-
When trying to run the hello-world on docker using the command from the docs (the linux variation) or a blank pull of jinaai/jina I receive the following error:
docker: error pulling image configur…
-
I think that the Dockerfiles for Transformer encoders could be improved. Namely:
- they should all be based on a CUDA image. After all, most people using transformer models are using them with GPUs…
-
I have trained a model myself, and the structure is different from the current models, how to make my model as serviceusing jina?
-
**Issue**
Something wrong with hub images, ModuleNotFoundError found even when package has been installed inside image.
One can see in most of the hub executors that failed to bump versions that t…
-
**Describe the feature**
```
jina hub new
```
```
Select executor_type:
1 - Encoder
2 - Crafter
3 - Indexer
4 - Ranker
```
As one can see, the new type of `EvaluateExecutor` introduce…
-
**Describe the bug**
FaissIndexer could not be built. ( I think the reason why many Hub executors failed to be built with latest jina version is also related to this).
The reason is that some tes…
-
**Describe the bug**
I have been trying to update `faiss-search-example` (https://github.com/jina-ai/examples/pull/318) and I am struggling to get it right.
So the problem is related to `workspace…
-
When run 'python app.py -t index' in webqa-search example with the latest jina 0.6.8, it shows error as below. Need help to fix the problem.
Appreciate your attention.
```
(py37-jina) sz@air:/mnt…
-
Hey everyone,
we have build a very complicated executor and want to split the source code into different files/modules. I would expect that it works like for the [MongoDBIndexer](https://github.com…
-
Refactor TransformerTorch and TransformerTF tests with a matrix of parameters by using `pytest parametrize`. The matrix should contain a few popular models worth testing with different `pooling` and `…