运行generate.sh后一直报model为NoneType

[JupyterChu]

操作系统： CentOS Linux 7 GPU: 3090，一张 python版本：3.8.16 库的版本和requirements.txt中一致 llama-7b-hf数据提前下好了 cuda版本：11.4 nvcc --version： nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2021 NVIDIA Corporation Built on Sun_Aug_15_21:14:11_PDT_2021 Cuda compilation tools, release 11.4, V11.4.120 Build cuda_11.4.r11.4/compiler.30300941_0

报错信息： CUDA SETUP: CUDA runtime path found: ...................../.conda/envs/py38/lib/libcudart.so CUDA SETUP: Highest compute capability among GPUs detected: 8.6 CUDA SETUP: Detected CUDA version 118 CUDA SETUP: Loading binary ...................../python3.8/site-packages/bitsandbytes/libbitsandbytes_cuda118.so... Namespace(lora_path='Chinese-Vicuna/Chinese-Vicuna-lora-7b-belle-and-guanaco', model_path='decapoda-research/llama-7b-hf', use_local=0, use_typewriter=1) Chinese-Vicuna/Chinese-Vicuna-lora-7b-belle-and-guanaco/adapter_model.bin

Loading checkpoint shards: 0%| | 0/33 [00:00<?, ?it/s] Loading checkpoint shards: 3%|▎ | 1/33 [00:02<01:20, 2.51s/it] Loading checkpoint shards: 6%|▌ | 2/33 [00:02<00:40, 1.32s/it] Loading checkpoint shards: 9%|▉ | 3/33 [00:03<00:28, 1.06it/s] Loading checkpoint shards: 12%|█▏ | 4/33 [00:03<00:22, 1.32it/s] Loading checkpoint shards: 15%|█▌ | 5/33 [00:04<00:18, 1.52it/s] Loading checkpoint shards: 18%|█▊ | 6/33 [00:04<00:16, 1.67it/s] Loading checkpoint shards: 21%|██ | 7/33 [00:05<00:14, 1.79it/s] Loading checkpoint shards: 24%|██▍ | 8/33 [00:05<00:13, 1.87it/s] Loading checkpoint shards: 27%|██▋ | 9/33 [00:06<00:12, 1.94it/s] Loading checkpoint shards: 30%|███ | 10/33 [00:06<00:11, 1.98it/s] Loading checkpoint shards: 33%|███▎ | 11/33 [00:07<00:10, 2.02it/s] Loading checkpoint shards: 36%|███▋ | 12/33 [00:07<00:10, 2.03it/s] Loading checkpoint shards: 39%|███▉ | 13/33 [00:08<00:09, 2.04it/s] Loading checkpoint shards: 42%|████▏ | 14/33 [00:08<00:09, 2.06it/s] Loading checkpoint shards: 45%|████▌ | 15/33 [00:09<00:08, 2.06it/s] Loading checkpoint shards: 48%|████▊ | 16/33 [00:09<00:08, 2.07it/s] Loading checkpoint shards: 52%|█████▏ | 17/33 [00:10<00:07, 2.07it/s] Loading checkpoint shards: 55%|█████▍ | 18/33 [00:10<00:07, 2.08it/s] Loading checkpoint shards: 58%|█████▊ | 19/33 [00:11<00:06, 2.08it/s] Loading checkpoint shards: 61%|██████ | 20/33 [00:11<00:06, 2.09it/s] Loading checkpoint shards: 64%|██████▎ | 21/33 [00:12<00:05, 2.09it/s] Loading checkpoint shards: 67%|██████▋ | 22/33 [00:12<00:05, 2.09it/s] Loading checkpoint shards: 70%|██████▉ | 23/33 [00:13<00:04, 2.10it/s] Loading checkpoint shards: 73%|███████▎ | 24/33 [00:13<00:04, 2.09it/s] Loading checkpoint shards: 76%|███████▌ | 25/33 [00:14<00:03, 2.08it/s] Loading checkpoint shards: 79%|███████▉ | 26/33 [00:14<00:03, 2.08it/s] Loading checkpoint shards: 82%|████████▏ | 27/33 [00:14<00:02, 2.09it/s] Loading checkpoint shards: 85%|████████▍ | 28/33 [00:15<00:02, 2.09it/s] Loading checkpoint shards: 88%|████████▊ | 29/33 [00:15<00:01, 2.09it/s] Loading checkpoint shards: 91%|█████████ | 30/33 [00:16<00:01, 2.09it/s] Loading checkpoint shards: 94%|█████████▍| 31/33 [00:16<00:00, 2.10it/s] Loading checkpoint shards: 97%|█████████▋| 32/33 [00:17<00:00, 2.09it/s] Loading checkpoint shards: 100%|██████████| 33/33 [00:17<00:00, 2.05it/s] Loading checkpoint shards: 100%|██████████| 33/33 [00:17<00:00, 1.84it/s]

Downloading (…)/adapter_config.json: 0%| | 0.00/370 [00:00<?, ?B/s] Downloading (…)/adapter_config.json: 100%|██████████| 370/370 [00:00<00:00, 110kB/s]

Downloading adapter_model.bin: 0%| | 0.00/16.8M [00:00<?, ?B/s] Downloading adapter_model.bin: 62%|██████▏ | 10.5M/16.8M [00:02<00:01, 4.31MB/s] Downloading adapter_model.bin: 100%|██████████| 16.8M/16.8M [00:03<00:00, 4.52MB/s] Downloading adapter_model.bin: 100%|██████████| 16.8M/16.8M [00:03<00:00, 4.47MB/s] Traceback (most recent call last): File "generate.py", line 110, in model.eval() AttributeError: 'NoneType' object has no attribute 'eval'

求助，谢谢！

[Facico]

还是peft版本的问题，见这个issue （其实类似的问题把报错信息在项目中搜一下说不定就有解决方案了）

[JupyterChu]

好的，谢谢！我试试