Spaces:
Sleeping
Sleeping
Commit
•
73159f3
1
Parent(s):
8a43a3a
Update app.py
Browse files
app.py
CHANGED
@@ -63,11 +63,11 @@ if page == "какая-то еще":
|
|
63 |
outputs = model(**tokens)
|
64 |
embeddings = outputs.last_hidden_state.mean(dim=1)
|
65 |
return embeddings
|
66 |
-
|
67 |
embeddings = pd.read_pickle('embeddings.pkl')
|
68 |
user_input = st.text_area('Введите описание фильма')
|
69 |
input_embedding = encode_description(user_input)
|
70 |
-
embeddings_tensor = torch.stack(
|
71 |
|
72 |
# Рассчитайте косинусное сходство
|
73 |
similarity_scores = cosine_similarity(input_embedding.view(1, -1).detach().numpy(), embeddings_tensor.reshape(embeddings_tensor.shape[0], -1))[0]
|
@@ -76,6 +76,6 @@ if page == "какая-то еще":
|
|
76 |
sorted_indices = similarity_scores.argsort()[::-1]
|
77 |
|
78 |
# Используйте индексы для извлечения строк из DataFrame
|
79 |
-
recs =
|
80 |
recs.index = recs.index + 1
|
81 |
st.write(recs[['movie_title', 'description']])
|
|
|
63 |
outputs = model(**tokens)
|
64 |
embeddings = outputs.last_hidden_state.mean(dim=1)
|
65 |
return embeddings
|
66 |
+
df2 = pd.read_csv('data_with_embeddings.csv')
|
67 |
embeddings = pd.read_pickle('embeddings.pkl')
|
68 |
user_input = st.text_area('Введите описание фильма')
|
69 |
input_embedding = encode_description(user_input)
|
70 |
+
embeddings_tensor = torch.stack(df2['description_embedding'].tolist()).numpy()
|
71 |
|
72 |
# Рассчитайте косинусное сходство
|
73 |
similarity_scores = cosine_similarity(input_embedding.view(1, -1).detach().numpy(), embeddings_tensor.reshape(embeddings_tensor.shape[0], -1))[0]
|
|
|
76 |
sorted_indices = similarity_scores.argsort()[::-1]
|
77 |
|
78 |
# Используйте индексы для извлечения строк из DataFrame
|
79 |
+
recs = df2.iloc[sorted_indices[:10]].reset_index(drop=True)
|
80 |
recs.index = recs.index + 1
|
81 |
st.write(recs[['movie_title', 'description']])
|