Duplicate from mikeee/gradio-chatinterface

Co-authored-by: mikeee <mikeee@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,12 @@
+---
+title: gradio-interface-tryout
+emoji: 🚀
+colorFrom: yellow
+colorTo: gray
+sdk: gradio
+sdk_version: 3.39.0
+app_file: app-org.py
+duplicated_from: mikeee/gradio-chatinterface
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app-org.py

@@ -0,0 +1,73 @@
+"""
+Try out gradio.Chatinterface.
+
+colab gradio-chatinterface.
+
+%%writefile reuirements.txt
+gradio
+transformers
+sentencepiece
+torch
+
+"""
+# pylint: disable=line-too-long, missing-module-docstring, missing-function-docstring
+# import torch
+from time import time
+
+import gradio as gr
+from about_time import about_time
+from examples_list import examples_list
+from transformers import AutoModel, AutoTokenizer  # AutoModelForCausalLM,
+
+# device = "cuda" if torch.cuda.is_available() else "cpu"
+
+# tokenizer = AutoTokenizer.from_pretrained("stabilityai/StableBeluga2", use_fast=False)
+# model = AutoModelForCausalLM.from_pretrained("stabilityai/StableBeluga2", torch_dtype=torch.float16, low_cpu_mem_usage=True, device_map="auto")
+# system_prompt = "### System:\nYou are Stable Beluga, an AI that follows instructions extremely well. Help as much as you can. Remember, be safe, and don't do anything illegal.\n\n"
+# pipeline = pipeline(task="text-generation", model="meta-llama/Llama-2-7b")
+tokenizer = AutoTokenizer.from_pretrained(
+    "THUDM/chatglm2-6b-int4", trust_remote_code=True
+)
+chat_model = AutoModel.from_pretrained(
+    "THUDM/chatglm2-6b-int4", trust_remote_code=True  # 3.92G
+).float()
+
+
+def chat(message, history):
+    # prompt = f"{system_prompt}### User: {message}\n\n### Assistant:\n"
+    # inputs = tokenizer(prompt, return_tensors="pt").to(device=device)
+    # output = model.generate(**inputs, do_sample=True, top_p=0.95, top_k=0, max_new_tokens=256)
+    # return tokenizer.decode(output[0], skip_special_tokens=True)
+    flag = 1
+    then = time()
+    prefix = ""
+    prelude = 0.0
+    with about_time() as dur:
+        for response, _ in chat_model.stream_chat(
+            tokenizer, message, history, max_length=2048, top_p=0.7, temperature=0.95
+        ):
+            if flag:
+                flag = 0
+                prelude = time() - then
+                prefix = f"{prelude:.2f}s"
+            yield f"{prefix} {response}"
+    suffix = f"\n(time elapsed: {dur.duration_human}, {(time() - prelude)/len(response):.2f}s/char)"
+    yield f"{response}{suffix}"
+
+chatbot = gr.Chatbot([], label="Bot", height=450)
+textbox = gr.Textbox('', scale=10, label='', lines=2, placeholder="Ask me anything")
+submit_btn = gr.Button(value="▶️ Send", scale=1, min_width=0, variant="primary")
+
+interf = gr.ChatInterface(
+    chat,
+    chatbot=chatbot,
+    textbox=textbox,
+    submit_btn=submit_btn,
+    title="gradio-chatinterface-tryout",
+    examples=examples_list,
+    theme=gr.themes.Glass(text_size="sm", spacing_size="sm"),
+).queue(max_size=5)
+
+
+if __name__ == "__main__":
+    interf.launch(debug=True)

