text update in UI and readme
Browse files- README.md +10 -10
- app.py +30 -47
- utils/__pycache__/generator.cpython-310.pyc +0 -0
- utils/__pycache__/retriever.cpython-310.pyc +0 -0
README.md
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
---
|
2 |
-
title:
|
3 |
emoji: 🧐
|
4 |
colorFrom: yellow
|
5 |
colorTo: gray
|
@@ -14,21 +14,21 @@ models:
|
|
14 |
- BAAI/bge-reranker-v2-m3
|
15 |
- meta-llama/Llama-3.1-8B-Instruct
|
16 |
---
|
17 |
-
short_description: AI-powered conversational assistant for
|
18 |
license: apache-2.0
|
19 |
|
20 |
|
21 |
## Technical Documentation of the system in accordance with EU AI Act
|
22 |
|
23 |
-
**System Name:**
|
24 |
|
25 |
**Provider / Supplier:** GIZ Data Service Center
|
26 |
|
27 |
-
**As of:**
|
28 |
|
29 |
## 1. General Description of the System
|
30 |
|
31 |
-
|
32 |
|
33 |
It combines a generative language assistant with a knowledge base implemented via Retrieval-Augmented Generation (RAG). The scope and functionality of the tool is focused on EU Deforestation Regulation compliance and related documentation.
|
34 |
|
@@ -48,11 +48,11 @@ It combines a generative language assistant with a knowledge base implemented vi
|
|
48 |
|
49 |
## 3. Model Training Data
|
50 |
|
51 |
-
All the models mentioned above are being consumed without any fine-tuning or training being performed by the developer team of
|
52 |
|
53 |
## 4. Knowledge Base (Retrieval Component)
|
54 |
|
55 |
-
- **Data Sources:** Public
|
56 |
- **Embedding Model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3)
|
57 |
- **Embedding Dimension:** 1024
|
58 |
- **Vector Database:** Qdrant (via API)
|
@@ -71,14 +71,14 @@ All the models mentioned above are being consumed without any fine-tuning or tra
|
|
71 |
- The user interface clearly indicates the use of a generative AI model.
|
72 |
- An explanation of the RAG method is included.
|
73 |
- We collect usage statistics as detailed in Disclaimer tab of the app along with the explicit display in the user interface of the tool.
|
74 |
-
- Feedback mechanism available (via https://huggingface.co/spaces/GIZ/
|
75 |
|
76 |
## 7. Monitoring, Feedback, and Incident Reporting
|
77 |
|
78 |
-
- User can provide feedback via UI by giving (Thumbs-up or down to AI-Generated answer). Alternatively for more detailed feedback please use https://huggingface.co/spaces/GIZ/
|
79 |
- Technical development is carried out by the GIZ Data Service Center.
|
80 |
- No automated bias detection – but low risk due to content restrictions.
|
81 |
|
82 |
## 8. Contact
|
83 |
|
84 |
-
For any questions, please contact via https://huggingface.co/spaces/GIZ/
|
|
|
1 |
---
|
2 |
+
title: Gina Assistant
|
3 |
emoji: 🧐
|
4 |
colorFrom: yellow
|
5 |
colorTo: gray
|
|
|
14 |
- BAAI/bge-reranker-v2-m3
|
15 |
- meta-llama/Llama-3.1-8B-Instruct
|
16 |
---
|
17 |
+
short_description: AI-powered conversational assistant for Circular Economy
|
18 |
license: apache-2.0
|
19 |
|
20 |
|
21 |
## Technical Documentation of the system in accordance with EU AI Act
|
22 |
|
23 |
+
**System Name:** Gina ChatBot
|
24 |
|
25 |
**Provider / Supplier:** GIZ Data Service Center
|
26 |
|
27 |
+
**As of:** August 2025
|
28 |
|
29 |
## 1. General Description of the System
|
30 |
|
31 |
+
Gina Bot is an AI-powered conversational assistant designed to help you understand compliance with and analyze the Circular Economy topic. This tool leverages advanced language models to help you get clear and structured answers about Circular Economy requirements, compliance procedures, and regulatory guidance.
|
32 |
|
33 |
It combines a generative language assistant with a knowledge base implemented via Retrieval-Augmented Generation (RAG). The scope and functionality of the tool is focused on EU Deforestation Regulation compliance and related documentation.
