未验证 提交 832e8bf0 编写于 作者: S sys1874 提交者: GitHub

Merge pull request #122 from sys1874/main

Update README.MD
......@@ -44,5 +44,5 @@ Arxiv_dataset(Full Batch): Products_dataset(NeighborSampler):
| Products_baseline | 0.8023 ± 0.0026 | 0.9286 ± 0.0017 | 1,470,905 | Tesla V100 (32GB) |
| Products_UniPM | 0.8256 ± 0.0031 | 0.9308 ± 0.0017 | 1,475,605 | Tesla V100 (32GB) |
| Proteins_baseline | 0.8611 ± 0.0017 | 0.9128 ± 0.0007 | 1,879,664 | Tesla V100 (32GB) |
| Proteins_UniPM | 0.8643 ± 0.0016 | 0.9175 ± 0.0007 | 1,909,104 | Tesla V100 (32GB) |
| Proteins_UniPM | 0.8642 ± 0.0008 | 0.9175 ± 0.0007 | 1,909,104 | Tesla V100 (32GB) |
......@@ -268,7 +268,7 @@ class Arxiv_label_embedding_model():
hidden_size=self.hidden_size,
num_heads=self.num_heads,
concat=True,
layer_norm=True, relu=True)
layer_norm=True, relu=True, gate=True)
if self.dropout > 0:
feature_batch = L.dropout(feature_batch, dropout_prob=self.dropout,
dropout_implementation='upscale_in_train')
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册