feat: clip as external executor

.gitignore

@@ -1,3 +1,6 @@
+# OS
+.DS_Store
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -127,3 +130,6 @@ dmypy.json
 
 # Pyre type checker
 .pyre/
+
+# Jina
+.jina/

Dockerfile

@@ -23,7 +23,7 @@ RUN apt-get update \
 RUN if [ -n "${APT_PACKAGES}" ]; then apt-get update && apt-get install --no-install-recommends -y ${APT_PACKAGES}; fi && \
     git clone --depth=1 https://github.com/JingyunLiang/SwinIR.git  && \
     git clone --depth=1 https://github.com/CompVis/latent-diffusion.git && \
-    git clone --depth=1 https://github.com/hanxiao/glid-3-xl.git && \
+    git clone --depth=1 -b pref-remove-clip-as-service https://github.com/jina-ai/glid-3-xl.git && \
     pip install jax[cuda11_cudnn82]==0.3.13 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html && \
     pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113 && \
     cd latent-diffusion && pip install --timeout=1000 -e . && cd - && \

README.md

@@ -99,6 +99,7 @@ The 16 candidates are sorted by [CLIP-as-service](https://github.com/jina-ai/cli
 ```python
 fav_id = 3
 fav = da[fav_id]
+fav.embedding = da.embedding
 fav.display()
 ```
 
@@ -243,7 +244,7 @@ mkdir dalle && cd dalle
 git clone https://github.com/jina-ai/dalle-flow.git
 git clone https://github.com/JingyunLiang/SwinIR.git
 git clone https://github.com/CompVis/latent-diffusion.git
-git clone https://github.com/hanxiao/glid-3-xl.git
+git clone https://github.com/jina-ai/glid-3-xl.git
 ```
 
 You should have the following folder structure:
@@ -314,6 +315,10 @@ Congrats! Now you should be able to [run the client](#client).
 
 You can modify and extend the server flow as you like, e.g. changing the model, adding persistence, or even auto-posting to Instagram/OpenSea. With Jina and DocArray, you can easily make DALL·E Flow [cloud-native and ready for production](https://github.com/jina-ai/jina). 
 
+### Run your own CLIP
+
+By default [`CLIPTorchEncoder`](https://hub.jina.ai/executor/gzpbl8jh) runs as an [external executor](https://docs.jina.ai/fundamentals/flow/add-executors/#external-executors).
+If want to run your own CLIP, you can do that by just changing `external: true` to `external: false` in [`flow.yml`](./flow.yml).
 
 <!-- start support-pitch -->
 ## Support

client.ipynb

@@ -259,6 +259,7 @@
     "fav_id = 3\n",
     "\n",
     "fav = da[fav_id]\n",
+    "fav.embedding = da.embedding\n",
     "\n",
     "fav.display()"
    ]