Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Refactor madx_run_clm.py #27

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from
Draft

[WIP] Refactor madx_run_clm.py #27

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 1 addition & 29 deletions scripts/lang_adapt/madx_run_clm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -433,11 +433,6 @@ def group_texts(examples):
return lm_datasets

def modify_model(adapter_args, data_args, model_args, tokenizer, model):
#if "emb" in model_args.lang_adapt_strategies:
# if "replace" in model_args.embedding_strategies:
# for name, param in model.named_parameters():
# if "wte" not in name and "wpe" not in name and "lm_head" not in name:
# param.requires_grad = False

def get_adapter_config(adapter_args, model_args):
if adapter_args.adapter_config == "prefix_tuning":
Expand Down Expand Up @@ -566,16 +561,6 @@ def zero_grad(grad):

embedding_layer.weight.register_hook(lambda grad: zero_grad(grad))

#if model_args.embedding_strategies == "overlap-replace":
# if not tokenizer.name_or_path == model_args.model_name_or_path:
# orig_tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path)
# model.add_embeddings('lng_emb', tokenizer, reference_embedding='default', reference_tokenizer=orig_tokenizer )
# model._active_embedding = "lng_emb"
# model.delete_embeddings('default')
# model.tie_weights()
#elif model_args.embedding_strategies == "replace":
# model.resize_token_embeddings(len(tokenizer))

trainable_params = 0
frozen_params = 0
emb_params = 0
Expand Down Expand Up @@ -688,9 +673,7 @@ def main():
print("Model: 👇")
print(model)


# print("Embeddings at start of run:", model.get_input_embeddings().weight[250880:,:]) # get original weight for embedding layer
# orig_embeddings = model.get_input_embeddings().weight.detach().clone() # clone original weight for embedding layer

# Training
if training_args.do_train:
checkpoint = None
Expand Down Expand Up @@ -725,17 +708,6 @@ def main():
trainer.log_metrics("train", metrics)
trainer.save_metrics("train", metrics)
trainer.save_state()

# uncomment to test whether extending vocab gradient masking is working correctly.
# if model_args.embedding_strategies == "extend":
# print("Unsliced, post-training:", model.get_input_embeddings().weight) # get updated weight
# if not torch.equal(orig_embeddings[:250880, :], model.get_input_embeddings().weight[:250880, :]):
# raise ValueError("embedding layer is updated where it shouldn't....")

# if torch.equal(orig_embeddings[250880:, :], model.get_input_embeddings().weight[250880:, :]):
# print("original embeddings:", orig_embeddings[250880:, :])
# print("updated embeddings:", model.get_input_embeddings().weight[250880:, :])
# raise ValueError("embedding layer is not updated where it should....")


# Evaluation
Expand Down