clip-as-service
clip-as-service copied to clipboard
🏄 Scalable embedding, reasoning, ranking for images and sentences with CLIP
If the user encodes a document with uri and no blob, the current implementation will load blob in the client and send it to the server. This pr saves the...
When the doc to be encoded is broken (contains broken URIs, incomplete files, etc.), the server will raise an error and the whole encoding process is thus stopped. This is...
How can I specify what model to use. I have seen the documentation specifying which model is supported for pytorch, onnx and tensorrt, but there is no documentation for changing...
When using prebuilt docker images and demo code, such problems raised: code: > from clip_client import Client > clip_client = Client('grpc://0.0.0.0:51009') > r = clip_client.encode(['First do it', 'then do it...
如果我想使用bert第i层的输出作为后续网络的输入,并且仅微调bert的第i层,应该如何做?
Hi, I have products images and its description and want to generate product embeddings using the image and description for each of the products. Can CLIP can help to generate...
Hi, thanks for the fantastic framework; it helps me greatly in my projects. I want to ask if this framework supports micro batching like bentoml’s here https://docs.bentoml.org/en/0.13-lts/guides/micro_batching.html. I think it...
Hi, i guss this is a comon case, if we have structured data like product name, category and the image url, the recommendation would be generate one vector per field?...