(use_neg=False, hash_flag=False)
| 7 | |
| 8 | |
| 9 | def get_xy_fd(use_neg=False, hash_flag=False): |
| 10 | feature_columns = [SparseFeat('user', 4, embedding_dim=4, use_hash=hash_flag), |
| 11 | SparseFeat('gender', 2, embedding_dim=4, use_hash=hash_flag), |
| 12 | SparseFeat('item_id', 3 + 1, embedding_dim=8, use_hash=hash_flag), |
| 13 | SparseFeat('cate_id', 2 + 1, embedding_dim=4, use_hash=hash_flag), |
| 14 | DenseFeat('pay_score', 1)] |
| 15 | |
| 16 | feature_columns += [ |
| 17 | VarLenSparseFeat(SparseFeat('hist_item_id', vocabulary_size=3 + 1, embedding_dim=8, embedding_name='item_id'), |
| 18 | maxlen=4, length_name="seq_length"), |
| 19 | VarLenSparseFeat(SparseFeat('hist_cate_id', vocabulary_size=2 + 1, embedding_dim=4, embedding_name='cate_id'), |
| 20 | maxlen=4, |
| 21 | length_name="seq_length")] |
| 22 | |
| 23 | behavior_feature_list = ["item_id", "cate_id"] |
| 24 | uid = np.array([0, 1, 2, 3]) |
| 25 | gender = np.array([0, 1, 0, 1]) |
| 26 | item_id = np.array([1, 2, 3, 2]) # 0 is mask value |
| 27 | cate_id = np.array([1, 2, 1, 2]) # 0 is mask value |
| 28 | score = np.array([0.1, 0.2, 0.3, 0.2]) |
| 29 | |
| 30 | hist_item_id = np.array([[1, 2, 3, 0], [1, 2, 3, 0], [1, 2, 0, 0], [1, 2, 0, 0]]) |
| 31 | hist_cate_id = np.array([[1, 1, 2, 0], [2, 1, 1, 0], [2, 1, 0, 0], [1, 2, 0, 0]]) |
| 32 | |
| 33 | behavior_length = np.array([3, 3, 2, 2]) |
| 34 | |
| 35 | feature_dict = {'user': uid, 'gender': gender, 'item_id': item_id, 'cate_id': cate_id, |
| 36 | 'hist_item_id': hist_item_id, 'hist_cate_id': hist_cate_id, |
| 37 | 'pay_score': score, "seq_length": behavior_length} |
| 38 | |
| 39 | if use_neg: |
| 40 | feature_dict['neg_hist_item_id'] = np.array([[1, 2, 3, 0], [1, 2, 3, 0], [1, 2, 0, 0], [1, 2, 0, 0]]) |
| 41 | feature_dict['neg_hist_cate_id'] = np.array([[1, 1, 2, 0], [2, 1, 1, 0], [2, 1, 0, 0], [1, 2, 0, 0]]) |
| 42 | feature_columns += [ |
| 43 | VarLenSparseFeat( |
| 44 | SparseFeat('neg_hist_item_id', vocabulary_size=3 + 1, embedding_dim=8, embedding_name='item_id'), |
| 45 | maxlen=4, length_name="seq_length"), |
| 46 | VarLenSparseFeat( |
| 47 | SparseFeat('neg_hist_cate_id', vocabulary_size=2 + 1, embedding_dim=4, embedding_name='cate_id'), |
| 48 | maxlen=4, length_name="seq_length")] |
| 49 | |
| 50 | x = {name: feature_dict[name] for name in get_feature_names(feature_columns)} |
| 51 | y = np.array([1, 0, 1, 0]) |
| 52 | return x, y, feature_columns, behavior_feature_list |
| 53 | |
| 54 | |
| 55 | if __name__ == "__main__": |
no test coverage detected