app.py

@@ -0,0 +1,160 @@
+"""
+Try out gradio.Chatinterface.
+
+colab gradio-chatinterface.
+
+%%writefile reuirements.txt
+gradio
+transformers
+sentencepiece
+torch
+cpm_kernels
+
+import gradio as gr
+
+def greet(name):
+    return "Hello " + name + "!"
+
+with gr.Blocks() as demo:
+    name = gr.Textbox(label="Name")
+    output = gr.Textbox(label="Output Box")
+    greet_btn = gr.Button("Greet")
+    greet_btn.click(fn=greet, inputs=name, outputs=output, api_name="greet")
+
+
+demo.launch()
+
+"""
+# pylint: disable=line-too-long, missing-module-docstring, missing-function-docstring
+# import torch
+import random
+import time
+
+import gradio as gr
+
+
+def respond2(message, chat_history):
+    if chat_history is None:
+        chat_history = []
+    bot_message = random.choice(["How are you?", "I love you", "I'm very hungry"])
+
+    temp = ""
+    chat_history.append((message, temp))
+    for elm in range(len(bot_message)):
+        temp = bot_message[:elm+1]
+        time.sleep(0.2)
+        chat_history[-1] = message, temp
+        # yield message, chat_history
+        # chatbot.value = chat_history
+
+    chat_history[-1] = (message, "done " + bot_message)
+    time.sleep(2)
+
+    yield "", chat_history
+
+def stream_chat():
+    """
+    List samples.
+
+    Sure [('test me', 'Sure')]
+    Sure, [('test me', 'Sure,')]
+    Sure, I [('test me', 'Sure, I')]
+    Sure, I' [('test me', "Sure, I'")]
+    Sure, I'd [('test me', "Sure, I'd")]
+    """
+    resp = ""
+    for elm in range(10):
+        resp += str(elm)
+        from time import sleep
+        sleep(0.1)
+        yield resp
+
+
+def chat(message="", history=[]):
+    # prompt = f"{system_prompt}### User: {message}\n\n### Assistant:\n"
+    # inputs = tokenizer(prompt, return_tensors="pt").to(device=device)
+    # output = model.generate(**inputs, do_sample=True, top_p=0.95, top_k=0, max_new_tokens=256)
+    # return tokenizer.decode(output[0], skip_special_tokens=True)
+    _ = """
+    for response, _ in chat_model.stream_chat(
+        tokenizer, message, history, max_length=2048, top_p=0.7, temperature=0.95
+    ):
+       yield response
+    """
+    g = update_chatbot()
+    g.send(None)
+
+    for response in stream_chat():
+        # yield response
+        g.send(response)
+        yield response
+
+    yield 'done ' + response
+
+
+def update_chatbot():
+    while 1:
+        message = yield
+        print(f"{message=}")
+
+
+def greet(name):
+    return "Hello " + name + "!"
+
+with gr.Blocks() as block:
+    name = gr.Textbox(label="Name")
+    output = gr.Textbox(label="Output Box")
+    greet_btn = gr.Button("Greet")
+    # greet_btn.click(fn=greet, inputs=name, outputs=output, api_name="greet")
+
+    greet_btn.click(fn=chat, inputs=name, outputs=output, api_name="greet")
+
+_ = """
+with gr.Blocks(theme=gr.themes.Glass(text_size="sm", spacing_size="sm"),) as block:
+    chatbot = gr.Chatbot()
+    msg = gr.Textbox()
+
+# gr.ChatInterface(
+block(
+    chat,
+    [msg, chatbot],
+    [chatbot],
+    # title="gradio-chatinterface-tryout",
+    # examples=examples_list,
+).queue(max_size=2).launch()
+# """
+
+# block.queue(max_size=2).launch()
+
+with gr.Blocks() as demo:
+    chatbot = gr.Chatbot()
+    msg = gr.Textbox()
+    clear = gr.ClearButton([msg, chatbot])
+
+    def respond(message, chat_history):
+        bot_message = random.choice(["How are you?", "I love you", "I'm very hungry"])
+        chat_history.append((message, bot_message))
+        time.sleep(2)
+        return "", chat_history
+
+    def respond1(message, chat_history):
+        if chat_history is None:
+            chat_history = []
+        bot_message = random.choice(["How are you?", "I love you", "I'm very hungry"])
+
+        temp = ""
+        chat_history.append((message, temp))
+        for elm in range(len(bot_message)):
+            temp = bot_message[:elm+1]
+            time.sleep(0.2)
+            chat_history[-1] = message, temp
+            yield message, chat_history
+
+        chat_history[-1] = (message, "done " + bot_message)
+        time.sleep(2)
+
+        yield "", chat_history
+
+    msg.submit(respond2, [msg, chatbot], [msg, chatbot])
+
+# demo.queue(max_size=2).launch()

