ri], 3, 'byte')
img_batch_raw[i] = image.scale(img, 256, 256)
-- and record associated info as well
local info_struct = {}
info_struct.id = self.ids[ri]
info_struct.file_path = self.files[ri]
table.insert(infos, info_struct)
end
local data = {}
data.images = img_batch_raw
data.bounds = {it_pos_now = self.iterator, it_max = self.N, wrapped = wrapped}
data.infos = infos
return data
end
这段代码做了很多事情，但是请注意从第13行开始的__init函数：
function DataLoaderRaw:__init(opt)
local coco_json = utils.getopt(opt, 'coco_json', '')
-- load the json file which contains additional information about the dataset
print('DataLoaderRaw loading images from folder: ', opt.folder_path)
self.files = {}
self.ids = {}
if string.len(opt.coco_json) > 0 then
print('reading from ' .. opt.coco_json)
-- read in filenames from the coco-style json file
self.coco_annotation = utils.read_json(opt.coco_json)
for k,v in pairs(self.coco_annotation.images) do
local fullpath = path.join(opt.folder_path, v.file_name)
table.insert(self.files, fullpath)
table.insert(self.ids, v.id)
end
else
-- read in all the filenames from the folder
print('listing all images in directory ' .. opt.folder_path)
local function isImage(f)
local supportedExt = {'.jpg','.JPG','.jpeg','.JPEG','.png','.PNG','.ppm','.PPM'}
for _,ext in pairs(supportedExt) do
local _, end_idx = f:find(ext)
if end_idx and end_idx == f:len() then
return true
end
end
return false
end
local n = 1
for file in paths.files(opt.folder_path, isImage) do
local fullpath = path.join(opt.folder_path, file)
table.insert(self.files, fullpath)
table.insert(self.ids, tostring(n)) -- just order them sequentially
n=n+1
end
end
self.N = #self.files
print('DataLoaderRaw found ' .. self.N .. ' images')
self.iterator = 1
end
原始代码会随机将图片和标题输出到.hmtl文件中。因此，图片1将与图片10相邻，图片4将与图片38相邻（例如）。
我不能这样做，所以我编辑了它：
function DataLoaderRaw:__init(opt)
local coco_json = utils.getopt(opt, 'coco_json', '')
-- load the json file which contains additional information about the dataset
print('DataLoaderRaw loading images from folder: ', opt.folder_path)
self.files = {}
self.ids = {}
if string.len(opt.coco_json) > 0 then
print('reading from ' .. opt.coco_json)
-- read in filenames from the coco-style json file
self.coco_annotation = utils.read_json(opt.coco_json)
for k,v in pairs(self.coco_annotation.images) do
local fullpath = path.join(opt.folder_path, v.file_name)
table.insert(self.files, fullpath)
table.insert(self.ids, v.id)
end
else
-- read in all the filenames from the folder
print('listing all images in directory ' .. opt.folder_path)
local function isImage(f)
local supportedExt = {'.jpg','.JPG','.jpeg','.JPEG','.png','.PNG','.ppm','.PPM'}
for _,ext in pairs(supportedExt) do
local _, end_idx = f:find(ext)
if end_idx and end_idx == f:len() then
return true
end
end
return false
end
local n = 1
local m = 0
local counter = 1
while counter <= #paths.dir(opt.folder_path) - 2 do -- two directories "." and ".." returned by dir that are not images
local file_added = false
for file in paths.files(opt.folder_path, isImage) do
if tonumber(string.sub(file, 7, 10)) == n then -- insert files into self.files in order
if tonumber(string.sub(file, 12, 12)) == m then
local fullpath = path.join(opt.folder_path, file)
table.insert(self.files, fullpath)
table.insert(self.ids, tostring(n)) -- just order them sequentially
m=m+1
counter = counter + 1
file_added = true
break
end
end
end
if file_added == false then
n = n + 1
m = 0
end
end
print(self.files)
end
self.N = #self.files
print('DataLoaderRaw found ' .. self.N .. ' images')
self.iterator = 1
end
我写的代码应该改变它并使它成为图1,2,3,4,5 ... n彼此相邻。
然而，我陷入了while循环......它永远是循环的。我真的不知道为什么 - 这是我第一次在lua编码。希望......一个训练有素的专家会知道吗？
谢谢。