12111620 丁一轩

config/gmf_factor8neg4 .yaml

@@ -0,0 +1,20 @@
+experiment_type: gmf
+alias: gmf_factor8neg4
+num_epoch: 100
+batch_size: 1024
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mf: 8
+latent_dim_mlp: 8
+num_negative: 4
+layers: [16, 64, 32, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001
+weight_init_gaussian: true
+use_cuda: true
+device_id: 0
+pretrain: false
+pretrain_mf: checkpoints/gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+pretrain_mlp: checkpoints/mlp_factor8neg4_Epoch100_HR0.5606_NDCG0.2463.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4
+num_epoch: 100
+batch_size: 256  # 1024
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 8
+num_negative: 4
+layers: [16, 64, 32, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4_layer0.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4_layer0
+num_epoch: 200
+batch_size: 256  # 1024,
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 4
+num_negative: 4
+layers: [8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4_layer1.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4_layer1
+num_epoch: 100
+batch_size: 256  # 1024,
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 4
+num_negative: 4
+layers: [8, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4_layer2.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4_layer2
+num_epoch: 100
+batch_size: 256  # 1024,
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 4
+num_negative: 4
+layers: [8, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4_layer3.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4_layer3
+num_epoch: 100
+batch_size: 256  # 1024,
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 4
+num_negative: 4
+layers: [8, 32, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/mlp_factor8neg4_layer4.yaml

@@ -0,0 +1,18 @@
+experiment_type: mlp
+alias: mlp_factor8neg4_layer4
+num_epoch: 100
+batch_size: 256  # 1024,
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mlp: 4
+num_negative: 4
+layers: [8, 64, 32, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001  # MLP model is sensitive to hyper params
+weight_init_gaussian: True
+use_cuda: True
+device_id: 0
+pretrain: False
+pretrain_mf: checkpoints/{}'.format('gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

config/neumf_factor8neg4.yaml

@@ -0,0 +1,20 @@
+experiment_type: neumf
+alias: neumf_factor8neg4
+num_epoch: 100
+batch_size: 1024
+optimizer: adam
+adam_lr: 0.001
+num_users: 6040
+num_items: 3706
+latent_dim_mf: 8
+latent_dim_mlp: 8
+num_negative: 4
+layers: [16, 64, 32, 16, 8]  # layers[0] is the concat of latent user vector & latent item vector
+l2_regularization: 0.0000001
+weight_init_gaussian: true
+use_cuda: true
+device_id: 0
+pretrain: false
+pretrain_mf: checkpoints/gmf_factor8neg4_Epoch100_HR0.6391_NDCG0.2852.model
+pretrain_mlp: checkpoints/mlp_factor8neg4_Epoch100_HR0.5606_NDCG0.2463.model
+model_dir: checkpoints/{}_Epoch{}_HR{:.4f}_NDCG{:.4f}.model

data.py

@@ -0,0 +1,80 @@
+import torch
+import random
+import pandas as pd
+from copy import deepcopy
+from torch.utils.data import DataLoader, Dataset
+
+class UserItemRatingDataset(Dataset):
+    def __init__(self, user, item, target) -> None:
+        super().__init__()
+        self.user = user
+        self.item = item
+        self.target = target
+
+    def __getitem__(self, index):
+        return self.user[index], self.item[index], self.target[index]
+
+    def __len__(self):
+        return self.user.size(0)
+
+class Generator:
+    def __init__(self, ratings):
+        self.ratings = ratings
+        self.preprocess_ratings = self._binarize(ratings)
+        self.user_pool = set(self.ratings['userId'].unique())
+        self.item_pool = set(self.ratings['itemId'].unique())
+        self.negatives = self._negative(self.ratings)
+        self.train_ratings, self.test_ratings = self._split(self.preprocess_ratings)
+
+    # For explicit feedback
+    def _binarize(self, ratings):
+        ratings = deepcopy(ratings)
+        ratings['rating'][ratings['rating'] > 0] = 1.0
+        return ratings
+
+    def _negative(self, ratings):
+        interact_status = ratings.groupby('userId')['itemId'].apply(set).reset_index().rename(
+            columns={'itemId': 'interacted_items'}
+        )
+        interact_status['negative_items'] = interact_status['interacted_items'].apply(lambda x: self.item_pool - x)
+        interact_status['negative_samples'] = interact_status['negative_items'].apply(lambda x: random.sample(x, 99))
+        return interact_status[['userId', 'negative_items', 'negative_samples']]
+
+    def _split(self, ratings):
+        ratings['rank_latest'] = ratings.groupby('userId')['timestamp'].rank(method='first', ascending=False)
+        test = ratings[ratings['rank_latest'] == 1]
+        train = ratings[ratings['rank_latest'] > 1]
+        assert train['userId'].nunique() == test['userId'].nunique()
+        return train[['userId', 'itemId', 'rating']], test[['userId', 'itemId', 'rating']]
+
+    def get_train_loader(self, num_negatives, batch_size):
+        users, items, ratings = [], [], []
+        train_ratings = pd.merge(self.train_ratings, self.negatives[['userId', 'negative_items']], on='userId')
+        train_ratings['negatives'] = train_ratings['negative_items'].apply(lambda x: random.sample(x, num_negatives))
+        for row in train_ratings.itertuples():
+            users.append(int(row.userId))
+            items.append(int(row.itemId))
+            ratings.append(float(row.rating))
+            for i in range(num_negatives):
+                users.append(int(row.userId))
+                items.append(int(row.negatives[i]))
+                ratings.append(float(0)) # negative samples get 0 rating
+
+        dataset = UserItemRatingDataset(user = torch.LongTensor(users),
+                                        item = torch.LongTensor(items), 
+                                        target = torch.FloatTensor(ratings))
+
+        return DataLoader(dataset, batch_size=batch_size, shuffle=True)
+
+    @property
+    def get_evaluate_data(self):
+        test_ratings = pd.merge(self.test_ratings, self.negatives[['userId', 'negative_samples']], on='userId')
+        test_users, test_items, negative_users, negative_items = [], [], [], []
+        for row in test_ratings.itertuples():
+            test_users.append(int(row.userId))
+            test_items.append(int(row.itemId))
+            for i in range(len(row.negative_samples)):
+                negative_users.append(int(row.userId))
+                negative_items.append(int(row.negative_samples[i]))
+        return [torch.LongTensor(test_users), torch.LongTensor(test_items), torch.LongTensor(negative_users), torch.LongTensor(negative_items)]
+