Matthew Ong

I wanted to build a semantic frame database based on another language (Akkadian) and then use it to do frame parsing on my own Akkadian corpus. I assumed one could...

Indeed, which I am prepared to do (hopefully in Inception) if it can be used to train the parser. Michael Ringgaard ***@***.***> 于2023年3月29日周三 00:12写道： > You will need an annotated...

Hello? It seems that no one is maintaining this project anymore...

Hello. I also recently posted a question but have never received an answer. It seems this project is dead.

Hi, I believe I installed from the main branch of the repository (pip install git+https://github.com/mittagessen/kraken.git) on a VM and ran the same command as above, but I got the same...

Hmm. Then I wonder if anyone has managed to get ```party train``` to work....

> You need to compile one or more training and validation datasets > separately and then add manifest files containing paths with the > coomandline options -t/-e as explained in...

> Yes, exactly. > […](#) I used two files ```train.lst``` and ```val.list```, each listing just one .arrow file each. That part seemed to work. However then the following error arose:...

> On 25/06/17 02:46AM, Matthew Ong wrote: > megamattc left a comment ([mittagessen/party#15](https://github.com/mittagessen/party/issues/15)) > > > Yes, exactly. > > […](#) > > I used two files ```train.lst``` and ```val.list```,...

> --load-from-safetensors model.safetensors I downloaded the model.safetensors file into the local folder and executed the revised command: ``` party train --load-from-safetensors model.safetensors --workers 14 -t train.lst -e val.lst --prompt-mode curves...