Update handler.py
Browse files- handler.py +4 -5
handler.py
CHANGED
@@ -6,10 +6,9 @@ import torch.cuda
|
|
6 |
|
7 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
8 |
LOGGER = logging.getLogger(__name__)
|
9 |
-
|
10 |
class EndpointHandler():
|
11 |
def __init__(self, path=""):
|
12 |
-
self.model = AutoModelForCausalLM.from_pretrained("Ozgur98/pushed_model_mosaic_small")
|
13 |
self.tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b")
|
14 |
# Load the Lora model
|
15 |
|
@@ -18,13 +17,13 @@ class EndpointHandler():
|
|
18 |
Args:
|
19 |
data (Dict): The payload with the text prompt and generation parameters.
|
20 |
"""
|
21 |
-
print("CALLED")
|
22 |
LOGGER.info(data)
|
23 |
# Forward
|
24 |
LOGGER.info(f"Start generation.")
|
25 |
-
tokenized_example = tokenizer(data, return_tensors='pt')
|
26 |
outputs = self.model.generate(tokenized_example['input_ids'].to('cuda:0'), max_new_tokens=100, do_sample=True, top_k=10, top_p = 0.95)
|
|
|
27 |
# Postprocess
|
28 |
-
answer = tokenizer.batch_decode(outputs, skip_special_tokens=True)
|
29 |
prompt = answer[0].rstrip()
|
30 |
return prompt
|
|
|
6 |
|
7 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
8 |
LOGGER = logging.getLogger(__name__)
|
|
|
9 |
class EndpointHandler():
|
10 |
def __init__(self, path=""):
|
11 |
+
self.model = AutoModelForCausalLM.from_pretrained("Ozgur98/pushed_model_mosaic_small", trust_remote_code=True).to(device='cuda:0', dtype=torch.bfloat16)
|
12 |
self.tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b")
|
13 |
# Load the Lora model
|
14 |
|
|
|
17 |
Args:
|
18 |
data (Dict): The payload with the text prompt and generation parameters.
|
19 |
"""
|
|
|
20 |
LOGGER.info(data)
|
21 |
# Forward
|
22 |
LOGGER.info(f"Start generation.")
|
23 |
+
tokenized_example = self.tokenizer(data, return_tensors='pt')
|
24 |
outputs = self.model.generate(tokenized_example['input_ids'].to('cuda:0'), max_new_tokens=100, do_sample=True, top_k=10, top_p = 0.95)
|
25 |
+
|
26 |
# Postprocess
|
27 |
+
answer = self.tokenizer.batch_decode(outputs, skip_special_tokens=True)
|
28 |
prompt = answer[0].rstrip()
|
29 |
return prompt
|