Spaces:
Running
Running
update
Browse files- Dockerfile +1 -1
- README.md +12 -6
- app.py +61 -22
- polos/models/__init__.py +4 -7
- polos/models/__pycache__/__init__.cpython-39.pyc +0 -0
- polos/models/encoders/__pycache__/xlmr.cpython-39.pyc +0 -0
- polos/models/encoders/xlmr.py +1 -4
Dockerfile
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
FROM python:3.9.2-slim
|
2 |
RUN apt-get update && apt-get install -y gcc g++
|
3 |
WORKDIR /app
|
4 |
|
|
|
1 |
+
FROM --platform=linux/amd64 python:3.9.2-slim
|
2 |
RUN apt-get update && apt-get install -y gcc g++
|
3 |
WORKDIR /app
|
4 |
|
README.md
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
---
|
2 |
-
title:
|
3 |
-
emoji:
|
4 |
colorFrom: blue
|
5 |
colorTo: green
|
6 |
sdk: docker
|
@@ -8,8 +8,14 @@ app_port: 8501
|
|
8 |
pinned: false
|
9 |
---
|
10 |
|
11 |
-
##
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
-
|
14 |
-
|
15 |
-
|
|
|
|
1 |
---
|
2 |
+
title: Polos Demo
|
3 |
+
emoji: 🌟
|
4 |
colorFrom: blue
|
5 |
colorTo: green
|
6 |
sdk: docker
|
|
|
8 |
pinned: false
|
9 |
---
|
10 |
|
11 |
+
## Get Started on M1 Mac
|
12 |
+
```bash
|
13 |
+
git submodule update --init --recursive
|
14 |
+
docker build . -t polos_demo
|
15 |
+
docker run -it -d -v `pwd`:/workspace -p 8080:8080 --platform linux/amd64 polos_demo
|
16 |
+
docker exec -it $process_id bash
|
17 |
|
18 |
+
root@28cb354f7609:~# sh install.sh
|
19 |
+
root@28cb354f7609:~# poetry run python test.py
|
20 |
+
root@28cb354f7609:~# poetry run streamlit run test.py --server.port 8080
|
21 |
+
```
|
app.py
CHANGED
@@ -2,7 +2,8 @@ import streamlit as st
|
|
2 |
from PIL import Image
|
3 |
from polos.models import download_model, load_checkpoint
|
4 |
|
5 |
-
|
|
|
6 |
def load_model():
|
7 |
model_path = download_model("polos")
|
8 |
model = load_checkpoint(model_path)
|
@@ -10,29 +11,67 @@ def load_model():
|
|
10 |
|
11 |
model = load_model()
|
12 |
|
13 |
-
default_image = Image.open("test.jpg").convert("RGB")
|
14 |
-
default_refs = [
|
15 |
-
"there is a dog sitting on a couch with a person reaching out",
|
16 |
-
"a dog laying on a couch with a person",
|
17 |
-
'a dog is laying on a couch with a person'
|
18 |
-
]
|
19 |
-
|
20 |
-
data = [
|
21 |
-
{
|
22 |
-
"img": default_image,
|
23 |
-
"mt": "",
|
24 |
-
"refs": default_refs
|
25 |
-
}
|
26 |
-
]
|
27 |
-
|
28 |
# Streamlitインターフェースの設定
|
29 |
st.title('Polos Demo')
|
30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
# ユーザー入力のテキストフィールド
|
32 |
-
user_input = st.text_input("Enter the input sentence:",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
|
34 |
-
#
|
35 |
-
if
|
36 |
-
|
37 |
-
_, scores = model.predict(data, batch_size=1, cuda=False)
|
38 |
-
st.write("Score:", scores)
|
|
|
2 |
from PIL import Image
|
3 |
from polos.models import download_model, load_checkpoint
|
4 |
|
5 |
+
# モデルのロード
|
6 |
+
@st.cache_resource()
|
7 |
def load_model():
|
8 |
model_path = download_model("polos")
|
9 |
model = load_checkpoint(model_path)
|
|
|
11 |
|
12 |
model = load_model()
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
# Streamlitインターフェースの設定
|
15 |
st.title('Polos Demo')
|
16 |
|
17 |
+
# セッションステートの初期化
|
18 |
+
if 'image' not in st.session_state:
|
19 |
+
st.session_state.image = None
|
20 |
+
if 'user_input' not in st.session_state:
|
21 |
+
st.session_state.user_input = ''
|
22 |
+
if 'user_refs' not in st.session_state:
|
23 |
+
st.session_state.user_refs = [
|
24 |
+
"there is a dog sitting on a couch with a person reaching out",
|
25 |
+
"a dog laying on a couch with a person",
|
26 |
+
'a dog is laying on a couch with a person'
|
27 |
+
]
|
28 |
+
if 'score' not in st.session_state:
|
29 |
+
st.session_state.score = None
|
30 |
+
|
31 |
+
