aicloud
/
chatglm-6b
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix attention_mask and position_ids

This commit is contained in:
duzx16 2023-04-02 01:58:45 +08:00
parent e22cddf212
commit 373fd6b9d4
1 changed files with 22 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
tokenization_chatglm.py
View File

@ -340,7 +340,7 @@ class ChatGLMTokenizer(PreTrainedTokenizer):
token_ids_0 += [self.sp_tokenizer[self.bos_token]] token_ids_0 += [self.sp_tokenizer[self.bos_token]]
if token_ids_1 is not None: if token_ids_1 is not None:
if token_ids_1[-1] != eop_id: if not token_ids_1 or token_ids_1[-1] != eop_id:
token_ids_1 += [eop_id] token_ids_1 += [eop_id]
token_ids_0 += token_ids_1 token_ids_0 += token_ids_1
@ -397,7 +397,8 @@ class ChatGLMTokenizer(PreTrainedTokenizer):
needs_to_be_padded = padding_strategy != PaddingStrategy.DO_NOT_PAD and len(required_input) != max_length needs_to_be_padded = padding_strategy != PaddingStrategy.DO_NOT_PAD and len(required_input) != max_length
# Initialize attention mask if not present. # Initialize attention mask if not present.
if return_attention_mask: if max_length is not None:
if "attention_mask" not in encoded_inputs:
if bos_token_id in required_input: if bos_token_id in required_input:
context_length = required_input.index(bos_token_id) context_length = required_input.index(bos_token_id)
else: else:
@ -408,6 +409,7 @@ class ChatGLMTokenizer(PreTrainedTokenizer):
attention_mask = np.bool_(attention_mask < 0.5) attention_mask = np.bool_(attention_mask < 0.5)
encoded_inputs["attention_mask"] = attention_mask encoded_inputs["attention_mask"] = attention_mask
if "position_ids" not in encoded_inputs:
position_ids = np.arange(seq_length, dtype=np.int64) position_ids = np.arange(seq_length, dtype=np.int64)
mask_token = mask_token_id if mask_token_id in required_input else gmask_token_id mask_token = mask_token_id if mask_token_id in required_input else gmask_token_id
if mask_token in required_input: if mask_token in required_input: