Upload 15 files
Browse files- .gitattributes +2 -0
- app.py +123 -47
- img/.DS_Store +0 -0
- img/icon-dark.png +3 -0
- img/icon-light.png +3 -0
- img/test_img.png +0 -0
- microsofttt.py +154 -0
- pdf_with_tables/test.pdf +0 -0
- requirements.txt +7 -1
- style.css +6 -8
- theme.py +2 -2
.gitattributes
CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
img/icon-dark.png filter=lfs diff=lfs merge=lfs -text
|
37 |
+
img/icon-light.png filter=lfs diff=lfs merge=lfs -text
|
app.py
CHANGED
@@ -1,25 +1,35 @@
|
|
1 |
-
import os
|
2 |
-
import gradio
|
3 |
-
import time, asyncio
|
4 |
from theme import CustomTheme
|
5 |
-
from llama_index.llms import OpenAI
|
6 |
from llama_index import (
|
7 |
-
ServiceContext,
|
8 |
SimpleDirectoryReader,
|
9 |
-
VectorStoreIndex,
|
10 |
-
load_index_from_storage,
|
11 |
StorageContext,
|
12 |
-
set_global_service_context,
|
13 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
bot_examples = [
|
16 |
"Wie kannst du mir helfen?",
|
17 |
"Welche Sprachen sprichst du?",
|
18 |
"Wie trainiere ich meinen Bizeps?",
|
19 |
"Erstelle mir einen Trainingsplan, wenn ich nur 3 mal pro Woche trainieren kann.",
|
20 |
-
"Berechne meinen BMI, wenn ich 75kg bei 175cm Körpergröße wiege.",
|
21 |
-
"Berechne mir meinen Kaloriendefizit, wenn ich in der Woche 0,
|
22 |
-
"Berechne mir nochmal das Kaloriendefizit, wenn ich Männlich
|
23 |
"Wie wechsle ich meine Reifen?"
|
24 |
]
|
25 |
|
@@ -50,6 +60,61 @@ context_str = (
|
|
50 |
|
51 |
chat_engine = None
|
52 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
def setup_ai():
|
54 |
"""
|
55 |
Setup the AI for use with querying questions to OpenAI.
|
@@ -59,31 +124,29 @@ def setup_ai():
|
|
59 |
assigns the context_template and system_prompt used for manipulating
|
60 |
the AI responses.
|
61 |
"""
|
62 |
-
global
|
63 |
-
|
64 |
-
#
|
65 |
-
|
66 |
-
|
67 |
-
print("Building Index")
|
68 |
-
documents = SimpleDirectoryReader("data").load_data()
|
69 |
-
index = VectorStoreIndex.from_documents(documents)
|
70 |
-
index.storage_context.persist(persist_dir="storage")
|
71 |
-
else:
|
72 |
-
print("Directory does already exist")
|
73 |
-
print("Reusing index")
|
74 |
-
storage_context = StorageContext.from_defaults(persist_dir="storage")
|
75 |
-
index = load_index_from_storage(storage_context)
|
76 |
-
|
77 |
api_key = os.environ["OPENAI_API_KEY"]
|
78 |
|
79 |
-
|
|
|
|
|
|
|
80 |
|
81 |
-
|
|
|
|
|
|
|
|
|
82 |
|
83 |
-
|
84 |
-
|
|
|
|
|
85 |
)
|
86 |
-
set_global_service_context(service_context)
|
87 |
|
88 |
def response(message, history):
|
89 |
"""
|
@@ -99,18 +162,21 @@ def response(message, history):
|
|
99 |
# If we don't assign an empty list if nothing is present,
|
100 |
# then the program will-in the worst case-crash.
|
101 |
chat_history = chat_engine.chat_history if chat_engine.chat_history is not None else []
|
102 |
-
print("Sending request to ChatGPT")
|
103 |
-
response = chat_engine.stream_chat(message, chat_history)
|
104 |
|
105 |
-
|
|
|
106 |
|
107 |
-
|
108 |
-
|
|
|
|
|
109 |
output_text += token
|
110 |
yield output_text
|
111 |
|
|
|
|
|
112 |
# For debugging, just to check if the UI looks good.
|
113 |
-
def response_no_api(message, history):
|
114 |
"""
|
115 |
Returns a default message.
