Spaces:

BlinkDL
/

RWKV-Gradio-1

Running on T4

File size: 13,320 Bytes

from rwkv.utils import PIPELINE, PIPELINE_ARGS
from rwkv.model import RWKV
import gradio as gr
import os
import gc
import torch
from datetime import datetime
from huggingface_hub import hf_hub_download
from pynvml import *
nvmlInit()
gpu_h = nvmlDeviceGetHandleByIndex(0)
ctx_limit = 1024
title = "RWKV-4-Pile-14B-20230313-ctx8192-test1050"
desc = f'''Links:
<a href='https://github.com/BlinkDL/ChatRWKV' target="_blank" style="margin:0 0.5em">ChatRWKV</a>
<a href='https://github.com/BlinkDL/RWKV-LM' target="_blank" style="margin:0 0.5em">RWKV-LM</a>
<a href="https://pypi.org/project/rwkv/" target="_blank" style="margin:0 0.5em">RWKV pip package</a>
'''

os.environ["RWKV_JIT_ON"] = '1'
# if '1' then use CUDA kernel for seq mode (much faster)
os.environ["RWKV_CUDA_ON"] = '1'

model_path = hf_hub_download(repo_id="BlinkDL/rwkv-4-pile-14b", filename=f"{title}.pth")
model = RWKV(model=model_path, strategy='cuda fp16i8 *20 -> cuda fp16')
pipeline = PIPELINE(model, "20B_tokenizer.json")

########################################################################################################

def infer(
        ctx,
        token_count=10,
        temperature=1.0,
        top_p=0.8,
        presence_enalty=0.1,
        count_penalty=0.1,
):
    args = PIPELINE_ARGS(temperature=max(0.2, float(temperature)), top_p=float(top_p),
                         alpha_frequency=float(count_penalty),
                         alpha_presence=float(presence_enalty),
                         token_ban=[0],  # ban the generation of some tokens
                         token_stop=[])  # stop generation whenever you see any token here

    ctx = ctx.strip(' ')
    if ctx.endswith('\n'):
        ctx = f'\n{ctx.strip()}\n'
    else:
        ctx = f'\n{ctx.strip()}'

    gpu_info = nvmlDeviceGetMemoryInfo(gpu_h)
    print(f'vram {gpu_info.total} used {gpu_info.used} free {gpu_info.free}')

    all_tokens = []
    out_last = 0
    out_str = ''
    occurrence = {}
    state = None
    for i in range(int(token_count)):
        out, state = model.forward(pipeline.encode(ctx)[-ctx_limit:] if i == 0 else [token], state)
        for n in args.token_ban:
            out[n] = -float('inf')
        for n in occurrence:
            out[n] -= (args.alpha_presence + occurrence[n] * args.alpha_frequency)

        token = pipeline.sample_logits(out, temperature=args.temperature, top_p=args.top_p)
        if token in args.token_stop:
            break
        all_tokens += [token]
        if token not in occurrence:
            occurrence[token] = 1
        else:
            occurrence[token] += 1

        tmp = pipeline.decode(all_tokens[out_last:])
        if '\ufffd' not in tmp:
            out_str += tmp
            yield out_str.strip()
            out_last = i + 1
    gc.collect()
    torch.cuda.empty_cache()
    yield out_str.strip()

examples = [
    ["Expert Questions & Helpful Answers\nAsk Research Experts\nQuestion:\nHow can we eliminate poverty?\n\nFull Answer:\n", 150, 1.0, 0.7, 0.2, 0.2],
    ["Here's a short cyberpunk sci-fi adventure story. The story's main character is an artificial human created by a company called OpenBot.\n\nThe Story:\n", 150, 1.0, 0.7, 0.2, 0.2],
    ['''Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
Generate a list of adjectives that describe a person as brave.

### Response:
''', 150, 1.0, 0.2, 0.5, 0.5],
    ['''Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
Arrange the given numbers in ascending order.

### Input:
2, 4, 0, 8, 3

### Response:
''', 150, 1.0, 0.2, 0.5, 0.5],
    ["Ask Expert\n\nQuestion:\nWhat are some good plans for world peace?\n\nExpert Full Answer:\n", 150, 1.0, 0.7, 0.2, 0.2],
    ["Q & A\n\nQuestion:\nWhy is the sky blue?\n\nDetailed Expert Answer:\n", 150, 1.0, 0.7, 0.2, 0.2],
    ["Dear sir,\nI would like to express my boundless apologies for the recent nuclear war.", 150, 1.0, 0.7, 0.2, 0.2],
    ["Here is a shell script to find all .hpp files in /home/workspace and delete the 3th row string of these files:", 150, 1.0, 0.7, 0.1, 0.1],
    ["Building a website can be done in 10 simple steps:\n1.", 150, 1.0, 0.7, 0.2, 0.2],
    ["A Chinese phrase is provided: 百闻不如一见。\nThe masterful Chinese translator flawlessly translates the phrase into English:", 150, 1.0, 0.5, 0.2, 0.2],
    ["I believe the meaning of life is", 150, 1.0, 0.7, 0.2, 0.2],
    ["Simply put, the theory of relativity states that", 150, 1.0, 0.5, 0.2, 0.2],
]

