Update readme.md

Oceanusity · web-flow · commit 2089c8e7e86b · 2024-02-03T16:19:34.000-06:00
diff --git a/OpenDFT/QHNet/readme.md b/OpenDFT/QHNet/readme.md
@@ -46,7 +46,7 @@ python test_mixed.py dataset=all model=QHNet model.version=$Selected_version mod
 
 For the model version, the `QHNet_w_bias` applies self-interaction layer with bias in the `\ell=0` by setting `bias=True` in [self-interaction layer](https://github.com/divelab/AIRS/blob/e1abc14bb9424f563e4ff1a4abd9be73a7b4b52f/OpenDFT/QHNet/models/ori_QHNet_with_bias.py#L178) and normalize the path using [weights from the dictionary](https://github.com/divelab/AIRS/blob/e1abc14bb9424f563e4ff1a4abd9be73a7b4b52f/OpenDFT/QHNet/models/ori_QHNet_with_bias.py#L65). For the `QHNet_wo_bias`, it does not apply such bias and path_normalization operations.
 
-When conducting experiments for the paper, the model is developing and have minior setting difference for different experiments. In order to exactly reproduce the reported results, we provide the model with different versions. After our exploration, we clean our model and provide the suggested `QHNet` for following experiments. We applogize for the inconvience.
+When conducting experiments for the paper, the model is developing and have minior setting difference for different experiments. In order to exactly reproduce the reported results, we provide the model with different versions. After our exploration, we clean our model and provide the suggested `QHNet_w_bias` for following experiments. We applogize for the inconvience.
 
 
 

Original file line number	Diff line number	Diff line change
`@@ -46,7 +46,7 @@ python test_mixed.py dataset=all model=QHNet model.version=$Selected_version mod`
`46`	`46`
`47`	`47`	For the model version, the `QHNet_w_bias` applies self-interaction layer with bias in the `\ell=0` by setting `bias=True` in [self-interaction layer](https://github.com/divelab/AIRS/blob/e1abc14bb9424f563e4ff1a4abd9be73a7b4b52f/OpenDFT/QHNet/models/ori_QHNet_with_bias.py#L178) and normalize the path using [weights from the dictionary](https://github.com/divelab/AIRS/blob/e1abc14bb9424f563e4ff1a4abd9be73a7b4b52f/OpenDFT/QHNet/models/ori_QHNet_with_bias.py#L65). For the `QHNet_wo_bias`, it does not apply such bias and path_normalization operations.
`48`	`48`
`49`		-When conducting experiments for the paper, the model is developing and have minior setting difference for different experiments. In order to exactly reproduce the reported results, we provide the model with different versions. After our exploration, we clean our model and provide the suggested `QHNet` for following experiments. We applogize for the inconvience.
	`49`	+When conducting experiments for the paper, the model is developing and have minior setting difference for different experiments. In order to exactly reproduce the reported results, we provide the model with different versions. After our exploration, we clean our model and provide the suggested `QHNet_w_bias` for following experiments. We applogize for the inconvience.
`50`	`50`
`51`	`51`
`52`	`52`