使用python調(diào)用pytorch中的clip模型時報(bào)錯：AttributeError: partially initialized module ‘clip’ has no attribute ‘load’ (most likely due to a circular import)

現(xiàn)象

clip源碼及官方文檔

在使用pyton調(diào)用pytorch中的clip模型獲取圖片的關(guān)鍵詞時，根據(jù)官方文檔的代碼描述，代碼內(nèi)容如下

import torch
import clip
from PIL import Image
device = "cuda" if torch.cuda.is_available() else "cpu"
model, preprocess = clip.load("ViT-B/32", device=device)
image = preprocess(Image.open("C:/Users/HLY/Desktop/3.jpeg")).unsqueeze(0).to(device)
text = clip.tokenize(["a diagram", "a dog", "a cat"]).to(device)
with torch.no_grad():
    image_features = model.encode_image(image)
    text_features = model.encode_text(text)
    logits_per_image, logits_per_text = model(image, text)
    probs = logits_per_image.softmax(dim=-1).cpu().numpy()
print("Label probs:", probs)  # prints: [[0.9927937  0.00421068 0.00299572]]

根據(jù)官方文檔安裝pytorch，clip等版本后，報(bào)錯，報(bào)錯信息如下

Traceback (most recent call last):
  File "D:\WorkTest\pythonProject\resnetTest\clip.py", line 2, in <module>
    import clip
  File "D:\WorkTest\pythonProject\resnetTest\clip.py", line 6, in <module>
    model, preprocess = clip.load("ViT-B/32", device=device)
                        ^^^^^^^^^
AttributeError: partially initialized module 'clip' has no attribute 'load' (most likely due to a circular import)

解決方案

一、查看項(xiàng)目中是否有為clip名的文件

如果項(xiàng)目中有clip.py名文件，import clip會默認(rèn)為引入自己，則會出現(xiàn)上述所說循環(huán)導(dǎo)入的問題。將clip.py文件更名即可

二、查看clip是否安裝成功

如果上述方案沒有解決，查看是否為clip安裝的問題

1、先調(diào)用如下命令，卸載已安裝的clip

pip uninstall clip

2、再根據(jù)官方文檔的介紹，重新安裝clip

3、如果重新安裝后也無效，嘗試卸載后以如下命令

pip install openai-clip

到此這篇關(guān)于python調(diào)用pytorch的clip模型時報(bào)錯原因及解決方法的文章就介紹到這了,更多相關(guān)python調(diào)用pytorch報(bào)錯內(nèi)容請搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家！