add notebook

Signed-off-by: cmuhao <[email protected]>

Author: HaoXuAI

.gitignore

@@ -11,4 +11,5 @@ terraform.tfstate.backup
 .vscode/*
 **/derby.log
 **/metastore_db/*
-.env
+.env
+.idea

module_4_rag/.gitignore

@@ -0,0 +1 @@
+data/*

module_4_rag/.python-version

@@ -0,0 +1 @@
+3.9

module_4_rag/batch_score_documents.py

@@ -4,8 +4,8 @@
 import torch
 import torch.nn.functional as F
 
-INPUT_FILENAME = "city_wikipedia_summaries.csv"
-EXPORT_FILENAME = "city_wikipedia_summaries_with_embeddings.csv"
+INPUT_FILENAME = "./data/city_wikipedia_summaries.csv"
+EXPORT_FILENAME = "./data/city_wikipedia_summaries_with_embeddings.parquet"
 TOKENIZER = 'sentence-transformers/all-MiniLM-L6-v2'
 MODEL = 'sentence-transformers/all-MiniLM-L6-v2'
 
@@ -35,8 +35,10 @@ def score_data() -> None:
         print('shape = ', df.shape)
         df['Embeddings'] = list(embeddings.detach().cpu().numpy())
         print("embeddings generated...")
+        df['event_timestamp'] = pd.to_datetime('today')
+        df["item_id"] = df.index
         print(df.head())
-        df.to_csv(EXPORT_FILENAME, index=False)
+        df.to_parquet(EXPORT_FILENAME, index=False)
         print("...data exported. job complete")
     else:
         print("scored data found...skipping generating embeddings.")

module_4_rag/feature_repo/entities.py

@@ -0,0 +1,3 @@
+from feast import Entity
+
+item = Entity(name="item_id")

module_4_rag/feature_repo/feature_store.yaml

@@ -0,0 +1,20 @@
+project: feast_demo_local
+provider: local
+registry:
+  registry_type: sql
+  path: postgresql://@localhost:5432/feast
+online_store:
+  type: postgres
+  pgvector_enabled: true
+  vector_len: 384
+  host: 127.0.0.1
+  port: 5432
+  database: feast
+  user: ""
+  password: ""
+
+
+offline_store:
+  type: file
+entity_key_serialization_version: 2
+

module_4_rag/feature_repo/features.py

@@ -0,0 +1,28 @@
+from datetime import timedelta
+
+from feast import (
+    FeatureView,
+    Field, FileSource,
+)
+from feast.data_format import ParquetFormat
+from feast.types import Float32, Array
+from entities import item
+
+
+parquet_file_path = "../data/city_wikipedia_summaries_with_embeddings.parquet"
+
+source = FileSource(
+    file_format=ParquetFormat(),
+    path=parquet_file_path,
+    timestamp_field="event_timestamp",
+)
+
+city_embeddings_feature_view = FeatureView(
+    name="city_embeddings",
+    entities=[item],
+    schema=[
+        Field(name="Embeddings", dtype=Array(Float32)),
+    ],
+    source=source,
+    ttl=timedelta(hours=2),
+)

module_4_rag/feature_repo/module_1.ipynb

@@ -8,7 +8,7 @@ packages = [{include = "feast_rag"}]
 
 [tool.poetry.dependencies]
 python = "^3.9"
-feast = "^0.35.0"
+feast = "^0.37.0"
 torch = "^2.2.0"
 flasgger = "^0.9.7.1"
 wikipedia = "^1.4.0"