使用mindie下atb-models对llama3-70b fp16量化成llama3-70b w4a16，出现expected str, bytes or os.PathLike object, not NoneType报错
已解决发表于2024-09-05 14:03:25
0 查看
【服务器型号】：Atlas 800 I A2
【版本信息】：
--驱动版本：24.1.rc2
--固件版本：7.3.0.1.231
--CANN版本：CANN 8.0.RC2
--Mindie：1.0.RC2
--Python版本：Python3.10.12
--操作系统版本：ubuntu22.04
--pytorch版本：2.1.0
问题现象：TypeError: expected str, bytes or os.PathLike object, not NoneType
期望解决日期：2024.9.6
本帖最后由 匿名用户 于 2024/09/05 14:04:16 编辑
点赞
收藏
回复
分享
举报
【服务器型号】：Atlas 800 I A2
【版本信息】：
--驱动版本：24.1.rc2
--固件版本：7.3.0.1.231
--CANN版本：CANN 8.0.RC2
--Mindie：1.0.RC2
--Python版本：Python3.10.12
--操作系统版本：ubuntu22.04
--pytorch版本：2.1.0
问题现象：TypeError: expected str, bytes or os.PathLike object, not NoneType
python3 examples/models/llama3/convert_quant_weights.py --model_path /workspce/modelzoo/llama3-70b/ --save_directory /workspce/modelzoo/llama3-70b-w4a16 --w_bit 4 --a_bit 16 --w_sym True --mm_tensor False --is_lowbit True --group_size=128 --open_outlier False /usr/local/Ascend/atb-models/examples/models/llama3/convert_quant_weights.py:5: DeprecationWarning: Package API `modelslim` is deprecated and will be removed in future version. Please use `msmodelslim` instead. from modelslim.pytorch.llm_ptq.llm_ptq_tools import QuantConfig 2024-09-05 05:54:25,305 [INFO] [pid: 5791] env.py-55: {'use_ascend': True, 'max_memory_gb': None, 'reserved_memory_gb': 3, 'skip_warmup': False, 'visible_devices': None, 'use_host_chooser': True, 'bind_cpu': True} 2024-09-05 05:54:25,442 - msmodelslim-logger - WARNING - `cpu` is set as `dev_type`, `dev_id` cannot be specified manually! Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [00:22<00:00, 1.32it/s] Traceback (most recent call last): File "/usr/local/Ascend/atb-models/examples/models/llama3/convert_quant_weights.py", line 105, in <module> quant_weight_generator = Quantifier( File "/usr/local/Ascend/atb-models/examples/convert/model_slim/quantifier.py", line 86, in __init__ self.tokenizer = AutoTokenizer.from_pretrained( File "/usr/local/python3.10.12/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 846, in from_pretrained return tokenizer_class_py.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs) File "/usr/local/python3.10.12/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2048, in from_pretrained return cls._from_pretrained( File "/usr/local/python3.10.12/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2287, in _from_pretrained tokenizer = cls(*init_inputs, **init_kwargs) File "/usr/local/python3.10.12/lib/python3.10/site-packages/transformers/models/llama/tokenization_llama.py", line 182, in __init__ self.sp_model = self.get_spm_processor(kwargs.pop("from_slow", False)) File "/usr/local/python3.10.12/lib/python3.10/site-packages/transformers/models/llama/tokenization_llama.py", line 212, in get_spm_processor with open(self.vocab_file, "rb") as f: TypeError: expected str, bytes or os.PathLike object, not NoneType [ERROR] 2024-09-05-05:54:50 (PID:5791, Device:-1, RankID:-1) ERR99999 UNKNOWN application exception