|
116 |
"""
|
@@ -131,20 +197,30 @@ def main():
|
|
131 |
elem_classes=["ask-button"],
|
132 |
)
|
133 |
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
143 |
|
144 |
chat_interface.queue()
|
145 |
chat_interface.launch(
|
146 |
-
inbrowser=True
|
|
|
147 |
)
|
148 |
|
|
|
149 |
if __name__ == "__main__":
|
150 |
main()
|
|
|
1 |
+
import os, gradio, torch, openai, os, fitz, asyncio, qdrant_client, time, math
|
|
|
|
|
2 |
from theme import CustomTheme
|
|
|
3 |
from llama_index import (
|
|
|
4 |
SimpleDirectoryReader,
|
|
|
|
|
5 |
StorageContext,
|
|
|
6 |
)
|
7 |
+
from llama_index.multi_modal_llms import OpenAIMultiModal
|
8 |
+
from llama_index.vector_stores.qdrant import QdrantVectorStore
|
9 |
+
from llama_index.indices.multi_modal.base import MultiModalVectorStoreIndex
|
10 |
+
from PIL import Image
|
11 |
+
from microsofttt import detect_and_crop_save_table
|
12 |
+
|
13 |
+
from torchvision import transforms
|
14 |
+
|
15 |
+
from transformers import AutoModelForObjectDetection
|
16 |
+
|
17 |
+
from llama_index.vector_stores.qdrant import QdrantVectorStore
|
18 |
+
|
19 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
20 |
+
openai.api_key = os.environ["OPENAI_API_KEY"]
|
21 |
+
|
22 |
+
image_documents: None
|
23 |
+
openai_mm_llm: None
|
24 |
|
25 |
bot_examples = [
|
26 |
"Wie kannst du mir helfen?",
|
27 |
"Welche Sprachen sprichst du?",
|
28 |
"Wie trainiere ich meinen Bizeps?",
|
29 |
"Erstelle mir einen Trainingsplan, wenn ich nur 3 mal pro Woche trainieren kann.",
|
30 |
+
"Berechne meinen BMI, wenn ich männlich bin und 75kg bei 175cm Körpergröße wiege.",
|
31 |
+
"Berechne mir meinen Kaloriendefizit, wenn ich in der Woche 0,1kg abnehmen möchte.",
|
32 |
+
"Berechne mir nochmal das Kaloriendefizit, wenn ich Männlich 18 bin.",
|
33 |
"Wie wechsle ich meine Reifen?"
|
34 |
]
|
35 |
|
|
|
60 |
|
61 |
chat_engine = None
|
62 |
|
63 |
+
def setup_db():
|
64 |
+
"""
|
65 |
+
Setup the qdrant store as well as convert PDFs with tables into images
|
66 |
+
to then use with the Microsoft Table Transformer and extract table information.
