From 404c6fb1a3aa8e19db25a5c53098ce4f8a917fd7 Mon Sep 17 00:00:00 2001 From: navanchauhan Date: Sun, 22 May 2022 11:42:00 -0600 Subject: cleaned with black --- db2pc.py | 29 +++++++++++++++++------------ 1 file changed, 17 insertions(+), 12 deletions(-) (limited to 'db2pc.py') diff --git a/db2pc.py b/db2pc.py index 4acddcb..48d3c4a 100644 --- a/db2pc.py +++ b/db2pc.py @@ -20,18 +20,23 @@ model = SentenceTransformer("paraphrase-multilingual-MiniLM-L12-v2") batch_size = 32 df = pd.read_sql("Select * from movies", engine) -df["combined_text"] = df["title"] + ": " + df["overview"].fillna('') + " - " + df["tagline"].fillna('') + " Genres:- " + df["genres"].fillna('') +df["combined_text"] = ( + df["title"] + + ": " + + df["overview"].fillna("") + + " - " + + df["tagline"].fillna("") + + " Genres:- " + + df["genres"].fillna("") +) print(f'Length of Combined Text: {len(df["combined_text"].tolist())}') -for x in tqdm(range(0,len(df),batch_size)): - to_send = [] - trakt_ids = df["trakt_id"][x:x+batch_size].tolist() - sentences = df["combined_text"][x:x+batch_size].tolist() - embeddings = model.encode(sentences) - for idx, value in enumerate(trakt_ids): - to_send.append( - ( - str(value), embeddings[idx].tolist() - )) - index.upsert(to_send) \ No newline at end of file +for x in tqdm(range(0, len(df), batch_size)): + to_send = [] + trakt_ids = df["trakt_id"][x : x + batch_size].tolist() + sentences = df["combined_text"][x : x + batch_size].tolist() + embeddings = model.encode(sentences) + for idx, value in enumerate(trakt_ids): + to_send.append((str(value), embeddings[idx].tolist())) + index.upsert(to_send) -- cgit v1.2.3