mgbam commited on
Commit
366223a
Β·
verified Β·
1 Parent(s): a7d25a1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +102 -15
README.md CHANGED
@@ -9,28 +9,115 @@ app_file: app.py
9
  pinned: false
10
  ---
11
 
12
- # πŸš€ BizIntel AI Ultra
13
 
14
- **Next-gen Business Analytics App powered by Google ADK, Gemini Pro, and Streamlit**
 
 
15
 
16
- ...
 
 
 
 
 
 
17
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
 
 
 
19
 
 
 
 
 
20
 
21
- # πŸš€ BizIntel AI Ultra
22
 
23
- **Next-gen Business Analytics App powered by Google ADK, Gemini Pro, and Streamlit**
24
 
25
- ## Features
26
- - Upload CSV or connect to databases (PostgreSQL, MySQL, MSSQL, Oracle, SQLite)
27
- - ADK multi-agent orchestration: trend analysis, forecasting, strategic insights
28
- - Interactive Streamlit dashboard with visual charts
29
- - Hugging Face Spaces Docker-based deployment
30
 
31
- ## Usage
32
- 1. Select your data source: file upload or SQL DB connection
33
- 2. Run agent analysis pipeline
34
- 3. View KPIs, charts, forecasts, and strategy
 
 
 
 
 
 
 
 
 
 
 
35
 
36
- Built with ❀️ using Streamlit + Gemini + ADK
 
 
9
  pinned: false
10
  ---
11
 
12
+ # πŸ“Š BizIntel AI Ultra – Advanced Analytics & Agentic BI
13
 
14
+ **BizIntel AI Ultra** turns any data tableβ€”whether a quick CSV upload or a live
15
+ database tableβ€”into an interactive **insight & forecasting workspace** powered
16
+ by:
17
 
18
+ * **Google Gemini 1.5 Pro** β‡’ natural-language strategy briefs
19
+ * **ARIMA time-series pipeline** β‡’ 90-step (or 3-period) forecasts with
20
+ model explainability, back-testing, residual ACF & CI bands
21
+ * **Plotly Darkβ„’** trend, forecast & EDA visuals
22
+ * **Streamlit** front-end running inside a Hugging Face Space
23
+ * **Modular tools** (`csv_parser`, `plot_generator`, `forecaster`,
24
+ `visuals`, `db_connector`) for quick reuse in your own projects
25
 
26
+ ---
27
+
28
+ ## πŸš€ Live Demo (Hugging Face Space)
29
+
30
+ <https://huggingface.co/spaces/your-handle/BizIntel_AI>
31
+
32
+ Upload a CSV or Excel (≀ 500 MB) **or** connect to PostgreSQL / MySQL /
33
+ SQL Server / Oracle via a SQLAlchemy URI, select a date column and a numeric
34
+ metric, and watch:
35
+
36
+ 1. Trend chart
37
+ 2. 90-step forecast with confidence fan, summary stats & interpret-ability
38
+ 3. Gemini-generated strategy report
39
+ 4. Optional histograms, scatter matrices, correlation heat-maps
40
+
41
+ ---
42
+
43
+ ## ✨ Features
44
+
45
+ | Area | Details |
46
+ | ---- | ------- |
47
+ | **Data ingest** | CSV, XLS/XLSX, or DB fetch (`sqlalchemy` URI) |
48
+ | **Sampling** | Auto-samples giant CSVs (β‰₯ 1 M rows) to avoid OOM |
49
+ | **Quick scan** | Markdown dataset overview: schema, missing %, numeric `describe()` |
50
+ | **Trend** | Interactive line + markers (Plotly Dark) |
51
+ | **Forecast** | ARIMA(1,1,1) auto-freq, 90-step (daily) or 3-step (monthly, etc) |
52
+ | | CI fan, residual ACF plot, 20 % back-test (MAPE, RMSE) |
53
+ | | Seasonal decomposition (toggle) |
54
+ | **Explainability** | Model summary table, plain-English coef notes |
55
+ | **Gemini 1.5 Pro** | 5 insights, 3 strategies, risks, extra-visuals β†’ Markdown + download |
56
+ | **EDA** | Histograms (with KDE), box / violin plots, scatter-matrix, corr heat-map |
57
+ | **Static images** | All figures also saved to `/tmp/*.png` for downstream reports |
58
+ | **Extensible** | Each visual is a standalone helper in `tools/` |
59
+
60
+ ---
61
+
62
+ ## πŸ›  Local Dev Setup
63
+
64
+ ```bash
65
+ # 1. Clone
66
+ git clone https://github.com/your-handle/BizIntel_AI.git
67
+ cd BizIntel_AI
68
+
69
+ # 2. Python env
70
+ python -m venv .venv
71
+ source .venv/bin/activate
72
+ pip install -r requirements.txt
73
+
74
+ # 3. Gemini key
75
+ export GEMINI_APIKEY="your_google_ai_studio_key"
76
+
77
+ # 4. Run
78
+ streamlit run app.py
79
+ Note : ARIMA & seasonal decomposition come from
80
+ statsmodels; Plotly PNG export needs kaleido.
81
+ These are included in requirements.txt.
82
+
83
+ πŸ”Œ Database Connect String Examples
84
+ text
85
+ Copy code
86
+ # PostgreSQL
87
+ postgresql+psycopg2://user:password@host:port/dbname
88
+
89
+ # MySQL
90
+ mysql+pymysql://user:password@host/dbname
91
 
92
+ # SQL Server
93
+ mssql+pyodbc://user:password@host:1433/db?driver=ODBC+Driver+17+for+SQL+Server
94
 
95
+ # Oracle
96
+ oracle+cx_oracle://user:password@host:1521/?service_name=ORCL
97
+ 🌐 Deploying to Hugging Face Spaces
98
+ Space β–Έ Create β–Έ Docker template
99
 
100
+ Push repo (Dockerfile, app.py, tools/, requirements.txt)
101
 
102
+ Add a secret β†’ GEMINI_APIKEY
103
 
104
+ Space builds ; enjoy!
 
 
 
 
105
 
106
+ πŸ—Ί Project Structure
107
+ Copy code
108
+ β”œβ”€ app.py
109
+ β”œβ”€ requirements.txt
110
+ β”œβ”€ Dockerfile
111
+ β”œβ”€ tools/
112
+ β”‚ β”œβ”€ csv_parser.py
113
+ β”‚ β”œβ”€ plot_generator.py
114
+ β”‚ β”œβ”€ forecaster.py
115
+ β”‚ β”œβ”€ visuals.py
116
+ β”‚ └─ db_connector.py
117
+ └─ README.md
118
+ 🀝 Contributing
119
+ Issues & PRs are welcome!
120
+ Please run black -l 88 and ruff before submitting.
121
 
122
+ πŸ“œ License
123
+ MIT Β© 2025 BizIntel AI Ultra Team