From a6ff91c64f9ba1a54d12046f75c148708fdfbbeb Mon Sep 17 00:00:00 2001
From: wxywb <xy.wang@zilliz.com>
Date: Fri, 14 Oct 2022 11:11:04 +0800
Subject: [PATCH] add the doc.

Signed-off-by: wxywb <xy.wang@zilliz.com>
---
 README.md        | 104 ++++++++++++++++++++++++++++++++++++++++++++++-
 jclip.py         |   9 ++--
 requirements.txt |   3 ++
 3 files changed, 112 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index b1bde4b..c2eb97d 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,104 @@
-# japanese-clip
+#Japanese Image-Text Retrieval Embdding with CLIP
+
+*author: David Wang*
+
+
+<br />
+
+
+
+## Description
+
+This operator extracts features for image or text with [Japanese-CLIP](https://github.com/rinnakk/japanese-clip
+) developed by [rinna Co., Ltd](https://rinna.co.jp/), which can generate embeddings for Japanese text and image by jointly training an image encoder and text encoder to maximize the cosine similarity.
+
+<br />
+
+
+## Code Example
+
+Load an image from path './teddy.jpg' to generate an image embedding. 
+
+Read the text 'スケートボードに乗っているテディベア。' to generate an text embedding. 
+
+ *Write the pipeline in simplified style*:
+
+```python
+import towhee
+
+towhee.glob('./teddy.jpg') \
+      .image_decode() \
+      .image_text_embedding.japanese_clip(model_name='clip_vit_b32', modality='image') \
+      .show()
+
+towhee.dc(["スケートボードに乗っているテディベア。"]) \
+      .image_text_embedding.japanese_clip(model_name='clip_vit_b32', modality='text') \
+      .show()
+```
+<img src="https://towhee.io/image-text-embedding/clip/raw/branch/main/vec1.png" alt="result1" style="height:20px;"/>
+<img src="https://towhee.io/image-text-embedding/clip/raw/branch/main/vec2.png" alt="result2" style="height:20px;"/>
+
+*Write a same pipeline with explicit inputs/outputs name specifications:*
+
+```python
+import towhee
+
+towhee.glob['path']('./teddy.jpg') \
+      .image_decode['path', 'img']() \
+      .image_text_embedding.japanese_clip['img', 'vec'](model_name='clip_vit_b32', modality='image') \
+      .select['img', 'vec']() \
+      .show()
+
+towhee.dc['text'](["スケートボードに乗っているテディベア。"]) \
+      .image_text_embedding.japanese_clip['text','vec'](model_name='clip_vit_b32', modality='text') \
+      .select['text', 'vec']() \
+      .show()
+```
+<img src="https://towhee.io/image-text-embedding/clip/raw/branch/main/tabular1.png" alt="result1" style="height:60px;"/>
+<img src="https://towhee.io/image-text-embedding/clip/raw/branch/main/tabular2.png" alt="result2" style="height:60px;"/>
+
+
+<br />
+
+
+
+## Factory Constructor
+
+Create the operator via the following factory method
+
+***japanese_clip(model_name, modality)***
+
+**Parameters:**
+
+​   ***model_name:*** *str*
+
+​   The model name of CLIP. Supported model names: 
+- japanese-clip-vit-b-16      
+- japanese-cloob-vit-b-16
+
+
+​   ***modality:*** *str*
+
+​   Which modality(*image* or *text*) is used to generate the embedding. 
+
+<br />
+
+
+
+## Interface
+
+An image-text embedding operator takes a [towhee image](link/to/towhee/image/api/doc) or string as input and generate an embedding in ndarray.
+
+
+**Parameters:**
+
+​	***data:*** *towhee.types.Image (a sub-class of numpy.ndarray)*  or *str*
+
+​  The data (image or text based on specified modality) to generate embedding.	
+
+
+
+**Returns:** *numpy.ndarray*
+
+​   The data embedding extracted by model.
 
diff --git a/jclip.py b/jclip.py
index c11e8bf..6943442 100644
--- a/jclip.py
+++ b/jclip.py
@@ -33,9 +33,10 @@ class Jaclip(NNOperator):
         sys.path.append(path)
         import japanese_clip as ja_clip
         sys.path.pop()
+        cfg = self._configs()[model_name]
         self.device = "cuda" if torch.cuda.is_available() else "cpu"
         self._modality = modality
-        model, preprocess = ja_clip.load("rinna/japanese-clip-vit-b-16", cache_dir="{}/weights/japanese_clip".format(path), device=self.device)
+        model, preprocess = ja_clip.load(cfg['weights'], cache_dir="{}/weights/japanese_clip".format(path), device=self.device)
         self.model = model
         self.tfms = preprocess
         self.tokenizer = ja_clip.load_tokenizer()
@@ -75,6 +76,8 @@ class Jaclip(NNOperator):
 
     def _configs(self):
         config = {}
-        config['blip_base'] = {}
-        config['blip_base']['weights'] = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base.pth'
+        config['japanese-clip-vit-b-16'] = {}
+        config['japanese-clip-vit-b-16']['weights'] = 'rinna/japanese-clip-vit-b-16'
+        config['japanese-cloob-vit-b-16'] = {}
+        config['japanese-cloob-vit-b-16']['weights'] = 'rinna/japanese-cloob-vit-b-16'
         return config
diff --git a/requirements.txt b/requirements.txt
index e69de29..ce14b4a 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -0,0 +1,3 @@
+torch
+towhee
+