我是 pytorch 世界的新手,我使用搜索和其他几个来源来摆脱 CUDA 内存错误,但运气不佳,也许这里的任何人都有解决方案。
我有以下代码并想简单地运行它:
from PIL import Image
import torch
from transformers import AutoProcessor, LlavaForConditionalGeneration
model_id = "llava-hf/llava-1.5-13b-hf"
prompt = "USER: <image>\nWhat are these?\nASSISTANT:"
image_file = "http://images.cocodataset.org/val2017/000000039769.jpg"
model = LlavaForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.float16,
low_cpu_mem_usage=True,
).to(0)
processor = AutoProcessor.from_pretrained(model_id)
raw_image = Image.open(requests.get(image_file, stream=True).raw)
inputs = processor(prompt, raw_image, return_tensors='pt').to(0, torch.float16)
output = model.generate(**inputs, max_new_tokens=200, do_sample=False)
print(processor.decode(output[0][2:], skip_special_tokens=True))
如果我启动程序,我会立即收到标准 CUDA 内存不足错误。
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.146.02 Driver Version: 535.146.02 CUDA Version: 12.2 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA GeForce RTX 4070 Ti Off | 00000000:01:00.0 Off | N/A |
| 30% 57C P0 34W / 285W | 0MiB / 12282MiB | 0% Default |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+
+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| No running processes found |
+---------------------------------------------------------------------------------------+
难道显卡真的太弱了?我无法想象,因为使用 CPU 运行脚本大约需要 20 秒?尝试了所有与批量大小清除缓存重新启动。有谁知道或可以为我指出正确的方向来运行预训练模型?
您是否已尝试使用本地图像而不是 URL 中的图像?
您还可以尝试使用较小的图像样本或较小的标记。