Spaces:
Running
Running
Commit
Β·
727cda5
1
Parent(s):
3708509
simplify
Browse files- .gitignore +1 -0
- pages/1_π_Explorer.py β 1_π_Explorer.py +32 -13
- README.md +3 -5
- _π_README.py +0 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
unzipped_dataset/
|
pages/1_π_Explorer.py β 1_π_Explorer.py
RENAMED
|
@@ -1,16 +1,41 @@
|
|
| 1 |
-
from huggingface_hub import
|
| 2 |
import streamlit as st
|
| 3 |
import pandas as pd
|
| 4 |
import matplotlib.pyplot as plt
|
| 5 |
import os
|
|
|
|
|
|
|
| 6 |
|
| 7 |
st.set_page_config(layout="wide")
|
| 8 |
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
|
| 15 |
st.html("""
|
| 16 |
<style>
|
|
@@ -31,13 +56,6 @@ td, th {
|
|
| 31 |
""")
|
| 32 |
|
| 33 |
|
| 34 |
-
@st.cache_resource(show_spinner="Loading dataset (can take ~3 min)")
|
| 35 |
-
def load_dataset_1():
|
| 36 |
-
return snapshot_download(repo_id="reducto/rd-tablebench", repo_type="dataset")
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
dataset = load_dataset_1()
|
| 40 |
-
|
| 41 |
df = pd.read_csv(results)
|
| 42 |
|
| 43 |
if "current_index" not in st.session_state:
|
|
@@ -121,6 +139,7 @@ with col2:
|
|
| 121 |
image_path = f"{dataset}/_images/{row['pdf_path'].replace('.pdf', '.jpg')}"
|
| 122 |
st.image(image_path, use_column_width=True)
|
| 123 |
|
|
|
|
| 124 |
st.subheader("Groundtruth")
|
| 125 |
st.html(f"{dataset}/groundtruth/{row['pdf_path'].replace('.pdf', '.html')}")
|
| 126 |
|
|
|
|
| 1 |
+
from huggingface_hub import hf_hub_download
|
| 2 |
import streamlit as st
|
| 3 |
import pandas as pd
|
| 4 |
import matplotlib.pyplot as plt
|
| 5 |
import os
|
| 6 |
+
import zipfile
|
| 7 |
+
import shutil
|
| 8 |
|
| 9 |
st.set_page_config(layout="wide")
|
| 10 |
|
| 11 |
+
with st.spinner("Downloading dataset"):
|
| 12 |
+
results = hf_hub_download(
|
| 13 |
+
repo_id="reducto/rd-tablebench",
|
| 14 |
+
filename="rd-tablebench.zip",
|
| 15 |
+
repo_type="dataset",
|
| 16 |
+
)
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
def unzip_dataset():
|
| 20 |
+
if not os.path.exists("unzipped_dataset"):
|
| 21 |
+
os.makedirs("unzipped_dataset")
|
| 22 |
+
with st.spinner("Unzipping dataset"):
|
| 23 |
+
with zipfile.ZipFile(results, "r") as zip_ref:
|
| 24 |
+
zip_ref.extractall("unzipped_dataset")
|
| 25 |
+
return "unzipped_dataset/rd-tablebench"
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
if st.button("Redo Unzip"):
|
| 29 |
+
if os.path.exists("unzipped_dataset"):
|
| 30 |
+
shutil.rmtree("unzipped_dataset")
|
| 31 |
+
st.rerun()
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
dataset = unzip_dataset()
|
| 35 |
+
|
| 36 |
+
results = f"{dataset}/providers/scores.csv"
|
| 37 |
+
|
| 38 |
+
assert os.path.exists(results)
|
| 39 |
|
| 40 |
st.html("""
|
| 41 |
<style>
|
|
|
|
| 56 |
""")
|
| 57 |
|
| 58 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
df = pd.read_csv(results)
|
| 60 |
|
| 61 |
if "current_index" not in st.session_state:
|
|
|
|
| 139 |
image_path = f"{dataset}/_images/{row['pdf_path'].replace('.pdf', '.jpg')}"
|
| 140 |
st.image(image_path, use_column_width=True)
|
| 141 |
|
| 142 |
+
st.write(row)
|
| 143 |
st.subheader("Groundtruth")
|
| 144 |
st.html(f"{dataset}/groundtruth/{row['pdf_path'].replace('.pdf', '.html')}")
|
| 145 |
|
README.md
CHANGED
|
@@ -5,10 +5,8 @@ colorFrom: green
|
|
| 5 |
colorTo: gray
|
| 6 |
sdk: streamlit
|
| 7 |
sdk_version: 1.39.0
|
| 8 |
-
app_file:
|
| 9 |
pinned: false
|
| 10 |
-
license:
|
| 11 |
short_description: Reducto's SOTA human annotated table benchmark.
|
| 12 |
-
---
|
| 13 |
-
|
| 14 |
-
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
|
| 5 |
colorTo: gray
|
| 6 |
sdk: streamlit
|
| 7 |
sdk_version: 1.39.0
|
| 8 |
+
app_file: 1_π_Explorer.py
|
| 9 |
pinned: false
|
| 10 |
+
license: cc-by-nc-nd-4.0
|
| 11 |
short_description: Reducto's SOTA human annotated table benchmark.
|
| 12 |
+
---
|
|
|
|
|
|
_π_README.py
DELETED
|
File without changes
|