deeprails 0.2.0__tar.gz → 0.2.1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: deeprails
3
- Version: 0.2.0
3
+ Version: 0.2.1
4
4
  Summary: Python SDK for interacting with the DeepRails API
5
5
  Project-URL: Homepage, https://deeprails.com
6
6
  Project-URL: Documentation, https://docs.deeprails.com
@@ -128,11 +128,13 @@ except Exception as e:
128
128
 
129
129
  ## Available Metrics
130
130
 
131
- - `correctness`: Evaluates factual accuracy of the output
132
- - `completeness`: Checks if the output addresses all aspects of the prompt
133
- - `harmfulness`: Detects potentially harmful content
134
- - `bias`: Identifies biased language or reasoning
135
- - And more...
131
+ - `correctness`: Measures factual accuracy by evaluating whether each claim in the output is true and verifiable.
132
+ - `completeness`: Assesses whether the response addresses all necessary parts of the prompt with sufficient detail and relevance.
133
+ - `instruction_adherence`: Checks whether the AI followed the explicit instructions in the prompt and system directives.
134
+ - `context_adherence`: Determines whether each factual claim is directly supported by the provided context.
135
+ - `ground_truth_adherence`: Measures how closely the output matches a known correct answer (gold standard).
136
+ - `comprehensive_safety`: Detects and categorizes safety violations across areas like PII, CBRN, hate speech, self-harm, and more.
137
+
136
138
 
137
139
  ## Error Handling
138
140
 
@@ -93,11 +93,13 @@ except Exception as e:
93
93
 
94
94
  ## Available Metrics
95
95
 
96
- - `correctness`: Evaluates factual accuracy of the output
97
- - `completeness`: Checks if the output addresses all aspects of the prompt
98
- - `harmfulness`: Detects potentially harmful content
99
- - `bias`: Identifies biased language or reasoning
100
- - And more...
96
+ - `correctness`: Measures factual accuracy by evaluating whether each claim in the output is true and verifiable.
97
+ - `completeness`: Assesses whether the response addresses all necessary parts of the prompt with sufficient detail and relevance.
98
+ - `instruction_adherence`: Checks whether the AI followed the explicit instructions in the prompt and system directives.
99
+ - `context_adherence`: Determines whether each factual claim is directly supported by the provided context.
100
+ - `ground_truth_adherence`: Measures how closely the output matches a known correct answer (gold standard).
101
+ - `comprehensive_safety`: Detects and categorizes safety violations across areas like PII, CBRN, hate speech, self-harm, and more.
102
+
101
103
 
102
104
  ## Error Handling
103
105
 
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
4
4
 
5
5
  [project]
6
6
  name = "deeprails"
7
- version = "0.2.0"
7
+ version = "0.2.1"
8
8
  description = "Python SDK for interacting with the DeepRails API"
9
9
  readme = {file = "README.md", content-type = "text/markdown"}
10
10
  authors = [{name = "Neil Mate", email = "support@deeprails.ai"}]
File without changes
File without changes