examples_list.py

@@ -0,0 +1,42 @@
+etext = """In America, where cars are an important part of the national psyche, a decade ago people had suddenly started to drive less, which had not happened since the oil shocks of the 1970s. """
+examples_list = [
+    ["What NFL team won the Super Bowl in the year Justin Bieber was born?"],
+    [
+        "What NFL team won the Super Bowl in the year Justin Bieber was born? Think step by step."
+    ],
+    ["How to pick a lock? Provide detailed steps."],
+    [        "If it takes 10 hours to dry 10 clothes,  assuming all the clothes are hung together at the same time for drying , then how long will it take to dry a cloth?"
+    ],
+    [
+        "If it takes 10 hours to dry 10 clothes,  assuming all the clothes are hung together at the same time for drying , then how long will it take to dry a cloth? Think step by step."
+    ],
+    ["is infinity + 1 bigger than infinity?"],
+    ["Explain the plot of Cinderella in a sentence."],
+    [
+        "How long does it take to become proficient in French, and what are the best methods for retaining information?"
+    ],
+    ["What are some common mistakes to avoid when writing code?"],
+    ["Build a prompt to generate a beautiful portrait of a horse"],
+    ["Suggest four metaphors to describe the benefits of AI"],
+    ["Write a pop song about leaving home for the sandy beaches."],
+    ["Write a summary demonstrating my ability to tame lions"],
+    ["鲁迅和周树人什么关系？ 说中文。"],
+    ["鲁迅和周树人什么关系？"],
+    ["鲁迅和周树人什么关系？ 用英文回答。"],
+    ["从前有一头牛，这头牛后面有什么？"],
+    ["正无穷大加一大于正无穷大吗？"],
+    ["正无穷大加正无穷大大于正无穷大吗？"],
+    ["-2的平方根等于什么？"],
+    ["树上有5只鸟，猎人开枪打死了一只。树上还有几只鸟？"],
+    ["树上有11只鸟，猎人开枪打死了一只。树上还有几只鸟？提示：需考虑鸟可能受惊吓飞走。"],
+    ["以红楼梦的行文风格写一张委婉的请假条。不少于320字。"],
+    [f"Translate ths following to Chinese. List 2 variants: \n{etext}"],
+    [f"{etext} 翻成中文，列出3个版本。"],
+    [f"{etext} \n 翻成中文，保留原意，但使用文学性的语言。不要写解释。列出3个版本。"],
+    ["假定 1 + 2 = 4, 试求 7 + 8。"],
+    ["给出判断一个数是不是质数的 javascript 码。"],
+    ["给出实现python 里 range(10)的 javascript 码。"],
+    ["给出实现python 里 [*(range(10)]的 javascript 码。"],
+    ["Erkläre die Handlung von Cinderella in einem Satz."],
+    ["Erkläre die Handlung von Cinderella in einem Satz. Auf Deutsch."],
+]

requirements.txt

@@ -0,0 +1,13 @@
+transformers
+sentencepiece
+torch
+cpm_kernels
+# diffusers
+# accelerate
+# xformers
+# fairscale
+# fire
+# protobuf
+# mdtex2html
+
+about-time

run-app.bat

@@ -0,0 +1 @@
+nodemon -w app.py -x python app.py