Fix YAML metadata, add base model (Falconsai/text_summarization) and dataset references

README.md

@@ -1,9 +1,56 @@
+---
+language: en
+license: apache-2.0
+base_model: Falconsai/text_summarization
+tags:
+- summarization
+- email
+- t5
+- text2text-generation
+- brief-summary
+- full-summary
+datasets:
+- argilla/FinePersonas-Conversations-Email-Summaries
+metrics:
+- rouge
+widget:
+- text: "summarize_brief: Subject: Team Meeting Tomorrow. Body: Hi everyone, Just a reminder that we have our weekly team meeting tomorrow at 2 PM EST. Please prepare your status updates and any blockers you're facing. We'll also discuss the Q4 roadmap. Thanks!"
+  example_title: "Brief Summary"
+- text: "summarize_full: Subject: Project Update. Body: The development team has completed the first phase of the new feature implementation. We've successfully integrated the API, updated the UI components, and conducted initial testing. The performance improvements show a 40% reduction in load time. Next steps include user acceptance testing and documentation updates."
+  example_title: "Full Summary"
+- text: "summarize_brief: Subject: lunch mtg. Body: hey guys, cant make lunch today bc stuck in traffic. can we do tmrw at 1pm instead? lmk what works. thx!"
+  example_title: "Messy Email (Brief)"
+model-index:
+- name: t5-small-email-summarizer
+  results:
+  - task:
+      type: summarization
+      name: Email Summarization
+    dataset:
+      type: argilla/FinePersonas-Conversations-Email-Summaries
+      name: FinePersonas Email Summaries
+    metrics:
+    - type: rouge-l
+      value: 0.42
+      name: ROUGE-L
+pipeline_tag: summarization
+library_name: transformers
+---
+
 # T5 Email Summarizer - Brief & Full
 
 ## Model Description
 
 This is a fine-tuned T5-small model specialized for email summarization. The model can generate both brief (one-line) and detailed (comprehensive) summaries of emails, and is robust to messy, informal inputs with typos and abbreviations.
 
+### Model Details
+- **Base Model**: [Falconsai/text_summarization](https://huggingface.co/Falconsai/text_summarization) (T5-small)
+- **Fine-tuned by**: Wordcab Team
+- **Model type**: T5 (Text-to-Text Transfer Transformer)
+- **Language**: English
+- **License**: Apache 2.0
+- **Demo**: [Try it on Spaces](https://huggingface.co/spaces/wordcab/t5-email-summarizer-demo)
+
 ### Key Features
 - **Dual-mode summarization**: Supports both `summarize_brief:` and `summarize_full:` prefixes
 - **Robust to informal text**: Handles typos, abbreviations, and casual language
@@ -42,7 +89,7 @@ Training data was augmented with:
 ## Training Procedure
 
 ### Training Details
-- **Base model**: google/t5-v1_1-small
+- **Base model**: [Falconsai/text_summarization](https://huggingface.co/Falconsai/text_summarization)
 - **Training epochs**: 1
 - **Batch size**: 64
 - **Learning rate**: 3e-4
@@ -67,8 +114,8 @@ pip install transformers torch
 from transformers import T5ForConditionalGeneration, T5Tokenizer
 
 # Load model and tokenizer
-tokenizer = T5Tokenizer.from_pretrained("Wordcab/t5-small-email-summarizer")
-model = T5ForConditionalGeneration.from_pretrained("Wordcab/t5-small-email-summarizer")
+tokenizer = T5Tokenizer.from_pretrained("wordcab/t5-small-email-summarizer")
+model = T5ForConditionalGeneration.from_pretrained("wordcab/t5-small-email-summarizer")
 
 # Example email
 email = """Subject: Team Meeting Tomorrow. Body: Hi everyone, 
@@ -116,7 +163,7 @@ def summarize_long_email(email, model, tokenizer, mode="brief"):
 - **Coherence score on messy inputs**: 80%
 
 ### Comparison with Base Model
-- 8.3% improvement in quality over base Falconsai/text_summarization
+- 8.3% improvement in quality over base [Falconsai/text_summarization](https://huggingface.co/Falconsai/text_summarization)
 - Successfully differentiates brief vs full summaries (2.5x length difference)
 - Better handling of informal text and typos
 
@@ -126,7 +173,7 @@ def summarize_long_email(email, model, tokenizer, mode="brief"):
 ```python
 import requests
 
-API_URL = "https://api-inference.huggingface.co/models/Wordcab/t5-small-email-summarizer"
+API_URL = "https://api-inference.huggingface.co/models/wordcab/t5-small-email-summarizer"
 headers = {"Authorization": "Bearer YOUR_HF_TOKEN"}
 
 def query(payload):
@@ -143,7 +190,7 @@ output = query({
 docker run --gpus all -p 8080:80 \
   -v t5-small-email-summarizer:/model \
   ghcr.io/huggingface/text-generation-inference:latest \
-  --model-id Wordcab/t5-small-email-summarizer \
+  --model-id wordcab/t5-small-email-summarizer \
   --max-input-length 512 \
   --max-total-tokens 662
 ```
@@ -176,7 +223,7 @@ If you use this model, please cite:
   author={Wordcab Team},
   year={2025},
   publisher={HuggingFace},
-  url={https://huggingface.co/Wordcab/t5-small-email-summarizer}
+  url={https://huggingface.co/wordcab/t5-small-email-summarizer}
 }
 ```
 
@@ -186,11 +233,11 @@ This model is released under the Apache 2.0 License.
 
 ## Acknowledgments
 
-- Based on Google's T5 architecture
-- Fine-tuned using HuggingFace Transformers
-- Training data derived from public email datasets
+- Based on [Falconsai/text_summarization](https://huggingface.co/Falconsai/text_summarization) T5 model
+- Fine-tuned on [argilla/FinePersonas-Conversations-Email-Summaries](https://huggingface.co/datasets/argilla/FinePersonas-Conversations-Email-Summaries) dataset
+- Training performed using HuggingFace Transformers
 - Special thanks to the open-source community
 
 ## Contact
 
-For questions or feedback, please open an issue on the [model repository](https://huggingface.co/Wordcab/t5-small-email-summarizer/discussions).
+For questions or feedback, please open an issue on the [model repository](https://huggingface.co/wordcab/t5-small-email-summarizer/discussions).