|
34 |
|
|
|
48 |
|
49 |
## 3. Model Training Data
|
50 |
|
51 |
+
All the models mentioned above are being consumed without any fine-tuning or training being performed by the developer team of Gina Bot. And hence there is no training data which had been used by the development team of Gina Bot.
|
52 |
|
53 |
## 4. Knowledge Base (Retrieval Component)
|
54 |
|
55 |
+
- **Data Sources:** Public Circular Economy documentation, regulatory guidance, and compliance materials
|
56 |
- **Embedding Model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3)
|
57 |
- **Embedding Dimension:** 1024
|
58 |
- **Vector Database:** Qdrant (via API)
|
|
|
71 |
- The user interface clearly indicates the use of a generative AI model.
|
72 |
- An explanation of the RAG method is included.
|
73 |
- We collect usage statistics as detailed in Disclaimer tab of the app along with the explicit display in the user interface of the tool.
|
74 |
+
- Feedback mechanism available (via https://huggingface.co/spaces/GIZ/gina_dev/discussions/new).
|
75 |
|
76 |
## 7. Monitoring, Feedback, and Incident Reporting
|
77 |
|
78 |
+
- User can provide feedback via UI by giving (Thumbs-up or down to AI-Generated answer). Alternatively for more detailed feedback please use https://huggingface.co/spaces/GIZ/gina_dev/discussions/new to report any issue.
|
79 |
- Technical development is carried out by the GIZ Data Service Center.
|
80 |
- No automated bias detection – but low risk due to content restrictions.
|
81 |
|
82 |
## 8. Contact
|
83 |
|
84 |
+
For any questions, please contact via https://huggingface.co/spaces/GIZ/gina_dev/discussions/new or send us email to [email protected]
|
app.py
CHANGED
@@ -9,21 +9,15 @@ from utils.generator import generate
|
|
9 |
|
10 |
# Sample questions for examples
|
11 |
SAMPLE_QUESTIONS = {
|
12 |
-
"
|
13 |
-
"¿
|
14 |
-
"
|
15 |
-
"¿
|
16 |
],
|
17 |
-
"
|
18 |
-
"¿
|
19 |
-
"¿Cómo
|
20 |
-
"¿Qué
|
21 |
-
],
|
22 |
-
"Evaluación de riesgos": [
|
23 |
-
"¿Cuál es el nivel de riesgo de deforestación en esta región?",
|
24 |
-
"¿Cómo evalúo los riesgos de la cadena de suministro?",
|
25 |
-
"¿Cuáles son los plazos de cumplimiento?"
|
26 |
-
]
|
27 |
}
|
28 |
|
29 |
# Global variable to cache API results and prevent double calls
|
@@ -193,25 +187,19 @@ custom_css = """
|
|
193 |
"""
|
194 |
|
195 |
init_prompt = """
|
196 |
-
Hola, soy
|
197 |
-
|
198 |
-
la deforestación. Responderé a sus preguntas utilizando los informes EUDR y los archivos
|
199 |
-
GeoJSON cargados.
|
200 |
-
|
201 |
-
💡 **Cómo utilizarlo (pestañas a la derecha)**
|
202 |
-
|
203 |
-
**Fuentes de datos:** elige entre subir un archivo GeoJSON para su análisis o consultar los informes EUDR filtrados por país.
|
204 |
-
**Ejemplos:** selecciona entre preguntas de ejemplo seleccionadas de diferentes categorías.
|
205 |
-
**Fuentes:** consulta las fuentes de contenido utilizadas para generar respuestas para la verificación de datos.
|
206 |
-
|
207 |
-
⚠️ Para conocer las limitaciones y la información sobre la recopilación de datos, consulta la pestaña «Aviso legal».
