scGPT
scGPT copied to clipboard
bowang-lab
Hi, thanks a lot for the great work! It seems that the data module is missing, and there is module import related to this. https://github.com/bowang-lab/scGPT/blob/bc8939504fc62dd617360618a840ca9b67419b5a/scgpt/scbank/databank.py#L10 I cannot find a file...
Hi, Congratulations on the Nature paper. I was wondering if it is only possible to generate gene expression data - after pre-training and finetuning - when the cell embedding is...
I met the same error while using the model.py file. [https://github.com/bowang-lab/scGPT/issues/69] which is AttributeError: 'FlashMHA' object has no attribute 'batch_first' then I modified the code referred to the Issue #69...
Hello, I am trying to run scg.tasks.embed_data on this dataset: https://storage.googleapis.com/linnarsson-lab-human/human_dev_GRCh38-3.0.0.h5ad from https://github.com/linnarsson-lab/developing-human-brain/ I will get to the point where the tqdm progress bar shows up for embedding cells, but...
Hi developer, Thanks to develop the great tool to annotate the single cell, I wander that this scGPT must be need GPU on centos7.9, and i hadn't the GPU, what...
In your paper, you mentioned that you used MS data for fine-tuning cell type annotation task, where the training data consist of normal patients and testing data consist of MS...
Hi, This is a great tool and I am trying to implement it in mouse brain data. I have downloaded the model using the link. I wanted to predict the...
Hi, I tried to fine tune the model on my data for integration. However, the sparsity of my data is higher than the common scRNA-seq dataset. And when I used...
```py # in purtabation toturial .... model_dict = model.state_dict() pretrained_dict = torch.load(model_file) pretrained_dict = { k: v for k, v in pretrained_dict.items() if any([k.startswith(prefix) for prefix in load_param_prefixs]) } for...
`torch.load(model_file)` does not work if the device is set to CPU, so I replaced all of these calls with `torch.load(model_file, map_location=device)` which is more device-agnostic.
