Skip to content

Commit

Permalink
update
Browse files Browse the repository at this point in the history
  • Loading branch information
blazerye committed Jul 26, 2024
1 parent 528e440 commit e701a67
Show file tree
Hide file tree
Showing 2 changed files with 3 additions and 3 deletions.
2 changes: 1 addition & 1 deletion evaluate/evaluate.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@

## Precautions:
Our evaluation process adopts the workflow of [ChatDrug](https://github.com/chao1224/ChatDrug) for evaluating small molecules. Therefore, the evaluation code is based on the script provided by this work. Please note that, unlike this work, our success rate calculation method is `the number of successful optimizations` / `the total number of molecules`, while this work is `the number of successful optimizations` / `the number of valid molecules`.
In addition, in our optimization tasks involving "esol", "bbbp", and "herg" properties, the property values are given by the idrug model. Currently, the API call of the idrug model is not open to the public, but the [web query](https://drug.ai.tencent.com/cn) is free for everyone. You can query the aforementioned property values of molecules through this method. If you want to evaluate the above tasks in batches, please implement your own property prediction model at the code entry we reserved.
In addition, in our optimization tasks involving "esol", "bbbp", and "herg" properties, the property values are given by the idrug model. Currently, the API call of the idrug model is not open to the public, but the [web query](https://drug.ai.tencent.com/cn) is available for everyone. You can query the aforementioned property values of molecules through this method. If you want to evaluate the above tasks in batches, please implement your own property prediction model at the code entry we reserved.

## Script Parameter Explanation:
`--task`: Task name, currently supported values are: `qed+`, `acceptor+`, `donor+`, `esol+`, `bbbp+`, `herg-`, `esol+acceptor+`, `qed+bbbp+`.
Expand Down
4 changes: 2 additions & 2 deletions evaluate/main_eval.py
Original file line number Diff line number Diff line change
Expand Up @@ -175,14 +175,14 @@ def main_assit(args):
parser.add_argument('--num_round', required=False, type=int, default=2, help='number of conversation round')
parser.add_argument('--client_add', required=True, type=str, help='DrugAssist client address set by gradio service')
parser.add_argument('--database', required=False, type=str, default='mol_DB.csv',help='molecule database')
parser.add_argument('--test_mol', required=False, type=str, default='testset.csv', help='testset ')
parser.add_argument('--test_mol', required=False, type=str, default='testset.csv', help='testset')
args = parser.parse_args()
args = vars(args)


if 'esol' in args['task'] or 'bbbp' in args['task'] or 'herg' in args['task']:
choice = input("The property value involved in this task will be given by a specific prediction model, make sure you have implemented it yourself!\
If you have any questions, please refer to xxx. Enter y to confirm, enter n to exit\n")
If you have any questions, please refer to https://github.com/blazerye/DrugAssist/blob/main/evaluate/evaluate.md. Enter y to confirm, enter n to exit\n")
if choice == 'y':
main_assit(args)
else:
Expand Down

0 comments on commit e701a67

Please sign in to comment.