|
67 |
+
"""
|
68 |
+
if not os.path.exists("./qdrant_db"):
|
69 |
+
|
70 |
+
if not os.path.exists("./table_images"):
|
71 |
+
os.mkdir("./table_images/")
|
72 |
+
# Convert PDFs to images
|
73 |
+
for file in os.listdir("./pdf_with_tables"):
|
74 |
+
pdf_document = fitz.open("./pdf_with_tables/"+file)
|
75 |
+
|
76 |
+
for page_number in range(pdf_document.page_count):
|
77 |
+
# Get the page
|
78 |
+
page = pdf_document[page_number]
|
79 |
+
|
80 |
+
# Convert the page to an image
|
81 |
+
pix = page.get_pixmap()
|
82 |
+
|
83 |
+
# Create a Pillow Image object from the pixmap
|
84 |
+
image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
|
85 |
+
|
86 |
+
# Save the image
|
87 |
+
image.save(f"./table_images/page_{page_number + 1}_{math.floor(time.time())}.png")
|
88 |
+
|
89 |
+
pdf_document.close()
|
90 |
+
|
91 |
+
# Crop images to tables
|
92 |
+
for image in os.listdir("./table_images"):
|
93 |
+
detect_and_crop_save_table("./table_images/"+image)
|
94 |
+
# Delete old uncropped image
|
95 |
+
os.remove("./table_images/"+image)
|
96 |
+
|
97 |
+
# Read text documents and images
|
98 |
+
text_documents = SimpleDirectoryReader("./data/").load_data()
|
99 |
+
image_documents = SimpleDirectoryReader("./table_images/").load_data()
|
100 |
+
|
101 |
+
# Create the text and image databases
|
102 |
+
client = qdrant_client.QdrantClient(path="qdrant_db")
|
103 |
+
|
104 |
+
text_store = QdrantVectorStore(
|
105 |
+
client=client, collection_name="text_collection"
|
106 |
+
)
|
107 |
+
image_store = QdrantVectorStore(
|
108 |
+
client=client, collection_name="image_collection"
|
109 |
+
)
|
110 |
+
|
111 |
+
# Create a storage_context for the chatbot from the databases
|
112 |
+
storage_context = StorageContext.from_defaults(
|
113 |
+
vector_store=text_store, image_store=image_store
|
114 |
+
)
|
115 |
+
|
116 |
+
return (text_documents, image_documents, storage_context)
|
117 |
+
|
118 |
def setup_ai():
|
119 |
"""
|
120 |
Setup the AI for use with querying questions to OpenAI.
|
|
|
124 |
assigns the context_template and system_prompt used for manipulating
|
125 |
the AI responses.
|
126 |
"""
|
127 |
+
global openai_mm_llm, context_str, system_prompt, chat_engine
|
128 |
+
|
129 |
+
# Setup database
|
130 |
+
text_documents, image_documents, storage_context = setup_db()
|
131 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
132 |
api_key = os.environ["OPENAI_API_KEY"]
|
133 |
|
134 |
+
# Define the model used
|
135 |
+
openai_mm_llm = OpenAIMultiModal(
|
136 |
+
model="gpt-4-vision-preview", api_key=api_key, max_new_tokens=1500
|
137 |
+
)
|
138 |
|
139 |
+
# Give the model the storage_context
|
140 |
+
index = MultiModalVectorStoreIndex.from_documents(
|
141 |
+
documents=text_documents + image_documents,
|
142 |
+
storage_context=storage_context
|
143 |
+
)
|
144 |
|
145 |
+
# Create a chat engine from the index
|
146 |
+
chat_engine = index.as_chat_engine(
|
147 |
+
system_prompt=system_prompt,
|
148 |
+
context_str=context_str
|
149 |
)
|
|
|
150 |
|
151 |
def response(message, history):
|
152 |
"""
|
|
|
162 |
# If we don't assign an empty list if nothing is present,
|
163 |
# then the program will-in the worst case-crash.
|
164 |
chat_history = chat_engine.chat_history if chat_engine.chat_history is not None else []
|
|
|
|
|
165 |
|
166 |
+
# Send query
|
167 |
+
_response = chat_engine.stream_chat(message, chat_history)
|
168 |
|
169 |
+
# Stream chat answer
|
170 |
+
output_text: str = ""
|
171 |
+
for token in _response.response_gen:
|
172 |
+
time.sleep(0.02)
|
173 |
output_text += token
|
174 |
yield output_text
|
175 |
|
176 |
+
|
177 |
+
|
178 |
# For debugging, just to check if the UI looks good.
|
179 |
+
def response_no_api(message, history) -> str:
|
180 |
"""
|
181 |
Returns a default message.
