AlGouvea commited on
Commit
facf41f
·
1 Parent(s): 1cf536f

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +4 -4
app.py CHANGED
@@ -14,7 +14,7 @@ merged_data = ratings.merge(books)
14
 
15
  print(merged_data.columns)
16
 
17
- dls = CollabDataLoaders.from_df(ratings, item_name='Book-Title', bs=8192)
18
 
19
  class CollabNN(Module):
20
  def __init__(self, user_sz, item_sz, y_range=(0,10.5), n_act=150):
@@ -59,7 +59,7 @@ print("Loaded Both")
59
 
60
  # Function to get top N recommendations for a user using the Dot Product Model
61
  def get_top_recommendations_dot(user_id, dls, n=3):
62
- item_ids = dls.classes['Book-Title'].o2i[ratings['Book-Title']]
63
  user_ids = [user_id] * len(item_ids)
64
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(learn_dot.model.device)
65
  predictions_dot = learn_dot.model(user_item_pairs).squeeze().tolist()
@@ -69,12 +69,12 @@ def get_top_recommendations_dot(user_id, dls, n=3):
69
 
70
  # Function to get top N recommendations for a user using the CollabNN Model
71
  def get_top_recommendations_nn(user_id, n=3):
72
- item_ids = dls.classes['Book-Title'].o2i[ratings['Book-Title']]
73
  user_ids = [user_id] * len(item_ids)
74
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(deep_learner.model.device)
75
  predictions_nn = deep_learner.model(user_item_pairs).squeeze().tolist()
76
  top_indices = sorted(range(len(predictions_nn)), key=lambda i: predictions_nn[i], reverse=True)[:n]
77
- top_recommendations = [dls.classes['Book-Title'].i2o[item_id] for item_id in top_indices]
78
  return top_recommendations
79
 
80
  # Create a Gradio interface
 
14
 
15
  print(merged_data.columns)
16
 
17
+ dls = CollabDataLoaders.from_df(ratings, item_name='title', bs=8192)
18
 
19
  class CollabNN(Module):
20
  def __init__(self, user_sz, item_sz, y_range=(0,10.5), n_act=150):
 
59
 
60
  # Function to get top N recommendations for a user using the Dot Product Model
61
  def get_top_recommendations_dot(user_id, dls, n=3):
62
+ item_ids = dls.classes['title'].o2i[ratings['title']]
63
  user_ids = [user_id] * len(item_ids)
64
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(learn_dot.model.device)
65
  predictions_dot = learn_dot.model(user_item_pairs).squeeze().tolist()
 
69
 
70
  # Function to get top N recommendations for a user using the CollabNN Model
71
  def get_top_recommendations_nn(user_id, n=3):
72
+ item_ids = dls.classes['title'].o2i[ratings['title']]
73
  user_ids = [user_id] * len(item_ids)
74
  user_item_pairs = torch.tensor(list(zip(user_ids, item_ids))).to(deep_learner.model.device)
75
  predictions_nn = deep_learner.model(user_item_pairs).squeeze().tolist()
76
  top_indices = sorted(range(len(predictions_nn)), key=lambda i: predictions_nn[i], reverse=True)[:n]
77
+ top_recommendations = [dls.classes['title'].i2o[item_id] for item_id in top_indices]
78
  return top_recommendations
79
 
80
  # Create a Gradio interface