-
Hi,
I am interested in using your Pre-trained Model for single-step prediction. `rxn4chemistry_wrapper.predict_automatic_retrosynthesis` will return the multi-step retrosynthesis pathways. Is there…
-
I have trouble reproducing the results of the paper, and I wanted to know if there was anything that I was doing wrong. I evaluated all the fine-tuned models provided (https://az.app.box.com/s/7eci3nd…
-
Hello,
Thanks for sharing this library! The results of https://torchdrug.ai/docs/tutorials/retrosynthesis.html are different from the [ G2Gs](https://arxiv.org/pdf/2003.12725.pdf).
For reaction clas…
-
hi,
Im trying to reproduce the results for single-step RT, by fine-tuning pretrained model on PubMed, with the USPTO_50k dataset. Im doing this by using PyTorch Trainer: Seq2SeqTrainer, and adding …
-
Hi, illuminated by the paper, "Artificial applicability labels for improving policies in retrosynthesis prediction", I wanna retrain the template applicability filter policy using USPTO data to enhanc…
-
### Week 1 - Get to know the community
- [X] Join the communication channels
- [X] Open a GitHub issue (this one!)
- [x] Install the Ersilia Model Hub and test the simplest model
- [x] Write a motiva…
-
Training the model G2G center identification on 50 epochs and synthon completion on 10 epochs with task = tasks.Retrosynthesis(reaction_task, synthon_task, center_topk=3,
…
-
Hi!
Awesome work and really good results!
I noticed you use a different preprocessing procedure than some of the compared works for USPTO-MIT and USPTO-FULL datasets. For example, USPTO-MIT in the L…
-
[https://figshare.com/articles/software/RetroBioCat_database_files/12696482](url)
![QQ图片20210730175653](https://user-images.githubusercontent.com/50348328/127636834-b8b1fd1d-f67e-416b-8ce9-368ea53a…
-
This issue will collect all feedback submitted via the feedback form at the end of each tutorial
----
Results have been [aggregated](https://nbviewer.jupyter.org/github/bebatut/galaxy-training-m…