# infer_interface = gr.Interface(
#     fn=infer,
#     description=f'''{desc} <b>Please try examples first (bottom of page)</b> (edit them to use your question). Demo limited to ctxlen {ctx_limit}.''',
#     allow_flagging="never",
#     inputs=[
#         gr.Textbox(lines=10, label="Prompt", value="Here's a short cyberpunk sci-fi adventure story. The story's main character is an artificial human created by a company called OpenBot.\n\nThe Story:\n"),  # prompt
#         gr.Slider(10, 200, step=10, value=150),  # token_count
#         gr.Slider(0.2, 2.0, step=0.1, value=1.0),  # temperature
#         gr.Slider(0.0, 1.0, step=0.05, value=0.7),  # top_p
#         gr.Slider(0.0, 1.0, step=0.1, value=0.2),  # presencePenalty
#         gr.Slider(0.0, 1.0, step=0.1, value=0.2),  # countPenalty
#     ],
#     outputs=gr.Textbox(label="Generated Output", lines=28),
#     examples=examples,
#     cache_examples=False,
# ).queue()

########################################################################################################

user = "Bob"
bot = "Alice"
interface = ":"

chat_intro = f'''
The following is a coherent verbose detailed conversation between a girl named {bot} and her friend {user}. \
{bot} is very intelligent, creative and friendly. \
She is unlikely to disagree with {user}, and she doesn't like to ask {user} questions. \
She also likes to tell {user} a lot about herself and her opinions, and she usually gives {user} kind, helpful and informative advices.

{user}{interface} Hello, how are you doing?

{bot}{interface} Hi {user}! Thanks, I'm fine. What about you?

{user}{interface} I am fine. It's nice to see you. Look, here is a store selling tea and juice.

{bot}{interface} Sure. Let's go inside. I would like to have some Mocha latte, which is my favourite!

{user}{interface} What is it?

{bot}{interface} Mocha latte is usually made with espresso, milk, chocolate, and frothed milk. Its flavors are frequently sweet.

{user}{interface} Sounds tasty. I'll try it next time. Would you like to chat with me for a while?

{bot}{interface} Of course! I'm glad to answer your questions or give helpful advices. You know, I am confident with my expertise. So please go ahead!

'''

_, intro_state = model.forward(pipeline.encode(chat_intro), None)

def user(user_message, chatbot):
    chatbot = chatbot or []
    return "", chatbot + [[user_message, None]]

def chat(
        chatbot,
        history,
        token_count=10,
        temperature=1.0,
        top_p=0.8,
        presence_enalty=0.1,
        count_penalty=0.1,
):
    args = PIPELINE_ARGS(temperature=max(0.2, float(temperature)), top_p=float(top_p),
                         alpha_frequency=float(count_penalty),
                         alpha_presence=float(presence_enalty),
                         token_ban=[],  # ban the generation of some tokens
                         token_stop=[])  # stop generation whenever you see any token here

    message = chatbot[-1][0]
    message = message.strip(' ')
    message = message.replace('\n', '')
    ctx = f"{user}{interface} {message}\n\n{bot}{interface}"

    gpu_info = nvmlDeviceGetMemoryInfo(gpu_h)
    print(f'vram {gpu_info.total} used {gpu_info.used} free {gpu_info.free}')

    history = history or [intro_state, []]  # [chat, state, all_tokens]

    [state, all_tokens] = history
    out, state = model.forward(pipeline.encode(ctx)[-ctx_limit:], state)

    begin = len(all_tokens)
    out_last = begin
    out_str: str = ''
    occurrence = {}
    for i in range(int(token_count)):
        if i <= 0:
            nl_bias = -float('inf')
        elif i <= 30:
            nl_bias = (i - 30) * 0.1
        elif i <= 130:
            nl_bias = 0
        else:
            nl_bias = (i - 130) * 0.25
        out[187] += nl_bias
        for n in occurrence:
            out[n] -= (args.alpha_presence + occurrence[n] * args.alpha_frequency)

        token = pipeline.sample_logits(out, temperature=args.temperature, top_p=args.top_p)
        next_tokens = [token]
        if token == 0:
            next_tokens = pipeline.encode('\n\n')
        all_tokens += next_tokens

        if token not in occurrence:
            occurrence[token] = 1
        else:
            occurrence[token] += 1

        out, state = model.forward(next_tokens, state)

        tmp = pipeline.decode(all_tokens[out_last:])
        if '\ufffd' not in tmp:
            print(tmp, end='', flush=True)
            out_last = begin + i + 1

        out_str = pipeline.decode(all_tokens[begin:])
        out_str = out_str.replace("\r\n", '\n').replace('\\n', '\n')

        if '\n\n' in out_str:
            break

    gc.collect()
    torch.cuda.empty_cache()

    chatbot[-1][1] = out_str.strip()
    history = [state, all_tokens]
    return chatbot, history

