Apache Solr open-source search software
solr/example/films/vectors/README.md tweaks (#2121)
Christine Poerschkee1c700c277a8
diff --git a/solr/example/films/vectors/README.md b/solr/example/films/vectors/README.mdindex 5a2a7c01751..0145ac0994e 100644--- a/solr/example/films/vectors/README.md+++ b/solr/example/films/vectors/README.md@@ -23,6 +23,12 @@ We present in this directory the Python scripts that were used to create the `fiTo replicate the example you have to run the `create_model.py` script first, followed by `create_dataset.py`. We will describe and discuss each of these scripts below.+## Setup++```+pip install sentence-transformers+```+## Creating the Model (`create_model.py`)There are several approaches that one could use to create vectors (embeddings) to represent documents. In the case of our example we decided to use a _textual_ approach, where we use the text of the document as input for calculating its vector.diff --git a/solr/example/films/vectors/films.py b/solr/example/films/vectors/films.pyindex 72947c01860..cd1a712eb39 100644--- a/solr/example/films/vectors/films.py+++ b/solr/example/films/vectors/films.py@@ -20,7 +20,7 @@ import csvfrom lxml import etreefrom sentence_transformers import SentenceTransformer-PATH_FILMS_DATASET = "./data/films.json"+PATH_FILMS_DATASET = "../films.json"PATH_FILMS_MODEL = "./models/films-model-size_10"PATH_FILMS_VECTORS_JSON = "./data/films-vectors.json"PATH_FILMS_VECTORS_XML = "./data/films-vectors.xml"