Ferrxni commited on
Commit
285aa76
·
1 Parent(s): 84b61cf
Files changed (2) hide show
  1. README.md +38 -14
  2. app.py +1 -1
README.md CHANGED
@@ -17,6 +17,8 @@ hf_oauth: false
17
  ---
18
  Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
19
 
 
 
20
  #TO-DO - Brainstorming
21
 
22
  ** Système de carte interactive collaborative, AI-powered par Mistral **
@@ -33,31 +35,53 @@ Check out the configuration reference at https://huggingface.co/docs/hub/spaces-
33
  - vidéos postées par des agriculteurs de ta région
34
  - personnalisation du fond de carte IGN (cf. api IGN) ; geopandas pour Python je crois
35
  - data.gouv.fr https://www.data.gouv.fr/fr/reuses/la-carte-interactive-des-types-delevages-et-des-regions-agricoles-en-france/
 
 
 
 
 
 
 
 
 
 
 
 
 
36
 
37
- # GAIA Hackathon
38
 
39
- Welcome to the GAIA Hackathon! This hackathon is a collaborative effort between Comexposium, La Ferme Digitale, and Mistral AI, aimed at developing innovative solutions to address key challenges in the agriculture sector.
 
40
 
41
- ## Hackathon Description
42
 
43
- Participants in this hackathon will be tasked with designing tools capable of answering questions posed through one-minute videos. To accomplish this, teams will have access to a kit containing 50 solutions and 150 datasets provided by the hackathon organizers, starting from January 29, 2024. This approach will enable the rapid development of concrete and operational proof-of-concepts (POCs).
44
 
45
- As part of this hackathon, Comexposium and La Ferme Digitale have partnered with Mistral AI to provide participants with privileged access to the first multimodal agricultural AI model. This model combines natural language, voice, images, and data to create a comprehensive and innovative solution. The POCs developed during the hackathon will contribute to the utilization and enhancement of this model, opening up new possibilities for the application of AI in the agricultural domain. Two demonstrations have been set up, which participants can access (level 1 and level 2).
46
 
47
- For getting started, we highly recommend using [Hugging Face](https://huggingface.co/).
 
 
 
48
 
49
- If you have any technical questions, please visit the Mistral support site on [Discord](https://discord.gg/pCePuRhG).
50
 
51
- If you wish to conduct local tests and do not have a MISTRAL API key, please contact us directly via email.
 
 
52
 
53
- The elements will continue to evolve for 15 days, but all the main concepts are already present. Participants in the GAIA hackathon will be confronted with crucial questions, including farm succession, the attractiveness of agricultural professions and training, crop management, livestock management, purchasing and selling, as well as deciphering standards and regulations.
54
 
55
- The goal is to find innovative, practical, and pragmatic solutions that can be implemented immediately to address these essential challenges for the future of agriculture.
 
 
56
 
57
- ## Contact Information
58
 
59
- For inquiries or assistance, please reach out to us:
60
 
61
 
 
62
 
63
- Website: [www.gaiacorp.com](https://www.gaiacorp.com)
 
17
  ---
18
  Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
19
 
20
+ Checkout the space HuggingFace to visualize the solution at https://huggingface.co/spaces/Ferrxni/Gaia_Mistral
21
+
22
  #TO-DO - Brainstorming
23
 
24
  ** Système de carte interactive collaborative, AI-powered par Mistral **
 
35
  - vidéos postées par des agriculteurs de ta région
36
  - personnalisation du fond de carte IGN (cf. api IGN) ; geopandas pour Python je crois
37
  - data.gouv.fr https://www.data.gouv.fr/fr/reuses/la-carte-interactive-des-types-delevages-et-des-regions-agricoles-en-france/
38
+
39
+ # Gaia Mistral Chat Demo
40
+
41
+ ## Introduction
42
+
43
+ This project showcases a simple chatbot application built with Gradio and the Mistral AI API. The chatbot is designed to answer questions related to agriculture. Users can ask questions in French, and the chatbot, powered by the Mistral AI's `mistral` model, will provide responses.
44
+
45
+ ## Features
46
+
47
+ - Simple chat interface for querying agricultural topics. (MVP 1)
48
+ - Interactive map with chatbot experience (MVP 2)
49
+
50
+ ## Requirements
51
 
52
+ To run this application, you'll need:
53
 
54
+ - Python 3.6 or later.
55
+ - An API key from Mistral AI.
56
 
57
+ ## Setup
58
 
59
+ 1. **Clone the Repository**
60
 
61
+ Start by cloning this repository to your local machine:
62
 
63
+ ```bash
64
+ git clone <repository-url>
65
+ cd <repository-directory>
66
+ ```
67
 
68
+ 2. **Install Dependencies**
69
 
70
+ ``bash
71
+ pip install -r requirements.txt
72
+ ```
73
 
74
+ 3. **Set Up Your API Key**
75
 
76
+ ``bash
77
+ API_KEY=your_mistral_ai_api_key_here
78
+ ```
79
 
80
+ ## Running the Application
81
 
82
+ To launch the chatbot, run the following command in the terminal from the project's root directory:
83
 
84
+ ```bash
85
+ python3 app.py
86
+ ```
87
 
 
app.py CHANGED
@@ -24,7 +24,7 @@ def chat_with_mistral(user_input):
24
 
25
  app = gr.Interface(
26
  fn=chat_with_mistral,
27
- inputs=gr.inputs.Textbox(lines=2, placeholder=placeholder),
28
  outputs="text",
29
  title=title,
30
  description=description,
 
24
 
25
  app = gr.Interface(
26
  fn=chat_with_mistral,
27
+ inputs=gr.Textbox(lines=2, placeholder=placeholder),
28
  outputs="text",
29
  title=title,
30
  description=description,