Replace boto with boto3.

targeted-marketing-python/build_model.py

@@ -20,7 +20,7 @@
     python build_model.py ["Optional name for created objects"]
 """
 import base64
-import boto
+import boto3
 import json
 import os
 import sys
@@ -32,7 +32,7 @@
 def build_model(data_s3_url, schema_fn, recipe_fn, name, train_percent=70):
     """Creates all the objects needed to build an ML Model & evaluate its quality.
     """
-    ml = boto.connect_machinelearning()
+    ml = boto3.client('machinelearning')
     (train_ds_id, test_ds_id) = create_data_sources(ml, data_s3_url, schema_fn,
                                                     train_percent, name)
     ml_model_id = create_model(ml, train_ds_id, recipe_fn, name)
@@ -59,10 +59,10 @@ def create_data_sources(ml, data_s3_url, schema_fn, train_percent, name):
         "DataSchema": open(schema_fn).read(),
     }
     ml.create_data_source_from_s3(
-        data_source_id=train_ds_id,
-        data_spec=spec,
-        data_source_name=name + " - training split",
-        compute_statistics=True
+        DataSourceId=train_ds_id,
+        DataSpec=spec,
+        DataSourceName=name + " - training split",
+        ComputeStatistics=True
     )
     print("Created training data set %s" % train_ds_id)
 
@@ -74,10 +74,10 @@ def create_data_sources(ml, data_s3_url, schema_fn, train_percent, name):
         }
     })
     ml.create_data_source_from_s3(
-        data_source_id=test_ds_id,
-        data_spec=spec,
-        data_source_name=name + " - testing split",
-        compute_statistics=True
+        DataSourceId=test_ds_id,
+        DataSpec=spec,
+        DataSourceName=name + " - testing split",
+        ComputeStatistics=True
     )
     print("Created test data set %s" % test_ds_id)
     return (train_ds_id, test_ds_id)
@@ -91,18 +91,18 @@ def create_model(ml, train_ds_id, recipe_fn, name):
     """
     model_id = 'ml-' + base64.b32encode(os.urandom(10))
     ml.create_ml_model(
-        ml_model_id=model_id,
-        ml_model_name=name + " model",
-        ml_model_type="BINARY",  # we're predicting True/False values
-        parameters={
+        MLModelId=model_id,
+        MLModelName=name + " model",
+        MLModelType="BINARY",  # we're predicting True/False values
+        Parameters={
             # Refer to the "Machine Learning Concepts" documentation
             # for guidelines on tuning your model
             "sgd.maxPasses": "100",
             "sgd.maxMLModelSizeInBytes": "104857600",  # 100 MiB
             "sgd.l2RegularizationAmount": "1e-4",
         },
-        recipe=open(recipe_fn).read(),
-        training_data_source_id=train_ds_id
+        Recipe=open(recipe_fn).read(),
+        TrainingDataSourceId=train_ds_id
     )
     print("Created ML Model %s" % model_id)
     return model_id
@@ -111,10 +111,10 @@ def create_model(ml, train_ds_id, recipe_fn, name):
 def create_evaluation(ml, model_id, test_ds_id, name):
     eval_id = 'ev-' + base64.b32encode(os.urandom(10))
     ml.create_evaluation(
-        evaluation_id=eval_id,
-        evaluation_name=name + " evaluation",
-        ml_model_id=model_id,
-        evaluation_data_source_id=test_ds_id
+        EvaluationId=eval_id,
+        EvaluationName=name + " evaluation",
+        MLModelId=model_id,
+        EvaluationDataSourceId=test_ds_id
     )
     print("Created Evaluation %s" % eval_id)
     return eval_id

targeted-marketing-python/use_model.py

@@ -24,7 +24,7 @@
     python use_model.py ml-12345678901 0.77 s3://your-bucket/prefix
 """
 import base64
-import boto
+import boto3
 import datetime
 import os
 import random
@@ -39,28 +39,28 @@
 def use_model(model_id, threshold, schema_fn, output_s3, data_s3url):
     """Creates all the objects needed to build an ML Model & evaluate its quality.
     """
-    ml = boto.connect_machinelearning()
+    ml = boto3.client('machinelearning') 
 
     poll_until_completed(ml, model_id)  # Can't use it until it's COMPLETED
-    ml.update_ml_model(model_id, score_threshold=threshold)
+    ml.update_ml_model(MLModelId=model_id, ScoreThreshold=threshold)
     print("Set score threshold for %s to %.2f" % (model_id, threshold))
 
     bp_id = 'bp-' + base64.b32encode(os.urandom(10))
     ds_id = create_data_source_for_scoring(ml, data_s3url, schema_fn)
     ml.create_batch_prediction(
-        batch_prediction_id=bp_id,
-        batch_prediction_name="Batch Prediction for marketing sample",
-        ml_model_id=model_id,
-        batch_prediction_data_source_id=ds_id,
-        output_uri=output_s3
+        BatchPredictionId=bp_id,
+        BatchPredictionName="Batch Prediction for marketing sample",
+        MLModelId=model_id,
+        BatchPredictionDataSourceId=ds_id,
+        OutputUri=output_s3
     )
     print("Created Batch Prediction %s" % bp_id)
 
 
 def poll_until_completed(ml, model_id):
     delay = 2
     while True:
-        model = ml.get_ml_model(model_id)
+        model = ml.get_ml_model(MLModelId=model_id)
         status = model['Status']
         message = model.get('Message', '')
         now = str(datetime.datetime.now().time())
@@ -76,13 +76,13 @@ def poll_until_completed(ml, model_id):
 def create_data_source_for_scoring(ml, data_s3url, schema_fn):
     ds_id = 'ds-' + base64.b32encode(os.urandom(10))
     ml.create_data_source_from_s3(
-        ds_id,
-        data_source_name="DS for Batch Prediction %s" % data_s3url,
-        data_spec={
+        DataSourceId=ds_id,
+        DataSourceName="DS for Batch Prediction %s" % data_s3url,
+        DataSpec={
             "DataLocationS3": data_s3url,
             "DataSchema": open(schema_fn).read(),
         },
-        compute_statistics=False
+        ComputeStatistics=False
     )
     print("Created Datasource %s for batch prediction" % ds_id)
     return ds_id