Spaces:
Sleeping
Sleeping
File size: 4,696 Bytes
513882d 0b3ce2d 513882d 8765c2a 513882d 0b3ce2d 513882d 8765c2a 513882d 8765c2a 513882d 8765c2a 513882d 2623a01 513882d 0b3ce2d 513882d 0b3ce2d 513882d 8765c2a 513882d 8765c2a 513882d 0b3ce2d |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 |
import streamlit as st
import pandas as pd
import re
from openai import OpenAI
import concurrent.futures
import json
import os
def extract_and_parse_json_from_markdown(markdown_text: str) -> dict:
code_block_pattern = r"```(?:json)?\s*([\s\S]*?)```"
code_block_match = re.search(code_block_pattern, markdown_text)
if code_block_match:
json_str = code_block_match.group(1).strip()
else:
json_str = markdown_text.strip()
try:
return json.loads(json_str)
except json.JSONDecodeError as e:
raise ValueError(f"Invalid JSON format: {e}")
def process_event(event):
openai = OpenAI(
api_key=os.environ.get('DEEP_API_KEY'),
base_url="https://api.deepinfra.com/v1/openai",
)
llm_prompt = f"""
You are a digital marketing campaign analyst designed to analyze and report digital marketing campaign data for Rod Wave concerts. Your job is to convert the given text into JSON.
Don't make any assumptions; if a value doesn't exist, consider it as zero.
{{
"market": "str",
"total_spend": "float",
"impressions": "float",
"clicks": "float",
"metrics_cpc": "float",
"metrics_cpm": "float",
"metrics_ctr": "float",
"metrics_cpa": "float",
"platform_spend_meta_total": "float",
"platform_spend_meta_instagram": "float",
"platform_spend_meta_facebook": "float",
"platform_spend_google_total": "float",
"platform_spend_google_youtube": "float",
"platform_spend_google_search_display": "float",
"platform_spend_programmatic": "float",
"revenue_average_ticket_price": "float",
"revenue_total_revenue": "float",
"revenue_roi": "float"
}}
Here is the text for it:
{event}
Return in only JSON adhering to the above schema.
"""
# Attempt to process the event and validate JSON
for attempt in range(2): # Try twice
chat_completion = openai.chat.completions.create(
model="Qwen/Qwen2.5-Coder-32B-Instruct",
messages=[{"role": "user", "content": llm_prompt}],
)
json_output = chat_completion.choices[0].message.content
try:
return extract_and_parse_json_from_markdown(json_output)
except ValueError:
if attempt == 0:
st.warning("JSON validation failed, retrying...")
else:
st.error("Failed to validate JSON after retrying.")
return None # Return None if it fails after retrying
def process_all_events(events):
json_all = []
progress_bar = st.progress(0)
with concurrent.futures.ThreadPoolExecutor(max_workers=8) as executor:
futures = [executor.submit(process_event, event) for event in events]
for i, future in enumerate(concurrent.futures.as_completed(futures)):
progress = (i + 1) / len(events)
progress_bar.progress(progress)
json_all.append(future.result())
return json_all
def main():
st.title("Rod Wave Concert Marketing Data Processor")
input_method = st.radio("Choose input method:", ["Text Area", "File Upload"])
text = None
if input_method == "Text Area":
text = st.text_area("Enter concert marketing data:", height=300)
else:
uploaded_file = st.file_uploader("Choose a text file", type="txt")
if uploaded_file is not None:
text = uploaded_file.read().decode("utf-8")
if text:
events = re.split(r'\n(?=Rod Wave Concert)', text)
events = [event for event in events if event.strip()]
st.write(f"Found **{len(events)}** events to process")
if st.button("Process Data"):
with st.spinner("Processing events..."):
json_all = process_all_events(events)
json_sanity = []
for ele in json_all:
if ele is not None: # Only process valid JSON
json_sanity.append(ele)
st.subheader("Processed JSON for Event")
st.json(ele)
df = pd.DataFrame(json_sanity)
st.success("Processing complete!")
st.write("Preview of processed data:")
st.dataframe(df.head())
csv = df.to_csv(index=False)
st.download_button(
label="Download CSV",
data=csv,
file_name="processed_concert_data.csv",
mime="text/csv"
)
if __name__ == "__main__":
main() |