TO TEST: App - register functions in serverledge

Author: matnar

examples/sentiment-analysis/scripts/create-functions.sh

@@ -0,0 +1,66 @@
+#!/bin/bash
+bin/serverledge-cli create -function sa_retrieve \ 
+    -memory 256 \
+    -runtime custom \ 
+    -custom_image sa-sentiment-analysis-retrieve \
+    --input "data_url:Text" \
+    --input "local_dir:Text" \
+    --input "object_name:Text" \
+    --output "status:Text" \ 
+    --output "local_download:Bool" \ 
+    --output "uploaded:Bool" \
+    --output "object_name:Text"
+
+
+bin/serverledge-cli create -function sa_extract \ 
+    -memory 256 \
+    -runtime custom \ 
+    -custom_image sa-sentiment-analysis-extract \
+    --input "tgz_input_object_name:Text" \
+    --input "subset:Float" \
+    --input "local_dataset_file:Text" \
+    --input "local_output_dir:Text" \
+    --input "output_train_object_name:Text" \
+    --input "output_test_object_name:Text" \
+    --output "status:Text" \
+    --output "train_object_name:Text" \
+    --output "test_object_name:Text" \
+    --output "running_time:Float"
+
+
+
+bin/serverledge-cli create -function sa_train \ 
+    -memory 256 \
+    -runtime custom \ 
+    -custom_image sa-sentiment-analysis-train \
+    --input "subset:Float" \
+    --input "max_features:Int" \
+    --input "train_object_data:Text" \
+    --input "local_train_file:Text" \
+    --input "local_model_file:Text" \
+    --input "local_vectorizer_file:Text" \
+    --input "output_model_object:Text" \
+    --input "output_vectorizer_object:Text" 
+    --output "status:Text" \
+    --output "model_object_name:Text" \
+    --output "vectorizer_object_name:Text" \
+    --output "running_time:Float"
+
+
+
+bin/serverledge-cli create -function sa_evaluate \ 
+    -memory 256 \
+    -runtime custom \ 
+    -custom_image sa-sentiment-analysis-evaluate \
+    --input "test_object_data:Text" \
+    --input "local_test_file:Text" \
+    --input "subset:Float" \
+    --input "local_model_file:Text" \
+    --input "local_vectorizer_file:Text" \
+    --input "input_model_object:Text" \
+    --input "input_vectorizer_object:Text" \
+    --output "status:Text" \
+    --output "accuracy:Float" \
+    --output "running_time:Float"
+
+## TODO: How to support environment variables?