|
182 |
"""
|
|
|
197 |
elem_classes=["ask-button"],
|
198 |
)
|
199 |
|
200 |
+
with gradio.Blocks(theme=CustomTheme(), css="style.css") as chat_interface:
|
201 |
+
gradio.Markdown(
|
202 |
+
"""<div style='display: flex; justify-content: center; align-items: center; margin-right: 12px;'>
|
203 |
+
<img width='48px' style='margin-right: 12px;' src='/file/img/icon-light.png'/>
|
204 |
+
ARNOLD
|
205 |
+
</div>""",
|
206 |
+
elem_classes=["arnold-title"]
|
207 |
+
)
|
208 |
+
gradio.ChatInterface(
|
209 |
+
fn=response,
|
210 |
+
theme=CustomTheme(),
|
211 |
+
submit_btn=submit_button,
|
212 |
+
chatbot=chatbot,
|
213 |
+
examples=bot_examples,
|
214 |
+
css="style.css",
|
215 |
+
)
|
216 |
+
|
217 |
|
218 |
chat_interface.queue()
|
219 |
chat_interface.launch(
|
220 |
+
inbrowser=True,
|
221 |
+
allowed_paths=["./img/"]
|
222 |
)
|
223 |
|
224 |
+
|
225 |
if __name__ == "__main__":
|
226 |
main()
|
img/.DS_Store
ADDED
Binary file (6.15 kB). View file
|
|
img/icon-dark.png
ADDED
Git LFS Details
|
img/icon-light.png
ADDED
Git LFS Details
|
img/test_img.png
ADDED
microsofttt.py
ADDED
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""
|
2 |
+
Microsoft Table Transformer Extension
|
3 |
+
By Neils:
|
4 |
+
https://docs.llamaindex.ai/en/stable/examples/multi_modal/multi_modal_pdf_tables.html#experiment-3-let-s-use-microsoft-table-transformer-to-crop-tables-from-the-images-and-see-if-it-gives-the-correct-answer
|
5 |
+
"""
|
6 |
+
import matplotlib.pyplot as plt
|
7 |
+
import matplotlib.patches as patches
|
8 |
+
from matplotlib.patches import Patch
|
9 |
+
import io
|
10 |
+
from PIL import Image, ImageDraw
|
11 |
+
import numpy as np
|
12 |
+
import csv
|
13 |
+
import pandas as pd
|
14 |
+
|
15 |
+
from torchvision import transforms
|
16 |
+
|
17 |
+
from transformers import AutoModelForObjectDetection
|
18 |
+
import torch
|
19 |
+
import openai
|
20 |
+
import os
|
21 |
+
import fitz
|
22 |
+
|
23 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
24 |
+
|
25 |
+
class MaxResize(object):
|
26 |
+
def __init__(self, max_size=800):
|
27 |
+
self.max_size = max_size
|
28 |
+
|
29 |
+
def __call__(self, image):
|
30 |
+
width, height = image.size
|
31 |
+
current_max_size = max(width, height)
|
32 |
+
scale = self.max_size / current_max_size
|
33 |
+
resized_image = image.resize(
|
34 |
+
(int(round(scale * width)), int(round(scale * height)))
|
35 |
+
)
|
36 |
+
|
37 |
+
return resized_image
|
38 |
+
|
39 |
+
|
40 |
+
detection_transform = transforms.Compose(
|
41 |
+
[
|
42 |
+
MaxResize(800),
|
43 |
+
transforms.ToTensor(),
|
44 |
+
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
|
45 |
+
]
|
46 |
+
)
|
47 |
+
|
48 |
+
structure_transform = transforms.Compose(
|
49 |
+
[
|
50 |
+
MaxResize(1000),
|
51 |
+
transforms.ToTensor(),
|
52 |
+
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
|
53 |
+
]
|
54 |
+
)
|
55 |
+
|
56 |
+
# load table detection model
|
57 |
+
# processor = TableTransformerImageProcessor(max_size=800)
|
