Quantifying the Carbon Emissions of Machine Learning
Paper
•
1910.09700
•
Published
•
32
Model Card for Model ID
This model performs intelligent battery cell fault diagnosis using real-time voltage, temperature, and current inputs. It consists of a hybrid architecture:
A RandomForestClassifier predicts the cell fault status.
A GPT-2 model, fine-tuned on labeled diagnosis data, generates recommendations based on sensor readings.
The model is deployed with an TTS modules can be added to explain results via speech or video.
Model Details
Model Description
A RandomForestClassifier for predicting battery cell status based on voltage, temperature, and current.
A fine-tuned GPT-2 language model to generate recommendation text tailored to the predicted fault.
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: Bhoomika Patil (IEEE Member, AESS Vice Chair 2024, PES Chair 2025)
- Funded by [optional]: [More Information Needed]
- Shared by [optional]: [More Information Needed]
- Model type: Classification + Text Generation
- Language(s) (NLP): [More Information Needed]
- License: [More Information Needed]
- Finetuned from model [optional]: gpt2 by Hugging Face
Model Sources [optional]
- Repository: [More Information Needed]
- Paper [optional]: [More Information Needed]
- Demo [optional]: [More Information Needed]
Uses
Direct Use
This model is intended to be used in:
Embedded EV BMS fault diagnosis
Lab-scale battery monitoring systems
Educational/academic demonstration kits
Downstream Use [optional]
Can be used for:
Real-time IoT fault detection dashboards
Voice-based battery recommendation bots
Integration into EV charging systems
Out-of-Scope Use
Not suitable for grid-level battery systems without retraining
Not a substitute for hardware-based protections
Bias, Risks, and Limitations
May give inaccurate predictions outside of the trained sensor range
GPT-2 recommendations may not always be technically feasible or safe
Does not consider time-series or historical trends
Limited robustness for deployment in production-grade EVs
Recommendations
Validate GPT-generated text before acting
Add confidence thresholds for alerts
Complement with hardware-level fault detection
How to Get Started with the Model
Use the code below to get started with the model.
from transformers import GPT2Tokenizer, GPT2LMHeadModel
tokenizer = GPT2Tokenizer.from_pretrained("your-username/gpt2-cell-diagnosis-system") model = GPT2LMHeadModel.from_pretrained("your-username/gpt2-cell-diagnosis-system")
prompt = "Voltage: 3.2V, Temperature: 48C, Current: 0.7A, Status: Overheat\nRecommendation:" inputs = tokenizer(prompt, return_tensors="pt") outputs = model.generate(**inputs, max_length=60) print(tokenizer.decode(outputs[0]))
Training Details
Training Data
Custom Excel dataset (CellDD.xlsx)
Includes labeled examples of Voltage, Temperature, Current, Status, and textual recommendations
Training Procedure
Preprocessing [optional]
Numerical columns normalized
Text cleaned and used to create prompts for GPT-2
Training Hyperparameters
- Training regime: [More Information Needed]
Speeds, Sizes, Times [optional]
[More Information Needed]
Evaluation
Testing Data, Factors & Metrics
Testing Data
20% split from original dataset (not public)
Factors
Classifier Accuracy: ~96%
Manual evaluation of GPT-2 text output
Metrics
Classifier Accuracy: ~96%
Manual evaluation of GPT-2 text output
Results
[More Information Needed]
Summary
Model Examination [optional]
[More Information Needed]
Environmental Impact
Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
- Hardware Type: Google Colab (1x T4 GPU)
- Hours used: ~25 minutes
- 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]
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
[More Information Needed]
More Information [optional]
[More Information Needed]
Model Card Authors [optional]
[More Information Needed]
Model Card Contact
[More Information Needed]
- Downloads last month
- 1