Update README.md

README.md

@@ -1,3 +1,82 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+---
+
+# Triplex
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+Triplex is a model for creating knowledge graphs from unstructured data. It works by extracting triplets - simple statements consisting of a subject, predicate, and object - from text or other data sources. Try the demo here: [kg.sciphi.ai](kg.sciphi.ai)
+
+## Model Details
+
+It is a finetuned version of Phi3-3.8B on a high quality proprietary dataset constructed using DBPedia, Wikidata, and other data sources.
+
+### Model Description
+
+
+
+- **Developed by:** [https://www.SciPhi.ai](SciPhi.ai)
+
+### Model Sources
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [https://www.github.com/SciPhi-AI/R2R](https://www.github.com/SciPhi-AI/R2R)
+- **Blog:** [https://www.sciphi.ai/blog/triplex](https://www.sciphi.ai/blog/triplex)
+- **Demo:** [kg.sciphi.ai](kg.sciphi.ai)
+
+
+```python
+
+import json
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+
+def triplextract(model, tokenizer, text, entity_types, predicates):
+
+    input_format = """
+        **Entity Types:**
+        {entity_types}
+
+        **Predicates:**
+        {predicates}
+
+        **Text:**
+        {text}
+        """
+
+    message = input_format.format(entity_types = json.dumps({"entity_types": entity_types}), predicates = json.dumps({"predicates": predicates}), text = text)
+
+    messages = [{'role': 'user', 'content': message}]
+    input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt = True, return_tensors="pt").to("cuda")
+    output = tokenizer.decode(model.generate(input_ids=input_ids, max_length=2048)[0], skip_special_tokens=True)
+    print(output)
+    return output
+
+
+
+tokenizer = AutoTokenizer.from_pretrained("sciphi/triplex", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("sciphi/triplex", trust_remote_code=True)
+
+
+model.to("cuda")
+
+model.eval()
+
+
+entity_types = [ "LOCATION", "POSITION", "DATE", "CITY", "COUNTRY", "NUMBER" ]
+
+predicates = [ "POPULATION", "AREA" ]
+
+text = """
+San Francisco,[24] officially the City and County of San Francisco, is a commercial, financial, and cultural center in Northern California. 
+
+With a population of 808,437 residents as of 2022, San Francisco is the fourth most populous city in the U.S. state of California behind Los Angeles, San Diego, and San Jose.
+"""
+
+prediction = triplextract(model, tokenizer, text, entity_types, predicates)
+print(prediction)
+
+
+```