Xin Yu
Xin Yu
I have the same qeustion. Can we extract features without using the MSA block?
Same issue here! Using 4GPU for distributed training on Ubuntu 22.04 with Tensorflow 2.13 hangs at the “compiled cluster using the XLA” line. Issue solved by downgrading to 2.12
Agree. Must have for me
Same issue here. Like @gramster I also have to turn off block embedding in order for it to work again.
I concur. It would be great to have an option to set global goal for the vault. I encountered some issues when adding goals to certain folders (while other folders...
This is great to hear. In addition to `pretrain.py`, would you be able to provide a slice of dataset to test the pertaining script? @subercui
I found a workaround, and want to post it here and see if it makes sense: In **Tutorial_Attention_GRN.ipynb**, section 1.1, add `domain_spec_batchnorm = "batchnorm"` to model initiation as follows, so...
I also suspect there is something with the data returned by the public Colab MMseq server. Tamarind Bio's Boltz server is seemingly running on a single L4 GPU and has...
same issue here. Thanks for the workaround!
@benseddikmo I have the same issue with the tabs too, in version 1.24.0. This was also brought up in https://discuss.streamlit.io/t/returning-to-the-right-tab-after-clicking-button-in-1-24-1/46577/9?u=tilleul and #6923. Hopefully a fix will be released soon!