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

Jump to bottom

[NFC] Minor NFC changes #1711

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+0 −21
Open

[NFC] Minor NFC changes #1711

Show file tree
Hide file tree
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
19 changes: 0 additions & 19 deletions apps/language_models/scripts/stablelm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,6 @@ def compile_stableLM(
):
from shark.shark_inference import SharkInference

# device = "cuda" # "cpu"
# TODO: vmfb and mlir name should include precision and device
vmfb_path = (
Path(model_name + f"_{device}.vmfb")
Expand Down Expand Up @@ -130,13 +129,6 @@ def get_tokenizer():
return tok


# sharkStableLM = compile_stableLM
# (
# None,
# tuple([input_ids, attention_mask]),
# "stableLM_linalg_f32_seqLen256",
# "/home/shark/vivek/stableLM_shark_f32_seqLen256"
# )
def generate(
new_text,
max_new_tokens,
Expand All @@ -148,18 +140,8 @@ def generate(
# Construct the input message string for the model by
# concatenating the current system message and conversation history
# Tokenize the messages string
# sharkStableLM = compile_stableLM
# (
# None,
# tuple([input_ids, attention_mask]),
# "stableLM_linalg_f32_seqLen256",
# "/home/shark/vivek/stableLM_shark_f32_seqLen256"
# )
words_list = []
for i in range(max_new_tokens):
# numWords = len(new_text.split())
# if(numWords>220):
# break
params = {
"new_text": new_text,
}
Expand Down Expand Up @@ -188,7 +170,6 @@ def generate_new_token(shark_model, tokenizer, params):
return_tensors="pt",
)
sum_attentionmask = torch.sum(model_inputs.attention_mask)
# sharkStableLM = compile_stableLM(None, tuple([input_ids, attention_mask]), "stableLM_linalg_f32_seqLen256", "/home/shark/vivek/stableLM_shark_f32_seqLen256")
output = shark_model(
"forward", [model_inputs.input_ids, model_inputs.attention_mask]
)
Expand Down
2 changes: 0 additions & 2 deletions apps/language_models/src/pipelines/stablelm_pipeline.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,7 +63,6 @@ def compile(self):
f"stableLM_linalg_{self.precision}_seqLen{self.max_sequence_len}"
)

# device = "cuda" # "cpu"
# TODO: vmfb and mlir name should include precision and device
model_vmfb_name = None
vmfb_path = (
Expand Down Expand Up @@ -120,7 +119,6 @@ def compile(self):
def get_tokenizer(self):
tok = AutoTokenizer.from_pretrained(self.hf_model_path)
tok.add_special_tokens({"pad_token": "<PAD>"})
# print("[DEBUG] Sucessfully loaded the tokenizer to the memory")
return tok

def generate(self, prompt):
Expand Down