58 |
+
model = AutoModelForObjectDetection.from_pretrained(
|
59 |
+
"microsoft/table-transformer-detection", revision="no_timm"
|
60 |
+
).to(device)
|
61 |
+
|
62 |
+
# load table structure recognition model
|
63 |
+
# structure_processor = TableTransformerImageProcessor(max_size=1000)
|
64 |
+
structure_model = AutoModelForObjectDetection.from_pretrained(
|
65 |
+
"microsoft/table-transformer-structure-recognition-v1.1-all"
|
66 |
+
).to(device)
|
67 |
+
|
68 |
+
|
69 |
+
# for output bounding box post-processing
|
70 |
+
def box_cxcywh_to_xyxy(x):
|
71 |
+
x_c, y_c, w, h = x.unbind(-1)
|
72 |
+
b = [(x_c - 0.5 * w), (y_c - 0.5 * h), (x_c + 0.5 * w), (y_c + 0.5 * h)]
|
73 |
+
return torch.stack(b, dim=1)
|
74 |
+
|
75 |
+
|
76 |
+
def rescale_bboxes(out_bbox, size):
|
77 |
+
width, height = size
|
78 |
+
boxes = box_cxcywh_to_xyxy(out_bbox)
|
79 |
+
boxes = boxes * torch.tensor(
|
80 |
+
[width, height, width, height], dtype=torch.float32
|
81 |
+
)
|
82 |
+
return boxes
|
83 |
+
|
84 |
+
|
85 |
+
def outputs_to_objects(outputs, img_size, id2label):
|
86 |
+
m = outputs.logits.softmax(-1).max(-1)
|
87 |
+
pred_labels = list(m.indices.detach().cpu().numpy())[0]
|
88 |
+
pred_scores = list(m.values.detach().cpu().numpy())[0]
|
89 |
+
pred_bboxes = outputs["pred_boxes"].detach().cpu()[0]
|
90 |
+
pred_bboxes = [
|
91 |
+
elem.tolist() for elem in rescale_bboxes(pred_bboxes, img_size)
|
92 |
+
]
|
93 |
+
|
94 |
+
objects = []
|
95 |
+
for label, score, bbox in zip(pred_labels, pred_scores, pred_bboxes):
|
96 |
+
class_label = id2label[int(label)]
|
97 |
+
if not class_label == "no object":
|
98 |
+
objects.append(
|
99 |
+
{
|
100 |
+
"label": class_label,
|
101 |
+
"score": float(score),
|
102 |
+
"bbox": [float(elem) for elem in bbox],
|
103 |
+
}
|
104 |
+
)
|
105 |
+
|
106 |
+
return objects
|
107 |
+
|
108 |
+
|
109 |
+
def detect_and_crop_save_table(
|
110 |
+
file_path, cropped_table_directory="./table_images/"
|
111 |
+
):
|
112 |
+
image = Image.open(file_path)
|
113 |
+
|
114 |
+
filename, _ = os.path.splitext(file_path.split("/")[-1])
|
115 |
+
|
116 |
+
if not os.path.exists(cropped_table_directory):
|
117 |
+
os.makedirs(cropped_table_directory)
|
118 |
+
|
119 |
+
# prepare image for the model
|
120 |
+
# pixel_values = processor(image, return_tensors="pt").pixel_values
|
121 |
+
pixel_values = detection_transform(image).unsqueeze(0).to(device)
|
122 |
+
|
123 |
+
# forward pass
|
124 |
+
with torch.no_grad():
|
125 |
+
outputs = model(pixel_values)
|
126 |
+
|
127 |
+
# postprocess to get detected tables
|
128 |
+
id2label = model.config.id2label
|
129 |
+
id2label[len(model.config.id2label)] = "no object"
|
130 |
+
detected_tables = outputs_to_objects(outputs, image.size, id2label)
|
131 |
+
|
132 |
+
print(f"number of tables detected {len(detected_tables)}")
|
133 |
+
|
134 |
+
for idx in range(len(detected_tables)):
|
135 |
+
# # crop detected table out of image
|
136 |
+
