From: Francois Fleuret <fleuret@elk.fleuret.org>
Date: Wed, 19 Oct 2016 15:59:59 +0000 (+0200)
Subject: The validation error was computed on the training data. That's embarassing.
X-Git-Url: https://www.fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=dyncnn.git;a=commitdiff_plain;h=4cab0b04a02e270f4e6bce11a763ddfe0a2ad2ae

The validation error was computed on the training data. That's embarassing.
---

diff --git a/dyncnn.lua b/dyncnn.lua
index 7cc5318..839431a 100755
--- a/dyncnn.lua
+++ b/dyncnn.lua
@@ -642,7 +642,7 @@ function trainModel(model,
          local startTime = sys.clock()
 
          for b = 1, validationData.nbSamples, batchSize do
-            fillBatch(trainData, b, batchSize, batch)
+            fillBatch(validationData, b, batchSize, batch)
             local output = model:forward(batch.input)
             accLoss = accLoss + l2criterion:forward(output, batch.target)
             nbBatches = nbBatches + 1