Ashish Garg created SYSTEMML-2515:
-------------------------------------
Summary: Python 3.5 Keras2DML not working
Key: SYSTEMML-2515
URL: https://issues.apache.org/jira/browse/SYSTEMML-2515
Project: SystemML
Issue Type: Bug
Reporter: Ashish Garg
Using Python notebook with Python 3.5 snapshot systemml 1-3 Snapshot
input_shape = (1,28,28) if K.image_data_format() == 'channels_first' else
(28,28, 1)
keras_model = Sequential()
keras_model.add(Conv2D(32, kernel_size=(5, 5), activation='relu',
input_shape=input_shape, padding='same'))
keras_model.add(MaxPooling2D(pool_size=(2, 2)))
keras_model.add(Conv2D(64, (5, 5), activation='relu', padding='same'))
keras_model.add(MaxPooling2D(pool_size=(2, 2)))
keras_model.add(Flatten())
keras_model.add(Dense(512, activation='relu'))
keras_model.add(Dropout(0.5))
keras_model.add(Dense(10, activation='softmax'))
keras_model.summary()
keras_model.compile(optimizer='adam', loss='mse', metrics=['accuracy'])
from pyspark.sql import SparkSession
from pyspark.sql.functions import *
spark = SparkSession.builder.enableHiveSupport().getOrCreate()
keras.backend.set_image_data_format("channels_first")
sysml_model.summary()
generate following error
Py4JJavaError: An error occurred while calling
None.org.apache.sysml.api.dl.Caffe2DML. :
com.google.protobuf.TextFormat$ParseException: 22:10: Couldn't parse integer:
For input string: "2.5" at
com.google.protobuf.TextFormat$Tokenizer.parseException(TextFormat.java:912) at
com.google.protobuf.TextFormat$Tokenizer.integerParseException(TextFormat.java:933)
at com.google.protobuf.TextFormat$Tokenizer.consumeUInt32(TextFormat.java:753)
at com.google.protobuf.TextFormat.mergeField(TextFormat.java:1179) at
com.google.protobuf.TextFormat.mergeField(TextFormat.java:1156) at
com.google.protobuf.TextFormat.mergeField(TextFormat.java:1156) at
com.google.protobuf.TextFormat.merge(TextFormat.java:1062) at
com.google.protobuf.TextFormat.merge(TextFormat.java:1028) at
com.google.protobuf.TextFormat.merge(TextFormat.java:998) at
org.apache.sysml.api.dl.Utils$.readCaffeNet(Utils.scala:148) at
org.apache.sysml.api.dl.CaffeNetwork.<init>(CaffeNetwork.scala:56) at
org.apache.sysml.api.dl.Caffe2DML.<init>(Caffe2DML.scala:196) at
sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
at
sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
at java.lang.reflect.Constructor.newInstance(Constructor.java:423) at
py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:247) at
py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:357)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
