This is an automated email from the ASF dual-hosted git repository.

fmcquillan pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/madlib.git

commit 96a44244f6110c34e2ca3742b0b6f72593e3ada8
Author: Domino Valdano <[email protected]>
AuthorDate: Tue Jan 14 17:52:02 2020 -0800

    Decrease the learning rate for transfer learning test
    
    This helps smooth out the learning curve, making the test be more
    predictable... much less likely to fail due to a random fluctuation.
    
    Co-authored-by: Orhan Kislal <[email protected]>
---
 .../modules/deep_learning/test/madlib_keras_transfer_learning.sql_in  | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git 
a/src/ports/postgres/modules/deep_learning/test/madlib_keras_transfer_learning.sql_in
 
b/src/ports/postgres/modules/deep_learning/test/madlib_keras_transfer_learning.sql_in
index d17ea20..92f2277 100644
--- 
a/src/ports/postgres/modules/deep_learning/test/madlib_keras_transfer_learning.sql_in
+++ 
b/src/ports/postgres/modules/deep_learning/test/madlib_keras_transfer_learning.sql_in
@@ -290,8 +290,8 @@ SELECT load_model_selection_table(
     'mst_table',
     ARRAY[1,3],
     ARRAY[
-        
$$loss='categorical_crossentropy',optimizer='Adam(lr=0.01)',metrics=['accuracy']$$,
-        $$loss='categorical_crossentropy', 
optimizer='Adam(lr=0.001)',metrics=['accuracy']$$
+        
$$loss='categorical_crossentropy',optimizer='Adam(lr=0.00001)',metrics=['accuracy']$$,
+        $$loss='categorical_crossentropy', 
optimizer='Adam(lr=0.00002)',metrics=['accuracy']$$
     ],
     ARRAY[
         $$batch_size=5,epochs=1$$

Reply via email to