|
208 |
|
|
|
|
|
|
|
|
|
209 |
"""
|
210 |
|
211 |
-
with gr.Blocks(title="
|
212 |
|
213 |
# Main Chat Interface
|
214 |
-
with gr.Tab("
|
215 |
with gr.Row():
|
216 |
# Left column - Chat interface (2/3 width)
|
217 |
with gr.Column(scale=2):
|
@@ -236,7 +224,7 @@ with gr.Blocks(title="EUDR Bot", theme=theme, css=custom_css) as demo:
|
|
236 |
# Input textbox
|
237 |
with gr.Row():
|
238 |
textbox = gr.Textbox(
|
239 |
-
placeholder="
|
240 |
show_label=False,
|
241 |
scale=7,
|
242 |
lines=1,
|
@@ -319,22 +307,21 @@ with gr.Blocks(title="EUDR Bot", theme=theme, css=custom_css) as demo:
|
|
319 |
# Guidelines Tab
|
320 |
with gr.Tab("Directrices"):
|
321 |
gr.Markdown("""
|
322 |
-
#### Welcome to
|
323 |
|
324 |
-
This AI-powered assistant helps you understand
|
325 |
|
326 |
## 💬 How to Ask Effective Questions
|
327 |
|
328 |
| ❌ Less Effective | ✅ More Effective |
|
329 |
|------------------|-------------------|
|
330 |
-
| "What is
|
331 |
-
| "Tell me about compliance" | "What
|
332 |
-
| "Show me data" | "What is the
|
333 |
|
334 |
## 🔍 Using Data Sources
|
335 |
|
336 |
-
**
|
337 |
-
**Talk to Reports:** Select Ecuador or Guatemala for country-specific EUDR analysis
|
338 |
|
339 |
## ⭐ Best Practices
|
340 |
|
@@ -347,23 +334,20 @@ with gr.Blocks(title="EUDR Bot", theme=theme, css=custom_css) as demo:
|
|
347 |
# About Tab
|
348 |
with gr.Tab("Información"):
|
349 |
gr.Markdown("""
|
350 |
-
## About
|
351 |
|
352 |
-
The **
|
353 |
-
placed on the EU market are deforestation-free and legally produced.
|
354 |
|
355 |
This AI-powered tool helps stakeholders:
|
356 |
-
- Understand
|
357 |
-
-
|
358 |
-
- Assess supply chain risks
|
359 |
- Navigate complex regulatory landscapes
|
360 |
|
361 |
-
**Developed by GIZ** to enhance accessibility and understanding of
|
362 |
through advanced AI and geographic data processing capabilities.
|
363 |
|
364 |
### Key Features:
|
365 |
-
-
|
366 |
-
- Country-specific EUDR compliance guidance
|
367 |
- Real-time question answering with source citations
|
368 |
- User-friendly interface for complex regulatory information
|
369 |
""")
|
@@ -374,12 +358,11 @@ with gr.Blocks(title="EUDR Bot", theme=theme, css=custom_css) as demo:
|
|
374 |
## Important Disclaimers
|
375 |
|
376 |
⚠️ **Scope & Limitations:**
|
377 |
-
- This tool is designed for
|
378 |
- Responses should not be considered official legal or compliance advice
|
379 |
- Always consult qualified professionals for official compliance decisions
|
380 |
|
381 |
⚠️ **Data & Privacy:**
|
382 |
-
- Uploaded GeoJSON files are processed via external WHISP API for analysis
|
383 |
- We collect usage statistics to improve the tool
|
384 |
- Files are processed temporarily and not permanently stored
|
385 |
|
|
|
9 |
|
10 |
# Sample questions for examples
|
11 |
SAMPLE_QUESTIONS = {
|
12 |
+
"Fundamentos y tendencias internacionales de EC": [
|
13 |
+
"¿Cómo se diferencia el modelo de economía circular del modelo lineal tradicional de 'tomar, hacer, desechar'?",
|
14 |
+
"¿Cuáles son algunos de los principios clave de la economía circular y cómo se aplican en la práctica?",
|
15 |
+
"¿Podrías dar ejemplos de industrias o empresas que estén implementando con éxito prácticas de economía circular?"
|
16 |
],
|
17 |
+
"EC en Colombia": [
|
18 |
+
"¿Qué políticas y normativas vigentes en Colombia impulsan la adopción de la economía circular?",
|
19 |
+
"¿Cómo pueden las regulaciones colombianas incentivar la innovación en el ecodiseño y la gestión de residuos?",
|
20 |
+
"¿Qué papel tienen los instrumentos económicos y fiscales en la promoción de la circularidad en el sector productivo de Colombia?"]