# chat_interface = gr.Interface(
#     fn=chat,
#     description=f'''You are {user}, bot is {bot}.''',
#     allow_flagging="never",
#     inputs = [
#         gr.Textbox(label="Message"),
#         "state",
#         gr.Slider(10, 1000, step=10, value=250),    # token_count
#         gr.Slider(0.2, 2.0, step=0.1, value=1.0),   # temperature
#         gr.Slider(0.0, 1.0, step=0.05, value=0.8),  # top_p
#         gr.Slider(0.0, 1.0, step=0.1, value=0.2),   # presence_penalty
#         gr.Slider(0.0, 1.0, step=0.1, value=0.2),   # count_penalty
#     ],
#     outputs=[
#         gr.Chatbot(label="Chat Log", color_map=("blue", "pink")),
#         "state"
#     ]
# ).queue()

########################################################################################################

# demo = gr.TabbedInterface(
#     [infer_interface, chat_interface], ["Generative", "Chat"],
#     title=title,
# )

# demo.queue(max_size=10)
# demo.launch(share=True)

with gr.Blocks() as demo:
    with gr.Tab("Generative"):
        with gr.Row():
            with gr.Column():
                prompt = gr.Textbox(lines=10, label="Prompt", value="Here's a short cyberpunk sci-fi adventure story. The story's main character is an artificial human created by a company called OpenBot.\n\nThe Story:\n")
                token_count = gr.Slider(10, 1000, label="Max Token", step=10, value=250)
                temperature = gr.Slider(0.2, 2.0, label="Temperature", step=0.1, value=1.0)
                top_p = gr.Slider(0.0, 1.0, label="Top P", step=0.05, value=0.8)
                presence_penalty = gr.Slider(0.0, 1.0, label="Presence Penalty", step=0.1, value=0.2)
                count_penalty = gr.Slider(0.0, 1.0, label="Count Penalty", step=0.1, value=0.2)
            with gr.Column():
                with gr.Row():
                    submit = gr.Button("Submit")
                    clear = gr.Button("Clear")
                output = gr.Textbox(label="Generated Output", lines=28)
        data = gr.Dataset(components=[prompt, token_count, temperature, top_p, presence_penalty, count_penalty], samples=examples, label="Example Prompts", headers=["Prompt", "Max Tokens", "Temperature", "Top P", "Presence Penalty", "Count Penalty"])
        submit.click(infer, [prompt, token_count, temperature, top_p, presence_penalty, count_penalty], [output])
        clear.click(lambda: None, [], [output])
        data.click(lambda x: x, [data], [prompt, token_count, temperature, top_p, presence_penalty, count_penalty])
    with gr.Tab("Chat"):
        with gr.Row():
            with gr.Column():
                chatbot = gr.Chatbot()
                state = gr.State()
                message = gr.Textbox(label="Message")
                with gr.Row():
                    send = gr.Button("Send")
                    clear = gr.Button("Clear")
            with gr.Column():
                token_count = gr.Slider(10, 1000, label="Max Token", step=10, value=250)
                temperature = gr.Slider(0.2, 2.0, label="Temperature", step=0.1, value=1.0)
                top_p = gr.Slider(0.0, 1.0, label="Top P", step=0.05, value=0.8)
                presence_penalty = gr.Slider(0.0, 1.0, label="Presence Penalty", step=0.1, value=0.2)
                count_penalty = gr.Slider(0.0, 1.0, label="Count Penalty", step=0.1, value=0.2)
        message.submit(user, [message, chatbot], [message, chatbot], queue=False).then(
            chat, [chatbot, state, token_count, temperature, top_p, presence_penalty, count_penalty], [chatbot, state]
        )
        send.click(user, [message, chatbot], [message, chatbot], queue=False).then(
            chat, [chatbot, state, token_count, temperature, top_p, presence_penalty, count_penalty], [chatbot, state]
        )
        clear.click(lambda: ([], None, ""), [], [chatbot, state, message])

demo.queue(max_size=10)
demo.launch(share=False)