Files
GPT2-Medium-MORIAI/README.md
ModelHub XC 494cc99b13 初始化项目,由ModelHub XC社区提供模型
Model: ShikoShin/GPT2-Medium-MORIAI
Source: Original Platform
2026-07-18 11:44:12 +08:00

222 lines
4.9 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
library_name: transformers
license: apache-2.0
language:
- en
base_model:
- openai-community/gpt2-medium
---
# MORI AI
## Model Details
### Model Description
MORI (Machine-Oriented Responsive Intelligence) is a transformer-based conversational AI developed as part of the research project:
**"Design and Implementation of an NLP-Driven Intelligent Web Companion for Real-Time User Interaction."**
This version of MORI is specifically designed to assist students, faculty, and stakeholders of ICCT Colleges San Mateo Campus by providing information related to academic programs, enrollment procedures, registrar services, school policies, and general campus inquiries.
The model focuses on answering institution-related questions and providing guidance based on publicly available academic and administrative information.
**Developed by:** ShikoShin
**Model Type:** Transformer-Based Conversational AI
**Language(s):** English
**License:** Apache 2.0
**Finetuned from:** DialoGPT-Large
---
# Model Sources
**Institution:** ICCT Colleges
**Research Project:** Design and Implementation of an NLP-Driven Intelligent Web Companion for Real-Time User Interaction
---
# Intended Use
## Primary Use Cases
MORI is intended to assist users with:
* Academic program inquiries
* Course information
* Admission requirements
* Enrollment procedures
* Registrar-related concerns
* School policies and guidelines
* Frequently asked questions about ICCT Colleges San Mateo Campus
## Educational Support
The model may be integrated into:
* Student assistance platforms
* School websites
* Mobile applications
* Information kiosks
* Academic support systems
---
# Out-of-Scope Use
MORI is not intended for:
* Medical advice
* Legal advice
* Financial consulting
* Psychological counseling
* Emergency response
* Safety-critical decision making
Users should consult official ICCT personnel for authoritative decisions regarding academic records, enrollment status, and institutional policies.
---
# Training Data
## Dataset Description
MORI was trained using a curated dataset consisting of institution-specific information related to:
* Academic courses and programs
* Enrollment workflows
* Admission requirements
* Registrar procedures
* Frequently asked student inquiries
* Publicly available institutional information
### Data Privacy Statement
No sensitive personal information was used during training.
The dataset does **not** contain:
* Student records
* Grades or transcripts
* Personal identification information
* Financial records
* Medical information
* Confidential institutional documents
Training data was limited to educational and administrative information intended for public dissemination.
---
# Limitations
MORI may:
* Generate incorrect responses
* Provide outdated information if institutional policies change
* Misinterpret ambiguous questions
* Require human verification for official transactions
The model should be treated as an informational assistant rather than an official source of record.
---
# How to Get Started
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "ShikoShin/DialoGPT-Large-MORIAI"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
prompt = "How do I enroll at ICCT San Mateo?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(
**inputs,
max_new_tokens=100,
do_sample=True,
temperature=0.8
)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
# Training Procedure
## Preprocessing
The dataset underwent:
* Data cleaning
* Duplicate removal
* Text normalization
* Formatting into question-answer pairs
* Tokenization
## Framework
* Python
* PyTorch
* Hugging Face Transformers
* Hugging Face Tokenizers
---
# Evaluation
## Evaluation Criteria
The model was evaluated based on:
* Response relevance
* Institutional information accuracy
* Conversational consistency
* User query understanding
## Results
MORI demonstrated the ability to answer common student and administrative questions related to ICCT Colleges San Mateo Campus while maintaining conversational coherence.
---
# Technical Specifications
## Architecture
Transformer-based autoregressive language model fine-tuned from DialoGPT-Large.
### Objective
Generate contextually appropriate responses to institution-related inquiries while maintaining conversational flow.
---
# Citation
### APA
ShikoShin. (2026). MORI: Intelligent Academic Assistant for ICCT Colleges San Mateo Campus. Hugging Face.
### BibTeX
@misc{mori2026,
author = {ShikoShin},
title = {MORI: Intelligent Academic Assistant for ICCT Colleges -- San Mateo Campus},
year = {2026},
publisher = {Hugging Face}
}
---
# Authors
ShikoShin
---
# Contact
For questions, feedback, or collaboration opportunities, please contact the repository owner through Hugging Face.