Training code for the networks described in "How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)" paper.
Hi guys! I am trying to train a face alignment model. So firstly I download the dataset(300W-LP) and landmarks(.t7) from the webpage. Downloaded landmarks(.t7) is then replaced with the landmarks(.mat) in the dataset. With the dataset, I place it into the face-alignment-training/dataset/ that I had clone from the github. I also changed the dataloader.lua script line15 from
[line15] for f in paths.files(base_dir..dirs[i],'.mat') do >>> for f in paths.files(base_dir..dirs[i],'.t7') do
With that I run
th main.lua
and it give the error shown below.
~/face-alignment-training$ th main.lua
=> Creating model from file: models/fan.lua
=> Model size: 23820176
=> Building dataset...
=> Dataset built. 57559 images were found.
Using lr_rate: 0.000250
=> Training epoch # 1
~/torch/install/bin/luajit: ...~/torch/install/share/lua/5.1/threads/threads.lua:183: [thread 1 callback] ./dataset-images.lua:16: attempt to index global 'opt' (a nil value)
stack traceback:
./dataset-images.lua:16: in function 'generateSampleFace'
./dataset-images.lua:34: in function 'get'
./dataloader.lua:86: in function <./dataloader.lua:80>
[C]: in function 'xpcall'
...~/torch/install/share/lua/5.1/threads/threads.lua:234: in function 'callback'
...e/~/torch/install/share/lua/5.1/threads/queue.lua:65: in function <...e/~/torch/install/share/lua/5.1/threads/queue.lua:41>
[C]: in function 'pcall'
...e/~/torch/install/share/lua/5.1/threads/queue.lua:40: in function 'dojob'
[string " local Queue = require 'threads.queue'..."]:15: in main chunk
stack traceback:
[C]: in function 'error'
...~/torch/install/share/lua/5.1/threads/threads.lua:183: in function 'dojob'
./dataloader.lua:122: in function '(for generator)'
./train.lua:49: in function 'train'
main.lua:40: in main chunk
[C]: in function 'dofile'
...~/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00405d50
Hi guys! I am trying to train a face alignment model. So firstly I download the dataset(300W-LP) and landmarks(.t7) from the webpage. Downloaded landmarks(.t7) is then replaced with the landmarks(.mat) in the dataset. With the dataset, I place it into the face-alignment-training/dataset/ that I had clone from the github. I also changed the dataloader.lua script line15 from [line15] for f in paths.files(base_dir..dirs[i],'.mat') do >>> for f in paths.files(base_dir..dirs[i],'.t7') do With that I run th main.lua and it give the error shown below.
~/face-alignment-training$ th main.lua => Creating model from file: models/fan.lua => Model size: 23820176
=> Building dataset...
=> Dataset built. 57559 images were found.
Using lr_rate: 0.000250 => Training epoch # 1
~/torch/install/bin/luajit: ...~/torch/install/share/lua/5.1/threads/threads.lua:183: [thread 1 callback] ./dataset-images.lua:16: attempt to index global 'opt' (a nil value) stack traceback: ./dataset-images.lua:16: in function 'generateSampleFace' ./dataset-images.lua:34: in function 'get' ./dataloader.lua:86: in function <./dataloader.lua:80> [C]: in function 'xpcall' ...~/torch/install/share/lua/5.1/threads/threads.lua:234: in function 'callback' ...e/~/torch/install/share/lua/5.1/threads/queue.lua:65: in function <...e/~/torch/install/share/lua/5.1/threads/queue.lua:41> [C]: in function 'pcall' ...e/~/torch/install/share/lua/5.1/threads/queue.lua:40: in function 'dojob' [string " local Queue = require 'threads.queue'..."]:15: in main chunk stack traceback: [C]: in function 'error' ...~/torch/install/share/lua/5.1/threads/threads.lua:183: in function 'dojob' ./dataloader.lua:122: in function '(for generator)' ./train.lua:49: in function 'train' main.lua:40: in main chunk [C]: in function 'dofile' ...~/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk [C]: at 0x00405d50
Thanks a lot for helping me :D