Only get_embedding once

scottleibrand · web-flow · commit ae45a48de36b · 2022-12-23T12:30:26.000-08:00
openai@fd181ec updated what was previously two different get_embedding() calls w/ different models to be two identical calls. Rather than doing the same API call twice in a row, we can now just set the second variable to be equal to the first one.
diff --git a/examples/Obtain_dataset.ipynb b/examples/Obtain_dataset.ipynb
@@ -162,7 +162,7 @@
     "\n",
     "# This will take just between 5 and 10 minutes\n",
     "df['ada_similarity'] = df.combined.apply(lambda x: get_embedding(x, engine='text-embedding-ada-002'))\n",
-    "df['ada_search'] = df.combined.apply(lambda x: get_embedding(x, engine='text-embedding-ada-002'))\n",
+    "df['ada_search'] = df['ada_similarity']\n",
     "df.to_csv('data/fine_food_reviews_with_embeddings_1k.csv')"
    ]
   }

Original file line number	Diff line number	Diff line change
`@@ -162,7 +162,7 @@`
`162`	`162`	`"\n",`
`163`	`163`	`"# This will take just between 5 and 10 minutes\n",`
`164`	`164`	`"df['ada_similarity'] = df.combined.apply(lambda x: get_embedding(x, engine='text-embedding-ada-002'))\n",`
`165`		`- "df['ada_search'] = df.combined.apply(lambda x: get_embedding(x, engine='text-embedding-ada-002'))\n",`
	`165`	`+ "df['ada_search'] = df['ada_similarity']\n",`
`166`	`166`	`"df.to_csv('data/fine_food_reviews_with_embeddings_1k.csv')"`
`167`	`167`	`]`
`168`	`168`	`}`