Basic recommendation engine

examples/movie-ratings/app.yaml

@@ -0,0 +1,2 @@
+- kind: app
+  name: movie-ratings

examples/movie-ratings/implementations/models/basic_embedding.py

@@ -0,0 +1,27 @@
+import tensorflow as tf
+
+
+def create_estimator(run_config, model_config):
+    user_id_index = model_config["aggregates"]["user_id_index"]
+    movie_id_index = model_config["aggregates"]["movie_id_index"]
+
+    feature_columns = [
+        tf.feature_column.embedding_column(
+            tf.feature_column.categorical_column_with_identity(
+                "user_id_indexed", len(user_id_index)
+            ),
+            model_config["hparams"]["embedding_size"],
+        ),
+        tf.feature_column.embedding_column(
+            tf.feature_column.categorical_column_with_identity(
+                "movie_id_indexed", len(movie_id_index)
+            ),
+            model_config["hparams"]["embedding_size"],
+        ),
+    ]
+
+    return tf.estimator.DNNRegressor(
+        feature_columns=feature_columns,
+        hidden_units=model_config["hparams"]["hidden_units"],
+        config=run_config,
+    )

examples/movie-ratings/movies.json

@@ -0,0 +1,12 @@
+{
+  "samples": [
+    {
+      "user_id": "71",
+      "movie_id": "91529"
+    },
+    {
+      "user_id": "71",
+      "movie_id": "174055"
+    }
+  ]
+}

examples/movie-ratings/resources/apis.yaml

@@ -0,0 +1,5 @@
+- kind: api
+  name: ratings
+  model_name: basic_embedding
+  compute:
+    replicas: 1

examples/movie-ratings/resources/environments.yaml

@@ -0,0 +1,24 @@
+- kind: environment
+  name: dev
+  data:
+    type: csv
+    path: s3a://cortex-examples/movie-ratings.csv
+    csv_config:
+      header: true
+    schema: ['user_id','movie_id','rating','timestamp']
+
+- kind: raw_column
+  name: user_id
+  type: STRING_COLUMN
+
+- kind: raw_column
+  name: movie_id
+  type: STRING_COLUMN
+
+- kind: raw_column
+  name: rating
+  type: FLOAT_COLUMN
+
+- kind: raw_column
+  name: timestamp
+  type: INT_COLUMN

examples/movie-ratings/resources/models.yaml

@@ -0,0 +1,15 @@
+- kind: model
+  name: basic_embedding
+  type: regression
+  target_column: rating
+  feature_columns: [user_id_indexed, movie_id_indexed]
+  aggregates: [user_id_index, movie_id_index]
+  hparams:
+    embedding_size: 10
+    hidden_units: [128]
+  data_partition_ratio:
+    training: 0.8
+    evaluation: 0.2
+  training:
+    batch_size: 50
+    num_steps: 8000

examples/movie-ratings/resources/transformed_columns.yaml

@@ -0,0 +1,31 @@
+- kind: aggregate
+  name: user_id_index
+  aggregator: cortex.index_string
+  inputs:
+    columns:
+      col: user_id
+
+- kind: transformed_column
+  name: user_id_indexed
+  transformer: cortex.index_string
+  inputs:
+    columns:
+      text: user_id
+    args:
+      index: user_id_index
+
+- kind: aggregate
+  name: movie_id_index
+  aggregator: cortex.index_string
+  inputs:
+    columns:
+      col: movie_id
+
+- kind: transformed_column
+  name: movie_id_indexed
+  transformer: cortex.index_string
+  inputs:
+    columns:
+      text: movie_id
+    args:
+      index: movie_id_index