Commit 041b901b authored by Ross Girshick's avatar Ross Girshick
Browse files

make cachedir setup same with a local override file

parent 9a4ac2d0
...@@ -7,3 +7,5 @@ data/caffe_nets ...@@ -7,3 +7,5 @@ data/caffe_nets
data/rcnn_models data/rcnn_models
data/selective_search_data data/selective_search_data
data/window_files data/window_files
rcnn_config_local.m
cachedir/*
...@@ -76,12 +76,6 @@ Let's use PASCAL VOC 2007 as an example. The basic pipeline is: ...@@ -76,12 +76,6 @@ Let's use PASCAL VOC 2007 as an example. The basic pipeline is:
You'll need about 200GB of disk space free for the feature cache (which is stored in `rcnn/feat_cache` by default; symlink `rcnn/feat_cache` elsewhere if needed). **It's best if the feature cache is on a fast, local disk.** Before running the pipeline, we first need to install the PASCAL VOC 2007 dataset. You'll need about 200GB of disk space free for the feature cache (which is stored in `rcnn/feat_cache` by default; symlink `rcnn/feat_cache` elsewhere if needed). **It's best if the feature cache is on a fast, local disk.** Before running the pipeline, we first need to install the PASCAL VOC 2007 dataset.
#### Model cache directory
The training and testing procedures save models and results under `rcnn/cachedir` by default. I make `rcnn/cachedir` a symlink to where I want this data stored.
From inside the `rcnn` directory: `$ ln -sf /path/to/your/cache/directory cachedir`.
#### Installing PASCAL VOC 2007 #### Installing PASCAL VOC 2007
1. Download the training and validation data [here](http://pascallin.ecs.soton.ac.uk/challenges/VOC/voc2007/VOCtrainval_06-Nov-2007.tar). 1. Download the training and validation data [here](http://pascallin.ecs.soton.ac.uk/challenges/VOC/voc2007/VOCtrainval_06-Nov-2007.tar).
...@@ -113,10 +107,15 @@ I use a symlink to hook the R-CNN codebase to the PASCAL VOC dataset: ...@@ -113,10 +107,15 @@ I use a symlink to hook the R-CNN codebase to the PASCAL VOC dataset:
#### Training R-CNN models and testing #### Training R-CNN models and testing
Now to run the training and testing code, use the following experiments script:
<pre> <pre>
>> test_results = rcnn_exp_train_and_test() >> test_results = rcnn_exp_train_and_test()
</pre> </pre>
**Note:** The training and testing procedures save models and results under `rcnn/cachedir` by default. You can customize this by creating a local config file named `rcnn_config_local.m` and defining the experiment directory variable `EXP_DIR`. Look at `rcnn_config_local.example.m` for an example.
### Training an R-CNN detector on another dataset ### Training an R-CNN detector on another dataset
It should be easy to train an R-CNN detector using another detection dataset as long as that dataset has *complete* bounding box annotations (i.e., all instances of all classes are labeled). It should be easy to train an R-CNN detector using another detection dataset as long as that dataset has *complete* bounding box annotations (i.e., all instances of all classes are labeled).
......
/work5/rbg
\ No newline at end of file
This is where saved models and test outputs live.
...@@ -30,24 +30,24 @@ function conf = rcnn_config(varargin) ...@@ -30,24 +30,24 @@ function conf = rcnn_config(varargin)
% this file (or any portion of it) in your project. % this file (or any portion of it) in your project.
% --------------------------------------------------------- % ---------------------------------------------------------
% ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
% Defaults config
% ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
%
% If you want to override any of these, create a script named
% rcnn_config_local.m and redefine these variables there.
% %
% ~~~~~~~~~~~~~~~~~~~~~~ BASIC SETUP ~~~~~~~~~~~~~~~~~~~~~~
% Please read the next few lines
% Experiments directory. The directory under which most outputs % Experiments directory. The directory under which most outputs
% generated by running this code will go. % generated by running this code will go.
EXP_DIR = './cachedir/rcnn/v1.1/fc7'; EXP_DIR = './cachedir';
% Set to false if you do not want to use a GPU. % Set to false if you do not want to use a GPU.
USE_GPU = true; USE_GPU = true;
% Load local overrides if rccn_config_local.m exists
% % See rcnn_config_local.example.m for an example
% You probably don't need to change configuration settings below this line. if exist('rcnn_config_local.m')
% rcnn_config_local;
end
% ~~~~~~~~~~~~~~~~~~~~~~ ADVANCED SETUP BELOW ~~~~~~~~~~~~~~~~~~~~~~ % ~~~~~~~~~~~~~~~~~~~~~~ ADVANCED SETUP BELOW ~~~~~~~~~~~~~~~~~~~~~~
......
% Set your own experiments directory
EXP_DIR = './cachedir/foo/bar';
% Do not use the GPU
USE_GPU = false;
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment