Spaces:
Sleeping
Sleeping
Update inference.py
Browse files- inference.py +14 -12
inference.py
CHANGED
@@ -1,6 +1,8 @@
|
|
1 |
import torch
|
2 |
-
from transformers import AutoTokenizer
|
3 |
from evo_model import EvoTransformerV22
|
|
|
|
|
4 |
|
5 |
# Load Evo model
|
6 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
@@ -12,7 +14,7 @@ evo_model.eval()
|
|
12 |
# Load tokenizer
|
13 |
tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
14 |
|
15 |
-
# 🧠 Evo logic
|
16 |
def get_evo_response(query, context):
|
17 |
combined = query + " " + context
|
18 |
inputs = tokenizer(combined, return_tensors="pt", truncation=True, padding="max_length", max_length=128)
|
@@ -20,24 +22,24 @@ def get_evo_response(query, context):
|
|
20 |
|
21 |
with torch.no_grad():
|
22 |
logits = evo_model(input_ids)
|
23 |
-
pred = torch.
|
24 |
|
25 |
-
return f"Evo suggests: Option {pred + 1}"
|
26 |
|
27 |
-
# 🤖 GPT-3.5 comparison
|
28 |
-
|
29 |
-
|
30 |
|
31 |
def get_gpt_response(query, context):
|
32 |
try:
|
33 |
prompt = f"Context: {context}\n\nQuestion: {query}\n\nAnswer:"
|
34 |
-
response =
|
35 |
model="gpt-3.5-turbo",
|
36 |
-
messages=[
|
|
|
|
|
37 |
temperature=0.3
|
38 |
)
|
39 |
-
return response
|
40 |
except Exception as e:
|
41 |
return f"Error from GPT: {e}"
|
42 |
-
|
43 |
-
#
|
|
|
1 |
import torch
|
2 |
+
from transformers import AutoTokenizer
|
3 |
from evo_model import EvoTransformerV22
|
4 |
+
from openai import OpenAI
|
5 |
+
import os
|
6 |
|
7 |
# Load Evo model
|
8 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
|
|
14 |
# Load tokenizer
|
15 |
tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
16 |
|
17 |
+
# 🧠 Evo logic (binary classification with sigmoid)
|
18 |
def get_evo_response(query, context):
|
19 |
combined = query + " " + context
|
20 |
inputs = tokenizer(combined, return_tensors="pt", truncation=True, padding="max_length", max_length=128)
|
|
|
22 |
|
23 |
with torch.no_grad():
|
24 |
logits = evo_model(input_ids)
|
25 |
+
pred = int(torch.sigmoid(logits).item() > 0.5)
|
26 |
|
27 |
+
return f"Evo suggests: Option {pred + 1}"
|
28 |
|
29 |
+
# 🤖 GPT-3.5 comparison using openai>=1.0.0
|
30 |
+
openai_api_key = os.environ.get("OPENAI_API_KEY", "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") # Replace with real key or set via HF secrets
|
31 |
+
client = OpenAI(api_key=openai_api_key)
|
32 |
|
33 |
def get_gpt_response(query, context):
|
34 |
try:
|
35 |
prompt = f"Context: {context}\n\nQuestion: {query}\n\nAnswer:"
|
36 |
+
response = client.chat.completions.create(
|
37 |
model="gpt-3.5-turbo",
|
38 |
+
messages=[
|
39 |
+
{"role": "user", "content": prompt}
|
40 |
+
],
|
41 |
temperature=0.3
|
42 |
)
|
43 |
+
return response.choices[0].message.content.strip()
|
44 |
except Exception as e:
|
45 |
return f"Error from GPT: {e}"
|
|
|
|