add optimizer

.woodpecker.yml

@@ -1,8 +1,16 @@
+matrix:
+  EPOCHS:
+    - 20
+    - 30
+  OPTIMIZER: 
+    - adam
+    - SGD
+
 steps:
   "train":
     image: nvcr.io/nvidia/tensorflow:23.10-tf2-py3
     commands:
-      - echo "starting python script   sd "
+      - echo "starting python script"
       - python run.py 
   "compress and upload":
     image: alpine:3

README.md

@@ -1,4 +0,0 @@
-# Long Running Trainings
-
-This repository contains an example pipeline for long running training tasks. 
-Detailed information can be found at the official [Sandbox Documentation](https://docs.sandbox.iuk.hdm-stuttgart.de/sandbox/training.html).

run.py

@@ -8,6 +8,10 @@ import os
 from keras import backend as K
 K.clear_session()
 
+EPOCHS = int(os.getenv("EPOCHS", default = 10))
+OPTIMIZER = os.getenv("OPTIMIZER", default = "SGD")
+
+
 gpus = tf.config.experimental.list_physical_devices('GPU')
 # if gpus:
 #   try:
@@ -17,7 +21,7 @@ gpus = tf.config.experimental.list_physical_devices('GPU')
 #     os.exit(1)
 
 print(tf.config.experimental.list_physical_devices())
-print("test")
+
 print(tf.__version__)
 
 print(tf.test.is_built_with_cuda())
@@ -68,10 +72,10 @@ with tf.device('/GPU:0'):
             keras.layers.Dense(10, activation='sigmoid')    
         ])
   # g
-    model.compile(optimizer='SGD',
+    model.compile(optimizer=OPTIMIZER,
                   loss='categorical_crossentropy',
                   metrics=['accuracy'])
-    model.fit(X_train_scaled, y_train_categorical, epochs=25)
+    model.fit(X_train_scaled, y_train_categorical, epochs=EPOCHS)
     model.save('mymodel.keras')
     
     print("finished training")