我正在关注 Gradient GPU 服务器上的 fast.ai 的 jupyter 笔记本教程。在第二个笔记本
02_production.ipynb
中,search_images_bing
失败。
代码:
!pip install -Uqq fastbook
import fastbook
fastbook.setup_book()
from fastbook import *
from fastai.vision.widgets import *
key = os.environ.get('AZURE_SEARCH_KEY', 'XXX') # not showing my key :)
results = search_images_bing(key, 'grizzly bear', min_sz=128)
ims = results.attrgot('content_url')
错误:
---------------------------------------------------------------------------
ErrorResponseException Traceback (most recent call last)
<ipython-input-107-8c0a1d6b3765> in <module>
----> 1 results = search_images_bing(key, 'grizzly bear', min_sz=128)
2 ims = results.attrgot('content_url')
3
4 # ims = search_bing_by_term('grizzly bear', 100)
5 len(ims)
/opt/conda/envs/fastai/lib/python3.8/site-packages/fastbook/__init__.py in search_images_bing(key, term, min_sz)
50 def search_images_bing(key, term, min_sz=128):
51 client = api('https://api.cognitive.microsoft.com', auth(key))
---> 52 return L(client.images.search(query=term, count=150, min_height=min_sz, min_width=min_sz).value)
53
54 def plot_function(f, tx=None, ty=None, title=None, min=-2, max=2, figsize=(6,4)):
/opt/conda/envs/fastai/lib/python3.8/site-packages/azure/cognitiveservices/search/imagesearch/operations/_images_operations.py in search(self, query, accept_language, user_agent, client_id, client_ip, location, aspect, color, country_code, count, freshness, height, id, image_content, image_type, license, market, max_file_size, max_height, max_width, min_file_size, min_height, min_width, offset, safe_search, size, set_lang, width, custom_headers, raw, **operation_config)
489
490 if response.status_code not in [200]:
--> 491 raise models.ErrorResponseException(self._deserialize, response)
492
493 deserialized = None
ErrorResponseException: Operation returned an invalid status code 'PermissionDenied'
我已在 Azure/Bing.Search.v7/F1(免费套餐)上正确注册并设置 API 并获得密钥。
如何通过与 bing API 对话来克服此错误?
似乎最新版本的 Bing 搜索断开了与 fast.ai 的
search_images_bing
功能的连接。
改进此处的解决方法https://forums.fast.ai/t/02-product-permissiondenied-error/65823/25?u=retuso,我能够创建适合的搜索功能的本地实例(几乎)完美地配合笔记本其余部分的自然使用。
本地功能:
def search_images_bing(key, term, max_images: int = 100, **kwargs):
params = {'q':term, 'count':max_images}
headers = {"Ocp-Apim-Subscription-Key":key}
search_url = "https://api.bing.microsoft.com/v7.0/images/search"
response = requests.get(search_url, headers=headers, params=params)
response.raise_for_status()
search_results = response.json()
# returns an L object to be identical to the original function.
return L(search_results['value'])
调用该函数的代码,注意新的 bing API 不使用原来的
content_url
标签,而是使用 contentUrl
:
# fits in nicely with the original code
results = search_images_bing(key, 'grizzly bear', min_sz=128)
ims = results.attrgot('contentUrl')
注意:我还没有使用 min_sz 参数。
如果这不适合你 从 fastai.vision.utils 导入 search_images_bing
然后这样做
from fastai.vision.all import *
def search_images_bing(key, term, min_sz=128, max_images=150):
params = {'q':term, 'count':max_images, 'min_height':min_sz, 'min_width':min_sz}
headers = {"Ocp-Apim-Subscription-Key":key}
search_url = "https://api.bing.microsoft.com/v7.0/images/search"
response = requests.get(search_url, headers=headers, params=params)
response.raise_for_status()
search_results = response.json()
return L(search_results['value'])