nashid
nashid
@KKGanguly were you able to run the model for training from scratch? Looks like code is now pushed.
@thibolu thanks for sharing the artefact. The steps appear to be incomplete and convoluted. I have downloaded the dataset from https://github.com/lin-tan/CoCoNut-Artifact/releases/tag/training_data_1.0.0 Dataset for python as an example: ``` coconut/coconut-dataset/python/2010$ ls...
@BBerabi @amilankovich-slab @KinuSan000 @dbr7 you were able to train the model with the dataset? What steps did you take?
@amilankovich-slab it appears to me, I have to take the following steps: - unzip data like for python.tar.lrz - then run the script for including preprocessing generate_data.py - Then I...
great, thanks for the pointer @jiang719. We really appreciate it.
@faysalhossain2007 were you able to run the model for training from scratch?
@lin-tan and @jiang719 we loaded the OpenAIGPTLMHeadModel(config)) to train a new model from scratch. This change looks like: ``` - gpt_loaded = torch.load(gpt_file) - config = gpt_loaded['config'] - gpt_model.load_state_dict(gpt_loaded['model']) +...
@jiang719 I cant find any model pushed to the repo. Where can I find the checkpoint?
@jiang719 can you please share the trained model or upload somewhere so that others can download?
@MysterAitch I was asking whether JavaParser would be a fit for performing static program slicing like forward slicing, backward slicing as an example. https://en.wikipedia.org/wiki/Program_slicing