shreyaspimpalgaonkar
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -14,8 +14,6 @@ It is a finetuned version of Phi3-3.8B on a high quality proprietary dataset con
|
|
14 |
|
15 |
### Model Description
|
16 |
|
17 |
-
|
18 |
-
|
19 |
- **Developed by:** [https://www.SciPhi.ai](SciPhi.ai)
|
20 |
|
21 |
### Model Sources
|
@@ -32,7 +30,6 @@ It is a finetuned version of Phi3-3.8B on a high quality proprietary dataset con
|
|
32 |
import json
|
33 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
34 |
|
35 |
-
|
36 |
def triplextract(model, tokenizer, text, entity_types, predicates):
|
37 |
|
38 |
input_format = """
|
@@ -46,29 +43,21 @@ def triplextract(model, tokenizer, text, entity_types, predicates):
|
|
46 |
{text}
|
47 |
"""
|
48 |
|
49 |
-
message = input_format.format(
|
|
|
|
|
|
|
50 |
|
51 |
messages = [{'role': 'user', 'content': message}]
|
52 |
input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt = True, return_tensors="pt").to("cuda")
|
53 |
output = tokenizer.decode(model.generate(input_ids=input_ids, max_length=2048)[0], skip_special_tokens=True)
|
54 |
-
print(output)
|
55 |
return output
|
56 |
|
57 |
-
|
58 |
-
|
59 |
tokenizer = AutoTokenizer.from_pretrained("sciphi/triplex", trust_remote_code=True)
|
60 |
-
model = AutoModelForCausalLM.from_pretrained("sciphi/triplex", trust_remote_code=True)
|
61 |
-
|
62 |
-
|
63 |
-
model.to("cuda")
|
64 |
-
|
65 |
-
model.eval()
|
66 |
-
|
67 |
|
68 |
entity_types = [ "LOCATION", "POSITION", "DATE", "CITY", "COUNTRY", "NUMBER" ]
|
69 |
-
|
70 |
predicates = [ "POPULATION", "AREA" ]
|
71 |
-
|
72 |
text = """
|
73 |
San Francisco,[24] officially the City and County of San Francisco, is a commercial, financial, and cultural center in Northern California.
|
74 |
|
|
|
14 |
|
15 |
### Model Description
|
16 |
|
|
|
|
|
17 |
- **Developed by:** [https://www.SciPhi.ai](SciPhi.ai)
|
18 |
|
19 |
### Model Sources
|
|
|
30 |
import json
|
31 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
32 |
|
|
|
33 |
def triplextract(model, tokenizer, text, entity_types, predicates):
|
34 |
|
35 |
input_format = """
|
|
|
43 |
{text}
|
44 |
"""
|
45 |
|
46 |
+
message = input_format.format(
|
47 |
+
entity_types = json.dumps({"entity_types": entity_types}),
|
48 |
+
predicates = json.dumps({"predicates": predicates}),
|
49 |
+
text = text)
|
50 |
|
51 |
messages = [{'role': 'user', 'content': message}]
|
52 |
input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt = True, return_tensors="pt").to("cuda")
|
53 |
output = tokenizer.decode(model.generate(input_ids=input_ids, max_length=2048)[0], skip_special_tokens=True)
|
|
|
54 |
return output
|
55 |
|
56 |
+
model = AutoModelForCausalLM.from_pretrained("sciphi/triplex", trust_remote_code=True).to('cuda').eval()
|
|
|
57 |
tokenizer = AutoTokenizer.from_pretrained("sciphi/triplex", trust_remote_code=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
58 |
|
59 |
entity_types = [ "LOCATION", "POSITION", "DATE", "CITY", "COUNTRY", "NUMBER" ]
|
|
|
60 |
predicates = [ "POPULATION", "AREA" ]
|
|
|
61 |
text = """
|
62 |
San Francisco,[24] officially the City and County of San Francisco, is a commercial, financial, and cultural center in Northern California.
|
63 |
|