YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
School Management AI Agent (Transfer-Based Seq2Seq)
This model is a transfer-learning fine-tuned checkpoint of google/flan-t5-small developed specifically for the School Management Database AI Assistant.
It translates natural-language queries from school administrators, faculty, and students into Domain-Specific Language (DSL) query specifications that compile directly to verified, parameterized PostgreSQL queries against the school schema.
Key Capabilities
- Strict Schema Grounding: Targets verified PostgreSQL tables and views (
v_student_attendance_report,v_exam_report,v_pending_leaves,v_timetable,students,faculty,marks). - Entity Resolution: Seamlessly parses student names, faculty names, grade/section identifiers, exam types, and dates.
- Zero-Cost Remote Access: Can be used locally or remotely loaded directly from Hugging Face Model Hub using Transformers.
Model Evaluation Metrics
- Base Architecture:
google/flan-t5-small - Training Samples: 100
- Validation Samples: 10
- Validation Loss: 1.9191
- Exact Match DSL Accuracy: 0.0%
- Training Time: 127.72s
Usage
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("jaay2/school-management-t5-agent")
model = AutoModelForSeq2SeqLM.from_pretrained("jaay2/school-management-t5-agent")
query = "Show attendance report for who is below 75%"
inputs = tokenizer(query, return_tensors="pt")
outputs = model.generate(**inputs, max_length=128)
dsl_spec = tokenizer.decode(outputs[0], skip_special_tokens=True)
print("Generated DSL:", dsl_spec)
- Downloads last month
- 17
Model tree for jaay2/school-management-t5-agent
Base model
google/flan-t5-small