AlGouvea commited on
Commit
fedecfd
·
1 Parent(s): 6b2fdc6

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +5 -5
app.py CHANGED
@@ -12,7 +12,7 @@ books = pd.read_csv('BX_Books.csv', encoding='latin-1', delimiter=';', header=No
12
  # Merge the data
13
  merged_data = ratings.merge(books)
14
 
15
- dls = CollabDataLoaders.from_df(ratings, item_name='title', bs=8192)
16
 
17
  class CollabNN(Module):
18
  def __init__(self, user_sz, item_sz, y_range=(0,10.5), n_act=150):
@@ -57,22 +57,22 @@ print("Loaded Both")
57
 
58
  # Function to get top N recommendations for a user using the Dot Product Model
59
  def get_top_recommendations_dot(user_id, dls, n=3):
60
- item_ids = dls.classes['title'].o2i[ratings['title']]
61
  user_ids = [user_id] * len(item_ids)
62
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(learn_dot.model.device)
63
  predictions_dot = learn_dot.model(user_item_pairs).squeeze().tolist()
64
  top_indices = sorted(range(len(predictions_dot)), key=lambda i: predictions_dot[i], reverse=True)[:n]
65
- top_recommendations = [dls.classes['title'].i2o[item_id] for item_id in top_indices]
66
  return top_recommendations
67
 
68
  # Function to get top N recommendations for a user using the CollabNN Model
69
  def get_top_recommendations_nn(user_id, n=3):
70
- item_ids = dls.classes['title'].o2i[ratings['title']]
71
  user_ids = [user_id] * len(item_ids)
72
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(deep_learner.model.device)
73
  predictions_nn = deep_learner.model(user_item_pairs).squeeze().tolist()
74
  top_indices = sorted(range(len(predictions_nn)), key=lambda i: predictions_nn[i], reverse=True)[:n]
75
- top_recommendations = [dls.classes['title'].i2o[item_id] for item_id in top_indices]
76
  return top_recommendations
77
 
78
  # Create a Gradio interface
 
12
  # Merge the data
13
  merged_data = ratings.merge(books)
14
 
15
+ dls = CollabDataLoaders.from_df(ratings, item_name='Book-Title', bs=8192)
16
 
17
  class CollabNN(Module):
18
  def __init__(self, user_sz, item_sz, y_range=(0,10.5), n_act=150):
 
57
 
58
  # Function to get top N recommendations for a user using the Dot Product Model
59
  def get_top_recommendations_dot(user_id, dls, n=3):
60
+ item_ids = dls.classes['Book-Title'].o2i[ratings['Book-Title']]
61
  user_ids = [user_id] * len(item_ids)
62
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(learn_dot.model.device)
63
  predictions_dot = learn_dot.model(user_item_pairs).squeeze().tolist()
64
  top_indices = sorted(range(len(predictions_dot)), key=lambda i: predictions_dot[i], reverse=True)[:n]
65
+ top_recommendations = [dls.classes['Book-Title'].i2o[item_id] for item_id in top_indices]
66
  return top_recommendations
67
 
68
  # Function to get top N recommendations for a user using the CollabNN Model
69
  def get_top_recommendations_nn(user_id, n=3):
70
+ item_ids = dls.classes['Book-Title'].o2i[ratings['Book-Title']]
71
  user_ids = [user_id] * len(item_ids)
72
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(deep_learner.model.device)
73
  predictions_nn = deep_learner.model(user_item_pairs).squeeze().tolist()
74
  top_indices = sorted(range(len(predictions_nn)), key=lambda i: predictions_nn[i], reverse=True)[:n]
75
+ top_recommendations = [dls.classes['Book-Title'].i2o[item_id] for item_id in top_indices]
76
  return top_recommendations
77
 
78
  # Create a Gradio interface