# デフォルト画像の取得
|
32 |
+
@st.cache_resource()
|
33 |
+
def get_default_image():
|
34 |
+
try:
|
35 |
+
return Image.open("test.jpg").convert("RGB")
|
36 |
+
except FileNotFoundError:
|
37 |
+
return Image.new('RGB', (200, 200), color = 'gray') # デフォルト画像が見つからない場合の代替画像
|
38 |
+
|
39 |
+
default_image = get_default_image()
|
40 |
+
|
41 |
+
# 画像アップロードのためのウィジェット
|
42 |
+
uploaded_image = st.file_uploader("Upload your image:", type=["jpg", "jpeg", "png"])
|
43 |
+
if uploaded_image is not None:
|
44 |
+
st.session_state.image = Image.open(uploaded_image).convert("RGB")
|
45 |
+
elif st.session_state.image is None:
|
46 |
+
st.session_state.image = default_image
|
47 |
+
|
48 |
+
# 常に画像を表示
|
49 |
+
st.image(st.session_state.image, caption="Displayed Image", use_column_width=True)
|
50 |
+
|
51 |
+
# 参照文の入力フィールド
|
52 |
+
user_refs = st.text_area("Enter reference sentences (separate each by a newline):", "\n".join(st.session_state.user_refs))
|
53 |
+
st.session_state.user_refs = user_refs.split("\n")
|
54 |
+
|
55 |
# ユーザー入力のテキストフィールド
|
56 |
+
user_input = st.text_input("Enter the input sentence:", value=st.session_state.user_input)
|
57 |
+
st.session_state.user_input = user_input
|
58 |
+
|
59 |
+
# Computeボタン
|
60 |
+
if st.button('Compute'):
|
61 |
+
# データの準備
|
62 |
+
data = [
|
63 |
+
{
|
64 |
+
"img": st.session_state.image,
|
65 |
+
"mt": st.session_state.user_input,
|
66 |
+
"refs": st.session_state.user_refs
|
67 |
+
}
|
68 |
+
]
|
69 |
+
|
70 |
+
# モデル予測
|
71 |
+
if st.session_state.user_input:
|
72 |
+
_, scores = model.predict(data, batch_size=1, cuda=False)
|
73 |
+
st.session_state.score = scores[0]
|
74 |
|
75 |
+
# スコアの表示
|
76 |
+
if st.session_state.score is not None:
|
77 |
+
st.metric(label="Score", value=f"{st.session_state.score:.5f}")
|
|
|
|
polos/models/__init__.py
CHANGED
@@ -19,13 +19,10 @@ str2model = {
|
|
19 |
}
|
20 |
|
21 |
def get_cache_folder():
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
return cache_directory
|
27 |
-
else:
|
28 |
-
raise Exception("HOME environment variable is not defined.")
|
29 |
|
30 |
def download_model(model: str, saving_directory: str = None) -> ModelBase:
|
31 |
"""Function that loads pretrained models from AWS.
|
|
|
19 |
}
|
20 |
|
21 |
def get_cache_folder():
|
22 |
+
cache_directory = "./.cache/"
|
23 |
+
if not os.path.exists(cache_directory):
|
24 |
+
os.makedirs(cache_directory)
|
25 |
+
return cache_directory
|
|
|
|
|
|
|
26 |
|
27 |
def download_model(model: str, saving_directory: str = None) -> ModelBase:
|
28 |
"""Function that loads pretrained models from AWS.
|
polos/models/__pycache__/__init__.cpython-39.pyc
CHANGED
Binary files a/polos/models/__pycache__/__init__.cpython-39.pyc and b/polos/models/__pycache__/__init__.cpython-39.pyc differ
|
|
polos/models/encoders/__pycache__/xlmr.cpython-39.pyc
CHANGED
Binary files a/polos/models/encoders/__pycache__/xlmr.cpython-39.pyc and b/polos/models/encoders/__pycache__/xlmr.cpython-39.pyc differ
|
|
polos/models/encoders/xlmr.py
CHANGED
@@ -29,10 +29,7 @@ XLMR_LARGE_V0_MODEL_NAME = "xlmr.large.v0/model.pt"
|
|
29 |
XLMR_BASE_V0_URL = "https://dl.fbaipublicfiles.com/fairseq/models/xlmr.base.v0.tar.gz"
|
30 |
XLMR_BASE_V0_MODEL_NAME = "xlmr.base.v0/model.pt"
|
31 |
|
32 |
-
|
33 |
-
saving_directory = os.environ["HOME"] + "/.cache/torch/yuigawada/"
|
34 |
-
else:
|
35 |
-
raise Exception("HOME environment variable is not defined.")
|
36 |
|
37 |
|
38 |
class XLMREncoder(Encoder):
|
|
|
29 |
XLMR_BASE_V0_URL = "https://dl.fbaipublicfiles.com/fairseq/models/xlmr.base.v0.tar.gz"
|
30 |
XLMR_BASE_V0_MODEL_NAME = "xlmr.base.v0/model.pt"
|
31 |
|
32 |
+
saving_directory = "./.cache/"
|
|
|
|
|
|
|
33 |
|
34 |
|
35 |
class XLMREncoder(Encoder):
|