From ef6b3ff35eab946ce99479ada2154a1fa2318ace Mon Sep 17 00:00:00 2001
From: AdamBear <adamxiong@gmail.com>
Date: Wed, 22 Mar 2023 15:11:21 +0800
Subject: [PATCH 1/3] support stream chat

---
 web_demo2.py | 29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

diff --git a/web_demo2.py b/web_demo2.py
index 6946a15..fafeb5d 100644
--- a/web_demo2.py
+++ b/web_demo2.py
@@ -11,8 +11,8 @@ st.set_page_config(
 
 @st.cache_resource
 def get_model():
-    tokenizer = AutoTokenizer.from_pretrained("/THUDM/chatglm-6b", trust_remote_code=True)
-    model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
+    tokenizer = AutoTokenizer.from_pretrained("/data/chatglm-6b", trust_remote_code=True)
+    model = AutoModel.from_pretrained("/data/chatglm-6b", trust_remote_code=True).half().cuda()
     model = model.eval()
     return tokenizer, model
 
@@ -25,26 +25,31 @@ def predict(input, history=None):
     tokenizer, model = get_model()
     if history is None:
         history = []
-    response, history = model.chat(tokenizer, input, history)
 
-    #updates = []
-    for i, (query, response) in enumerate(history):
-        #updates.append("用户：" + query)
-        message(query, avatar_style="big-smile", key=str(i) + "_user")
-        #updates.append("ChatGLM-6B：" + response)
-        message(response, avatar_style="bottts", key=str(i))
+    with container:
+        if len(history) > 0:
+            for i, (query, response) in enumerate(history):
+                message(query, avatar_style="big-smile", key=str(i) + "_user")
+                message(response, avatar_style="bottts", key=str(i))
 
-    # if len(updates) < MAX_BOXES:
-    #     updates = updates + [""] * (MAX_BOXES - len(updates))
+        message(input, avatar_style="big-smile", key=str(len(history)) + "_user")
+        st.write("AI正在回复:")
+        with st.empty():
+            for response, history in model.stream_chat(tokenizer, input, history):
+                query, response = history[-1]
+                st.write(response)
 
     return history
 
 
+container = st.container()
+
 # create a prompt text for the text generation
 prompt_text = st.text_area(label="用户命令输入",
             height = 100,
             placeholder="请在这儿输入您的命令")
 
+
 if 'state' not in st.session_state:
     st.session_state['state'] = []
 
@@ -53,4 +58,4 @@ if st.button("发送", key="predict"):
         # text generation
         st.session_state["state"] = predict(prompt_text, st.session_state["state"])
 
-    st.balloons()
\ No newline at end of file
+    st.session_state["state"]
\ No newline at end of file

From ee76342382e5894f45f44ad04bc6b047df26cce4 Mon Sep 17 00:00:00 2001
From: AdamBear <adamxiong@gmail.com>
Date: Wed, 22 Mar 2023 15:16:08 +0800
Subject: [PATCH 2/3] merge

---
 web_demo2.py | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/web_demo2.py b/web_demo2.py
index ede0872..ff78a99 100644
--- a/web_demo2.py
+++ b/web_demo2.py
@@ -11,13 +11,8 @@ st.set_page_config(
 
 @st.cache_resource
 def get_model():
-<<<<<<< HEAD
-    tokenizer = AutoTokenizer.from_pretrained("/data/chatglm-6b", trust_remote_code=True)
-    model = AutoModel.from_pretrained("/data/chatglm-6b", trust_remote_code=True).half().cuda()
-=======
     tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
     model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
->>>>>>> 292b5684c18a1765bf2570739e5cf2dcb03cca41
     model = model.eval()
     return tokenizer, model
 
@@ -31,7 +26,6 @@ def predict(input, history=None):
     if history is None:
         history = []
 
-<<<<<<< HEAD
     with container:
         if len(history) > 0:
             for i, (query, response) in enumerate(history):
@@ -45,12 +39,6 @@ def predict(input, history=None):
                 query, response = history[-1]
                 st.write(response)
 
-=======
-    for i, (query, response) in enumerate(history):
-        message(query, avatar_style="big-smile", key=str(i) + "_user")
-        message(response, avatar_style="bottts", key=str(i))
-
->>>>>>> 292b5684c18a1765bf2570739e5cf2dcb03cca41
     return history
 
 

From 5513dd7d2c61023adb5a48d4b723e8ff35163837 Mon Sep 17 00:00:00 2001
From: duzx16 <zx-du20@mails.tsinghua.edu.cn>
Date: Thu, 23 Mar 2023 14:46:05 +0800
Subject: [PATCH 3/3] Delete session_state show

---
 web_demo2.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/web_demo2.py b/web_demo2.py
index ff78a99..4e1f0e4 100644
--- a/web_demo2.py
+++ b/web_demo2.py
@@ -57,5 +57,3 @@ if st.button("发送", key="predict"):
     with st.spinner("AI正在思考，请稍等........"):
         # text generation
         st.session_state["state"] = predict(prompt_text, st.session_state["state"])
-
-    st.session_state["state"]
\ No newline at end of file