cropped_table = image.crop(detected_tables[idx]["bbox"])
|
137 |
+
cropped_table.save(f"./{cropped_table_directory}/{filename}_{idx}.png")
|
138 |
+
|
139 |
+
|
140 |
+
def plot_images(image_paths):
|
141 |
+
images_shown = 0
|
142 |
+
plt.figure(figsize=(16, 9))
|
143 |
+
for img_path in image_paths:
|
144 |
+
if os.path.isfile(img_path):
|
145 |
+
image = Image.open(img_path)
|
146 |
+
|
147 |
+
plt.subplot(2, 3, images_shown + 1)
|
148 |
+
plt.imshow(image)
|
149 |
+
plt.xticks([])
|
150 |
+
plt.yticks([])
|
151 |
+
|
152 |
+
images_shown += 1
|
153 |
+
if images_shown >= 9:
|
154 |
+
break
|
pdf_with_tables/test.pdf
ADDED
Binary file (39.7 kB). View file
|
|
requirements.txt
CHANGED
@@ -1,4 +1,10 @@
|
|
|
|
1 |
gradio==4.4.1
|
2 |
openai==1.1.1
|
3 |
llama-index==0.9.15
|
4 |
-
pypdf==3.17.1
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
clip @ git+https://github.com/openai/CLIP.git
|
2 |
gradio==4.4.1
|
3 |
openai==1.1.1
|
4 |
llama-index==0.9.15
|
5 |
+
pypdf==3.17.1
|
6 |
+
qdrant_client
|
7 |
+
pyMuPDF
|
8 |
+
tools
|
9 |
+
frontend
|
10 |
+
easyocr
|
style.css
CHANGED
@@ -12,19 +12,15 @@
|
|
12 |
padding: 0 !important;
|
13 |
}
|
14 |
|
15 |
-
div.gap>.stretch {
|
16 |
display: none !important;
|
17 |
-
}
|
18 |
|
19 |
div.gap.panel>div.gr-group {
|
20 |
position: absolute;
|
21 |
bottom: 0;
|
22 |
}
|
23 |
|
24 |
-
h1 {
|
25 |
-
font-size: 48px !important;
|
26 |
-
}
|
27 |
-
|
28 |
.ask-button {
|
29 |
background-color: var(--color-accent);
|
30 |
font-weight: bold;
|
@@ -39,6 +35,8 @@ div.message-wrap {
|
|
39 |
margin-bottom: 32px !important;
|
40 |
}
|
41 |
|
42 |
-
.
|
43 |
-
font-family: "
|
|
|
|
|
44 |
}
|
|
|
12 |
padding: 0 !important;
|
13 |
}
|
14 |
|
15 |
+
/*div.gap>.stretch {
|
16 |
display: none !important;
|
17 |
+
}*/
|
18 |
|
19 |
div.gap.panel>div.gr-group {
|
20 |
position: absolute;
|
21 |
bottom: 0;
|
22 |
}
|
23 |
|
|
|
|
|
|
|
|
|
24 |
.ask-button {
|
25 |
background-color: var(--color-accent);
|
26 |
font-weight: bold;
|
|
|
35 |
margin-bottom: 32px !important;
|
36 |
}
|
37 |
|
38 |
+
.arnold-title {
|
39 |
+
font-family: "Saira";
|
40 |
+
font-size: 48px !important;
|
41 |
+
text-align: center;
|
42 |
}
|
theme.py
CHANGED
@@ -7,7 +7,7 @@ class CustomTheme(Base):
|
|
7 |
|
8 |
def __init__(self):
|
9 |
super().__init__(
|
10 |
-
font=fonts.GoogleFont("
|
11 |
)
|
12 |
|
13 |
off_white = "#F0F0F0"
|
@@ -58,6 +58,6 @@ class CustomTheme(Base):
|
|
58 |
color_accent_soft_dark=accent_soft_dark,
|
59 |
border_color_accent_subdued_dark=accent_soft_dark,
|
60 |
|
61 |
-
block_radius="
|
62 |
container_radius="32px",
|
63 |
)
|
|
|
7 |
|
8 |
def __init__(self):
|
9 |
super().__init__(
|
10 |
+
font=(fonts.GoogleFont("Inter"), fonts.GoogleFont("Saira"))
|
11 |
)
|
12 |
|
13 |
off_white = "#F0F0F0"
|
|
|
58 |
color_accent_soft_dark=accent_soft_dark,
|
59 |
border_color_accent_subdued_dark=accent_soft_dark,
|
60 |
|
61 |
+
block_radius="16px",
|
62 |
container_radius="32px",
|
63 |
)
|