Create handler.py
Browse files- handler.py +38 -0
handler.py
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import pipeline
|
2 |
+
from greenery import parse
|
3 |
+
from greenery.parse import NoMatch
|
4 |
+
from listener import Listener, ListenerOutput
|
5 |
+
import time
|
6 |
+
import json
|
7 |
+
import torch
|
8 |
+
|
9 |
+
class EndpointHandler:
|
10 |
+
def __init__(self, path=""):
|
11 |
+
self.listener = Listener(path, {
|
12 |
+
"do_sample": True,
|
13 |
+
"max_new_tokens": 128,
|
14 |
+
"top_p": 0.9,
|
15 |
+
"num_return_sequences": 500,
|
16 |
+
"num_beams": 1
|
17 |
+
}, device="cuda" if torch.cuda.is_available() else "cpu")
|
18 |
+
|
19 |
+
def __call__(self, data):
|
20 |
+
# get inputs
|
21 |
+
inp = data.pop("inputs", None)
|
22 |
+
spec = inp["spec"]
|
23 |
+
true_program = inp["true_program"]
|
24 |
+
|
25 |
+
start = time.time()
|
26 |
+
outputs = self.listener.synthesize([[(s["string"], s["label"]) for s in spec]], return_scores=True)
|
27 |
+
consistent_program_scores = [outputs.decoded_scores[0][i] for i in outputs.idx[0]]
|
28 |
+
consistent_programs = [outputs.decoded[0][i] for i in outputs.idx[0]]
|
29 |
+
sorted_programs = sorted(set(zip(consistent_program_scores, consistent_programs)), reverse=True, key=lambda x: x[0])
|
30 |
+
end = time.time()
|
31 |
+
|
32 |
+
return {
|
33 |
+
"guess": sorted_programs[0][1],
|
34 |
+
"top_1_success": parse(sorted_programs[0][1]).equivalent(parse(true_program)),
|
35 |
+
"top_1_score": sorted_programs[0][0],
|
36 |
+
"top_5_success": any([parse(p).equivalent(parse(true_program)) for _, p in sorted_programs[:5]]),
|
37 |
+
"time": end - start
|
38 |
+
}
|