Initial commit.
authorFrancois Fleuret <francois@fleuret.org>
Wed, 14 Jun 2017 16:30:45 +0000 (18:30 +0200)
committerFrancois Fleuret <francois@fleuret.org>
Wed, 14 Jun 2017 16:30:45 +0000 (18:30 +0200)
README.md [new file with mode: 0644]

diff --git a/README.md b/README.md
new file mode 100644 (file)
index 0000000..44e9e1a
--- /dev/null
+++ b/README.md
@@ -0,0 +1,29 @@
+# Introduction #
+
+This is the port of the Synthetic Visual Reasoning Test to the pytorch
+framework.
+
+The main function is
+
+```
+torch.ByteTensor generate_vignettes(int problem_number, torch.LongTensor labels)
+```
+
+where
+
+ * `problem_number` indicates which of the 23 problem to use
+ * `labels` indicates the boolean labels of the vignettes to generate
+
+The returned ByteTensor has three dimensions:
+
+ * Vignette index
+ * Pixel row
+ * Pixel col
+
+# Installation and test #
+
+```
+make -j -k
+./built.py
+./test-svrt.py
+```