Update generate.py
Browse files- generate.py +18 -35
generate.py
CHANGED
@@ -1,47 +1,30 @@
|
|
1 |
-
# generate.py
|
2 |
import torch
|
3 |
-
from transformers import
|
4 |
from evo_model import EvoDecoderModel
|
5 |
-
from search_utils import web_search
|
6 |
|
7 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
8 |
|
9 |
-
# Load tokenizer
|
10 |
-
tokenizer =
|
11 |
-
vocab_size = tokenizer.vocab_size
|
12 |
|
13 |
-
model
|
|
|
|
|
14 |
model.load_state_dict(torch.load("evo_decoder_model.pt", map_location=device))
|
|
|
15 |
model.eval()
|
16 |
|
17 |
-
def generate_response(prompt,
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
web_context = web_search(prompt)
|
22 |
-
context += f"Relevant Info: {web_context}\n"
|
23 |
-
|
24 |
-
input_text = context + f"User: {prompt}\nAssistant:"
|
25 |
-
input_ids = tokenizer.encode(input_text, return_tensors="pt").to(device)
|
26 |
-
|
27 |
-
# Truncate to avoid exceeding model's positional encoding limit
|
28 |
-
if input_ids.size(1) > 512:
|
29 |
-
input_ids = input_ids[:, -512:]
|
30 |
-
|
31 |
-
for _ in range(max_length):
|
32 |
-
with torch.no_grad():
|
33 |
-
logits = model(input_ids)
|
34 |
-
next_token_logits = logits[:, -1, :]
|
35 |
-
|
36 |
-
# Top-k sampling
|
37 |
-
top_k_probs, top_k_indices = torch.topk(next_token_logits, top_k)
|
38 |
-
probs = torch.softmax(top_k_probs, dim=-1)
|
39 |
-
next_token = top_k_indices[0, torch.multinomial(probs, 1).item()].unsqueeze(0).unsqueeze(0)
|
40 |
|
41 |
-
|
|
|
|
|
42 |
|
43 |
-
|
44 |
-
|
45 |
|
46 |
-
|
47 |
-
return output.split("Assistant:")[-1].strip()
|
|
|
1 |
+
# generate.py
|
2 |
import torch
|
3 |
+
from transformers import BertTokenizer
|
4 |
from evo_model import EvoDecoderModel
|
|
|
5 |
|
6 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
7 |
|
8 |
+
# Load tokenizer
|
9 |
+
tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
|
|
10 |
|
11 |
+
# Initialize model architecture
|
12 |
+
vocab_size = tokenizer.vocab_size
|
13 |
+
model = EvoDecoderModel(vocab_size=vocab_size)
|
14 |
model.load_state_dict(torch.load("evo_decoder_model.pt", map_location=device))
|
15 |
+
model.to(device)
|
16 |
model.eval()
|
17 |
|
18 |
+
def generate_response(prompt, max_length=128, use_web=False):
|
19 |
+
with torch.no_grad():
|
20 |
+
input_ids = tokenizer(prompt, return_tensors="pt", padding=False, truncation=True).input_ids.to(device)
|
21 |
+
input_ids = input_ids[:, :128] # ✅ clip to trained length
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
|
23 |
+
logits = model(input_ids)
|
24 |
+
next_token_logits = logits[:, -1, :] # take last token's logits
|
25 |
+
predicted_id = torch.argmax(next_token_logits, dim=-1)
|
26 |
|
27 |
+
output_ids = torch.cat([input_ids, predicted_id.unsqueeze(0)], dim=1)
|
28 |
+
decoded = tokenizer.decode(output_ids[0], skip_special_tokens=True)
|
29 |
|
30 |
+
return decoded
|
|