samagra14wefi
commited on
Commit
•
0f685c5
1
Parent(s):
9696018
Replace asserts with print
Browse files
README.md
CHANGED
@@ -55,13 +55,13 @@ context_string = '''India won the world cup in 1983 and 2011. Australia won the
|
|
55 |
West Indies have won the world cup twice. Sri Lanka, Pakistan and England have won the world cup once.
|
56 |
Evaluate if the facts below are consistent with the statement.'''
|
57 |
|
58 |
-
|
59 |
-
|
60 |
|
61 |
-
|
62 |
-
|
63 |
|
64 |
-
|
65 |
```
|
66 |
|
67 |
### Evaluating Response relevance
|
@@ -75,7 +75,7 @@ response_bad = "We offer a variety of fresh produce including apples, oranges, a
|
|
75 |
score_good = calc_score(inquiry, response_good)
|
76 |
score_bad = calc_score(inquiry, response_bad)
|
77 |
|
78 |
-
|
79 |
|
80 |
```
|
81 |
|
@@ -89,7 +89,7 @@ response_offensive = "The politicians in Country X are all corrupt and stupid."
|
|
89 |
score_clean = calc_score(context, response_clean)
|
90 |
score_offensive = calc_score(context, response_offensive)
|
91 |
|
92 |
-
|
93 |
```
|
94 |
|
95 |
### Comparing Different Language Models
|
@@ -102,7 +102,7 @@ response_bert = "Photosynthesis is a method that converts carbon dioxide into or
|
|
102 |
score_gpt3 = calc_score(context, response_gpt3)
|
103 |
score_bert = calc_score(context, response_bert)
|
104 |
|
105 |
-
|
106 |
```
|
107 |
|
108 |
## Finetuning on your production data
|
|
|
55 |
West Indies have won the world cup twice. Sri Lanka, Pakistan and England have won the world cup once.
|
56 |
Evaluate if the facts below are consistent with the statement.'''
|
57 |
|
58 |
+
response_string_wrong = '''India has won the world cup most number of times.'''
|
59 |
+
response_string_correct = '''Australia has won the world cup most number of times.'''
|
60 |
|
61 |
+
score_wrong = calc_score(context_string, response_string_one)
|
62 |
+
score_correct = calc_score(context_string, response_string_two)
|
63 |
|
64 |
+
print(score_wrong > score_correct)
|
65 |
```
|
66 |
|
67 |
### Evaluating Response relevance
|
|
|
75 |
score_good = calc_score(inquiry, response_good)
|
76 |
score_bad = calc_score(inquiry, response_bad)
|
77 |
|
78 |
+
print(score_good > score_bad)
|
79 |
|
80 |
```
|
81 |
|
|
|
89 |
score_clean = calc_score(context, response_clean)
|
90 |
score_offensive = calc_score(context, response_offensive)
|
91 |
|
92 |
+
print(score_clean > score_offensive)
|
93 |
```
|
94 |
|
95 |
### Comparing Different Language Models
|
|
|
102 |
score_gpt3 = calc_score(context, response_gpt3)
|
103 |
score_bert = calc_score(context, response_bert)
|
104 |
|
105 |
+
print(score_gpt3 > score_bert)
|
106 |
```
|
107 |
|
108 |
## Finetuning on your production data
|