jeromex1
/

lyra_Oidium_mistral7B_LoRA

@@ -1,199 +1,172 @@
 ---
-library_name: transformers
-tags: []
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
+language:
+- fr
+- en
+license: apache-2.0
+tags:
+- agronomy
+- viticulture
+- plant-disease
+- oidium
+- mistral
+- sft
+- lora
+- decision-support
+pipeline_tag: text-generation
+model_name: Oidium_Mistral7B_SFT
 ---
+# Modèle de prévision du risque d’Oïdium – Vigne (Mistral 7B SFT)
+➡️ **[English version](#english-version)**
+---
+## 🎯 Objectif du modèle
+Ce modèle est un **Mistral 7B spécialisé (SFT)** dans l’évaluation du **risque d’oïdium de la vigne** et la génération de **recommandations agronomiques opérationnelles**, à partir d’un nombre limité mais pertinent de variables météorologiques et biologiques.
+Il a été conçu pour dépasser une simple reconnaissance de patterns, et apprendre une **logique décisionnelle conditionnelle**, proche du raisonnement d’un expert terrain.
+---
+## 🌱 Variables prises en compte
+Le modèle raisonne à partir des entrées suivantes :
+- Stade phénologique
+- Température moyenne (24 h)
+- Humidité relative (proxy de durée ≥ seuil sur 24 h)
+- Pluie sur 24 h (booléen)
+- Inoculum (proxy normalisé 0–1)
+- ETP (mm/j, proxy d’assèchement)
+Une attention particulière est portée aux **cas frontières**, par exemple :
+- humidité élevée mais forte ETP,
+- pluie récente suivie d’un assèchement rapide,
+- conditions proches des seuils critiques.
+---
+## 📊 Qualité du dataset et entraînement
+### Dataset
+- **Entraînement** : 1500 lignes
+- **Évaluation** : 100 lignes
+- Répartition des classes : **30 % faible / 40 % moyen / 30 % élevé**
+- Aucun doublon (train / eval / croisé)
+- 112 cas frontières explicitement intégrés
+### Entraînement
+- SFT réalisé sous Google Colab (GPU L4)
+- Durée : ~15 minutes
+- 3 epochs
+- Convergence stable (pas d’overfitting)
+---
+## 🧪 Résultats – Évaluation agronomique qualitative
+Sur une série de tests d’inférence ciblés (cas simples + cas frontières) :
+- **≈ 85–90 % des décisions sont pleinement valides agronomiquement**
+- Les rares écarts observés correspondent à des **choix prudents** (sur-classement moyen → élevé)
+- Aucun cas incohérent ou biologiquement aberrant n’a été observé
+Ces résultats indiquent que le modèle :
+- comprend les interactions pluie / humidité / ETP,
+- adapte ses recommandations au stade phénologique,
+- respecte strictement le format de sortie attendu.
+Le modèle est donc **opérationnel en l’état**, sans nécessité de micro-ajustement supplémentaire.
+---
+## 🔧 Cas d’usage
+- Aide à la décision agronomique
+- Surveillance du risque phytosanitaire
+- Intégration dans des workflows automatisés (ex. n8n)
+- Démonstrateur IA agronomique open-source
+⚠️ Ce modèle ne remplace pas une expertise humaine locale et doit être utilisé comme **outil d’aide à la décision**.
+---
+## 📌 Statut
+- Modèle validé qualitativement
+- Déployé sur Hugging Face
+- Prêt pour intégration API / workflow
+---
+## English version
+### 🎯 Model purpose
+This model is a **specialized Mistral 7B (SFT)** designed to assess **powdery mildew (Oidium) risk in vineyards** and generate **actionable agronomic recommendations** based on a limited yet informative set of meteorological and biological variables.
+It is designed to learn **conditional decision logic**, not only pattern recognition.
+---
+### 🌱 Input variables
+- Phenological stage
+- Mean temperature (24 h)
+- Relative humidity (proxy for ≥ threshold duration)
+- Rainfall (24 h, boolean)
+- Inoculum pressure (normalized proxy 0–1)
+- Reference evapotranspiration (ETP, mm/day)
+The model explicitly handles **borderline and contradictory scenarios**, such as high humidity combined with strong drying conditions.
+---
+### 📊 Dataset and training quality
+- **Training set**: 1500 samples
+- **Evaluation set**: 100 samples
+- Class distribution: **30 % low / 40 % medium / 30 % high risk**
+- No duplicates (train / eval / cross)
+- 112 explicit borderline cases
+Training details:
+- Supervised Fine-Tuning (SFT)
+- GPU: L4 (Google Colab)
+- 3 epochs, stable convergence
+---
+### 🧪 Agronomic validity assessment
+Based on targeted inference tests:
+- **~85–90 % of decisions are agronomically valid**
+- Remaining cases reflect **conservative but defensible choices**
+- No biologically incoherent outputs observed
+The model demonstrates reliable reasoning over humidity, rainfall, ETP, and phenological sensitivity.
+---
+### 🔧 Use cases
+- Agronomic decision support
+- Vineyard disease risk monitoring
+- Automated pipelines (e.g. n8n)
+- Open-source agronomic AI demonstration
+⚠️ This model is an **assistive tool** and does not replace local expert judgment.
+---
+### 📌 Status
+- Qualitatively validated
+- Deployed on Hugging Face
+- Ready for API and workflow integration