ImageSubnet/test.py at main · unconst/ImageSubnet · GitHub

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
# queries local miner for images

import bittensor as bt
from protocol import TextToImage, ImageToImage
import asyncio
import torchvision.transforms as transforms
import time
from PIL import Image

wallet = bt.wallet()
dendrite = bt.dendrite(wallet=wallet)
bt.trace()


# metagraph
metagraph = bt.metagraph(5, network="finney")
metagraph.sync()
axons = metagraph.axons

myaxonid = -1 # set this to be your axon id
if myaxonid == -1:
    print("Please set myaxonid in test.py:19 to be your axon id")
    exit()

query = TextToImage(
    text="an (anime:1.2) beautiful autumn forest scenery with the wind blowing the leaves",
    negative_prompt="worst quality, nsfw, xxx",
    width=768,
    height=768,
    num_images_per_prompt=1,
    seed=-1
)

call_single_uid = dendrite(
    axons[myaxonid],
    synapse=query,
    timeout=30.0
)


# await call_single_uid

async def query_async(call_single_uid):
    corutines = [call_single_uid]
    return await asyncio.gather(*corutines)

x = asyncio.run(query_async(call_single_uid))

_img = None

for image in x[0].images:
    # Convert the raw tensor from the Synapse into a PIL image and display it.
    transforms.ToPILImage()( bt.Tensor.deserialize(image) ).show()
    if(_img is None):
        _img = image

if(_img is None):
    bt.logging.warning("No image found")
    exit()


def i2i(t2i: TextToImage, **kwargs) -> ImageToImage:
    params = {
        'text': t2i.text,
        'negative_prompt': t2i.negative_prompt,
        'height': t2i.height,
        'width': t2i.width,
        'num_images_per_prompt': t2i.num_images_per_prompt,
        'seed': 696969,
    }

    # Update the parameters with the provided kwargs
    params.update(kwargs)

    print(params)

    query = ImageToImage(**params)

    call_single_uid = dendrite(
        axons[myaxonid],
        synapse=query,
        timeout=30.0
    )

    queried_async = asyncio.run(query_async(call_single_uid))

    return queried_async[0]


def show_images(i2i_result: ImageToImage) -> None:
    for image in i2i_result.images:
        # Convert the raw tensor from the Synapse into a PIL image and display it.
        transforms.ToPILImage()( bt.Tensor.deserialize(image) ).show()

show_images(i2i(query, image=_img, similarity="high", text="an (anime:1.2) woman walking on a path in an autumn forest"))
show_images(i2i(query, image=_img, similarity="medium", text="an (anime:1.2) woman walking on a path in an autumn forest"))
show_images(i2i(query, image=_img, similarity="low", text="an (anime:1.2) woman walking on a path in an autumn forest"))