Hugging Face's logo

Spaces:
BlinkDL
/
RWKV-Gradio-1
Running on T4

App Files Community
4
RWKV-Gradio-1
File size: 4,152 Bytes 
4a2c956
7c790c0
 
 
61b9ff7
 
 
4a2c956
db3b05f
 
 
 
 
7c790c0
4a2c956
7c790c0
 
 
 
2d6fa0c
db3b05f
2d6fa0c
 
7c790c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61b9ff7
 
 
7c790c0
 
 
 
 
 
2d6fa0c
7c790c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
 
import gradio as gr
import os
from datetime import datetime
from huggingface_hub import hf_hub_download
from pynvml import *
nvmlInit()
gpu_h = nvmlDeviceGetHandleByIndex(0)

title = "RWKV-4 14B fp16"
desc = '''DEMO limited to ctxlen 824, and slow because A10g does not have enough VRAM (some layers are computed on CPU instead). 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'
os.environ["RWKV_CUDA_ON"] = '1' # if '1' then use CUDA kernel for seq mode (much faster)

from rwkv.model import RWKV
model_path = hf_hub_download(repo_id="BlinkDL/rwkv-4-pile-14b", filename="RWKV-4-Pile-14B-20230213-8019.pth")
model = RWKV(model=model_path, strategy='cuda fp16 *30 -> cpu fp32')
# model_path = hf_hub_download(repo_id="BlinkDL/rwkv-4-pile-169m", filename="RWKV-4-Pile-169M-20220807-8023.pth")
# model = RWKV(model=model_path, strategy='cuda fp16')
from rwkv.utils import PIPELINE, PIPELINE_ARGS
pipeline = PIPELINE(model, "20B_tokenizer.json")

def infer(
        ctx,
        token_count=10,
        temperature=1.0,
        top_p=0.85,
        presencePenalty = 0.1,
        countPenalty = 0.1,
):
    args = PIPELINE_ARGS(temperature = max(0.2, float(temperature)), top_p = float(top_p),
                     alpha_frequency = countPenalty,
                     alpha_presence = presencePenalty,
                     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)[:824] 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
    yield out_str.strip()

examples = [
    ["Ask Expert\n\nQuestion:\nWhat are some good plans for world peace?\n\nExpert Full Answer:\n", 100, 1.0, 0.85, 0.1, 0.1],
    ["Q & A\n\nQuestion:\nWhy is the sky blue?\n\nDetailed Expert Answer:\n", 100, 1.0, 0.85, 0.1, 0.1],
    ["Expert Questions & Helpful Answers\nAsk Research Experts\nQuestion:\nCan you write a short story about an elf maiden named Julia that meets a warrior named Rallio and they go on an adventure together?\n\nFull Answer:\n", 100, 1.0, 0.85, 0.1, 0.1],
]


iface = gr.Interface(
    fn=infer,
    description=f'''{desc}''',
    allow_flagging="never",
    inputs=[
        gr.Textbox(lines=20, label="Prompt"),  # prompt
        gr.Slider(10, 200, step=10, value=100),  # 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=0.85),  # top_p
        gr.Slider(0.0, 1.0, step=0.1, value=0.1),  # presencePenalty
        gr.Slider(0.0, 1.0, step=0.1, value=0.1),  # countPenalty
    ],
    outputs=gr.Textbox(label="Generated Output", lines=35),
    examples=examples,
    cache_examples=False,
).queue()

demo = gr.TabbedInterface(
    [iface], ["Generative"],
    title=title,
)

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