jina-ai · hanxiao · Jun 4, 2021 · Jun 4, 2021 · Jun 4, 2021
diff --git a/.github/pages/hello-world.md b/.github/pages/hello-world.md
@@ -31,7 +31,9 @@ docker run -v "$(pwd)/j:/j" jinaai/jina hello fashion --workdir /j && open j/hel
 
 
 </details>
-This downloads the Fashion-MNIST training and test dataset and tells Jina to index 60,000 images from the training set. Then it randomly samples images from the test set as queries and asks Jina to retrieve relevant results. The whole process takes about 1 minute.
+This downloads the Fashion-MNIST training and test dataset and tells Jina to index 60,000 images from the training set.
+Then it randomly samples images from the test set as queries and asks Jina to retrieve relevant results.
+The whole process takes about 1 minute.
 
 
 ## 🤖 Covid-19 Chatbot
@@ -40,14 +42,17 @@ This downloads the Fashion-MNIST training and test dataset and tells Jina to ind
 <img align="right" width="25%" src="https://github.com/jina-ai/jina/blob/master/.github/images/helloworld-chatbot.gif?raw=true" />
 </a>
 
-For NLP engineers, we provide a simple chatbot demo for answering Covid-19 questions. To run that:
+For NLP engineers, we provide a simple chatbot demo for answering Covid-19 questions.
+To run that:
 ```bash
 pip install --pre "jina[chatbot]"
 
 jina hello chatbot
 ```
 
-This downloads [CovidQA dataset](https://www.kaggle.com/xhlulu/covidqa) and tells Jina to index 418 question-answer pairs with DistilBERT. The index process takes about 1 minute on CPU. Then it opens a web page where you can input questions and ask Jina.
+This downloads [CovidQA dataset](https://www.kaggle.com/xhlulu/covidqa) and tells Jina to index 418 question-answer pairs with MPNet.
+The index process takes about 1 minute on CPU.
+Then it opens a web page where you can input questions and ask Jina.
 
 <br><br><br><br>
 
@@ -57,11 +62,16 @@ This downloads [CovidQA dataset](https://www.kaggle.com/xhlulu/covidqa) and tell
 <img align="right" width="25%" src="https://github.com/jina-ai/jina/blob/master/.github/images/helloworld-multimodal.gif?raw=true" />
 </a>
 
-A multimodal-document contains multiple data types, e.g. a PDF document often contains figures and text. Jina lets you build a multimodal search solution in just minutes. To run our minimum multimodal document search demo:
+A multimodal-document contains multiple data types, e.g. a PDF document often contains figures and text.
+Jina lets you build a multimodal search solution in just minutes.
+To run our minimum multimodal document search demo:
 ```bash
 pip install --pre "jina[multimodal]"
 
 jina hello multimodal
 ```
 
-This downloads [people image dataset](https://www.kaggle.com/ahmadahmadzada/images2000) and tells Jina to index 2,000 image-caption pairs with MobileNet and DistilBERT. The index process takes about 3 minute on CPU. Then it opens a web page where you can query multimodal documents. We have prepared [a YouTube tutorial](https://youtu.be/B_nH8GCmBfc) to walk you through this demo.
+This downloads [people image dataset](https://www.kaggle.com/ahmadahmadzada/images2000) and tells Jina to index 2,000 image-caption pairs with MobileNet and MPNet.
+The index process takes about 3 minute on CPU.
+Then it opens a web page where you can query multimodal documents.
+We have prepared [a YouTube tutorial](https://youtu.be/B_nH8GCmBfc) to walk you through this demo.
diff --git a/jina/helloworld/chatbot/executors.py b/jina/helloworld/chatbot/executors.py
@@ -12,7 +12,7 @@ class MyTransformer(Executor):
 
     def __init__(
         self,
-        pretrained_model_name_or_path: str = 'sentence-transformers/distilbert-base-nli-stsb-mean-tokens',
+        pretrained_model_name_or_path: str = 'sentence-transformers/paraphrase-mpnet-base-v2',
         base_tokenizer_model: Optional[str] = None,
         pooling_strategy: str = 'mean',
         layer_index: int = -1,

diff --git a/jina/helloworld/multimodal/executors.py b/jina/helloworld/multimodal/executors.py
@@ -27,7 +27,7 @@ class TextEncoder(Executor):
 
     def __init__(
         self,
-        pretrained_model_name_or_path: str = 'sentence-transformers/distilbert-base-nli-stsb-mean-tokens',
+        pretrained_model_name_or_path: str = 'sentence-transformers/paraphrase-mpnet-base-v2',
         base_tokenizer_model: Optional[str] = None,
         pooling_strategy: str = 'mean',
         layer_index: int = -1,

diff --git a/tests/system/chatbot/test_chatbot.py b/tests/system/chatbot/test_chatbot.py
@@ -24,7 +24,7 @@ def post_uri():
 
 @pytest.fixture
 def expected_result():
-    return '''no evidence from the outbreak that eating garlic, sipping water every 15 minutes or taking vitamin C will protect people from the new coronavirus.'''
+    return '''It's not completely up to you.'''
 
 
 @pytest.fixture(autouse=True)
@@ -43,4 +43,6 @@ def test_chatbot(payload, post_uri, expected_result):
     """Regression test for chatbot example."""
     resp = requests.post(post_uri, json=payload)
     assert resp.status_code == 200
+    print('-------------------------1111111----')
+    print(resp.text)
     assert expected_result in resp.text