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
}
|
22 |
|
23 |
# Global variable to cache API results and prevent double calls
|
|
|
187 |
"""
|
188 |
|
189 |
init_prompt = """
|
190 |
+
Hola, soy Gina, una asistente conversacional con IA diseñada para ayudarte a comprender conceptos y ayudarte con el tema de la Economía Circular. Responderé a tus preguntas usando la base de datos de documentos sobre economía circular.
|
191 |
+
💡 **Cómo usarla (pestañas a la derecha)**
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
192 |
|
193 |
+
**Fuentes de datos:** Selecciona la sección de informes/documentos.
|
194 |
+
**Ejemplos:** Selecciona entre ejemplos de preguntas de diferentes categorías.
|
195 |
+
**Fuentes:** Consulta las fuentes de contenido utilizadas para generar las respuestas para la verificación de datos.
|
196 |
+
⚠️ Para conocer las limitaciones e información sobre la recopilación de datos, consulta la pestaña "Aviso legal"
|
197 |
"""
|
198 |
|
199 |
+
with gr.Blocks(title="Gina Bot", theme=theme, css=custom_css) as demo:
|
200 |
|
201 |
# Main Chat Interface
|
202 |
+
with gr.Tab("Gina Bot"):
|
203 |
with gr.Row():
|
204 |
# Left column - Chat interface (2/3 width)
|
205 |
with gr.Column(scale=2):
|
|
|
224 |
# Input textbox
|
225 |
with gr.Row():
|
226 |
textbox = gr.Textbox(
|
227 |
+
placeholder="Pregúntame cualquier cosa sobre Economía Circular",
|
228 |
show_label=False,
|
229 |
scale=7,
|
230 |
lines=1,
|
|
|
307 |
# Guidelines Tab
|
308 |
with gr.Tab("Directrices"):
|
309 |
gr.Markdown("""
|
310 |
+
#### Welcome to Gina Q&A!
|
311 |
|
312 |
+
This AI-powered assistant helps you understand Circular Economy.
|
313 |
|
314 |
## 💬 How to Ask Effective Questions
|
315 |
|
316 |
| ❌ Less Effective | ✅ More Effective |
|
317 |
|------------------|-------------------|
|
318 |
+
| "What is economy?" | "What are impact of circular economy on businesses?" |
|
319 |
+
| "Tell me about compliance" | "What are country guidelines on circular economy" |
|
320 |
+
| "Show me data" | "What is the trend on waste and how circular economy is helping in resolving this?" |
|
321 |
|
322 |
## 🔍 Using Data Sources
|
323 |
|
324 |
+
**Talk to Reports:** Select reports sections "Trend and fundamentals", "Financing Mechanisms", "Country Resource"
|
|
|
325 |
|
326 |
## ⭐ Best Practices
|
327 |
|
|
|
334 |
# About Tab
|
335 |
with gr.Tab("Información"):
|
336 |
gr.Markdown("""
|
337 |
+
## About Gina Q&A
|
338 |
|
339 |
+
The **Circular Economy** places some obligations on the manufacturers and business.
|
|
|
340 |
|
341 |
This AI-powered tool helps stakeholders:
|
342 |
+
- Understand circular Economy concepts and regulations
|
343 |
+
- Assess supply chain issues
|
|
|
344 |
- Navigate complex regulatory landscapes
|
345 |
|
346 |
+
**Developed by GIZ** for project in Colombia to enhance accessibility and understanding of circular Economy requirements
|
347 |
through advanced AI and geographic data processing capabilities.
|
348 |
|
349 |
### Key Features:
|
350 |
+
- Country-specific compliance guidance
|
|
|
351 |
- Real-time question answering with source citations
|
352 |
- User-friendly interface for complex regulatory information
|
353 |
""")
|
|
|
358 |
## Important Disclaimers
|
359 |
|
360 |
⚠️ **Scope & Limitations:**
|
361 |
+
- This tool is designed for Circular Economy assistance and geographic data analysis
|
362 |
- Responses should not be considered official legal or compliance advice
|
363 |
- Always consult qualified professionals for official compliance decisions
|
364 |
|
365 |
⚠️ **Data & Privacy:**
|
|
|
366 |
- We collect usage statistics to improve the tool
|
367 |
- Files are processed temporarily and not permanently stored
|
368 |
|
utils/__pycache__/generator.cpython-310.pyc
CHANGED
Binary files a/utils/__pycache__/generator.cpython-310.pyc and b/utils/__pycache__/generator.cpython-310.pyc differ
|
|
utils/__pycache__/retriever.cpython-310.pyc
CHANGED
Binary files a/utils/__pycache__/retriever.cpython-310.pyc and b/utils/__pycache__/retriever.cpython-310.pyc differ
|
|