xrtm-eval 0.1.0__py3-none-any.whl → 0.1.2__py3-none-any.whl
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.
- {xrtm_eval-0.1.0.dist-info → xrtm_eval-0.1.2.dist-info}/METADATA +24 -2
- {xrtm_eval-0.1.0.dist-info → xrtm_eval-0.1.2.dist-info}/RECORD +5 -5
- {xrtm_eval-0.1.0.dist-info → xrtm_eval-0.1.2.dist-info}/WHEEL +0 -0
- {xrtm_eval-0.1.0.dist-info → xrtm_eval-0.1.2.dist-info}/licenses/LICENSE +0 -0
- {xrtm_eval-0.1.0.dist-info → xrtm_eval-0.1.2.dist-info}/top_level.txt +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: xrtm-eval
|
|
3
|
-
Version: 0.1.
|
|
3
|
+
Version: 0.1.2
|
|
4
4
|
Summary: The Judge/Scoring engine for XRTM.
|
|
5
5
|
Author-email: XRTM Team <moy@xrtm.org>
|
|
6
6
|
License: Apache-2.0
|
|
@@ -23,15 +23,27 @@ Dynamic: license-file
|
|
|
23
23
|
|
|
24
24
|
[](https://opensource.org/licenses/Apache-2.0)
|
|
25
25
|
[](https://www.python.org/downloads/)
|
|
26
|
+
[](https://pypi.org/project/xrtm-eval/)
|
|
26
27
|
|
|
27
28
|
**The Judge for XRTM.**
|
|
28
29
|
|
|
29
30
|
`xrtm-eval` is the rigorous scoring engine used to grade probabilistic forecasts. It operates independently of the inference engine to ensure objective evaluation.
|
|
30
31
|
|
|
32
|
+
## Part of the XRTM Ecosystem
|
|
33
|
+
|
|
34
|
+
```
|
|
35
|
+
Layer 4: xrtm-train → (imports all)
|
|
36
|
+
Layer 3: xrtm-forecast → (imports eval, data)
|
|
37
|
+
Layer 2: xrtm-eval → (imports data) ← YOU ARE HERE
|
|
38
|
+
Layer 1: xrtm-data → (zero dependencies)
|
|
39
|
+
```
|
|
40
|
+
|
|
41
|
+
`xrtm-eval` provides scoring metrics AND trust primitives used by the forecast engine.
|
|
42
|
+
|
|
31
43
|
## Installation
|
|
32
44
|
|
|
33
45
|
```bash
|
|
34
|
-
|
|
46
|
+
pip install xrtm-eval
|
|
35
47
|
```
|
|
36
48
|
|
|
37
49
|
## Core Primitives
|
|
@@ -54,6 +66,16 @@ score = evaluator.score(prediction=0.7, ground_truth=1)
|
|
|
54
66
|
### 2. Expected Calibration Error (ECE)
|
|
55
67
|
Use the `ExpectedCalibrationErrorEvaluator` to measure the gap between confidence and accuracy across bin buckets.
|
|
56
68
|
|
|
69
|
+
### 3. Epistemic Trust Primitives (v0.1.1+)
|
|
70
|
+
`xrtm-eval` now includes trust scoring infrastructure:
|
|
71
|
+
|
|
72
|
+
```python
|
|
73
|
+
from xrtm.eval.core.epistemics import IntegrityGuardian, SourceTrustRegistry
|
|
74
|
+
|
|
75
|
+
registry = SourceTrustRegistry()
|
|
76
|
+
guardian = IntegrityGuardian(registry)
|
|
77
|
+
```
|
|
78
|
+
|
|
57
79
|
## Development
|
|
58
80
|
|
|
59
81
|
Prerequisites:
|
|
@@ -15,8 +15,8 @@ xrtm/eval/kit/eval/resilience.py,sha256=q95enu-mglq-S5MSIiO9kDN2Kn9898LvuxNeqo90
|
|
|
15
15
|
xrtm/eval/kit/eval/viz.py,sha256=LGss9nuGpubJdBbXRlQU2KiGqouTNnWCDyXvKAct3pE,2773
|
|
16
16
|
xrtm/eval/schemas/__init__.py,sha256=6nwuE6LpwosNhFR-xRPPWcGzP1ZxoXhN-vWrzK4XfbQ,75
|
|
17
17
|
xrtm/eval/schemas/forecast.py,sha256=vA5V5RTo7k2ZV3THs2h3fxmdusS06Vw92H5WKJmXVzA,624
|
|
18
|
-
xrtm_eval-0.1.
|
|
19
|
-
xrtm_eval-0.1.
|
|
20
|
-
xrtm_eval-0.1.
|
|
21
|
-
xrtm_eval-0.1.
|
|
22
|
-
xrtm_eval-0.1.
|
|
18
|
+
xrtm_eval-0.1.2.dist-info/licenses/LICENSE,sha256=BexUTTsX5WlzyJ0Tqajo1h_LFYfCtfFgWdRaGltpm5I,11328
|
|
19
|
+
xrtm_eval-0.1.2.dist-info/METADATA,sha256=h4-gh_4ofAluGCFqWZ6lHbv-absbh_-fs94_jqgt8W4,2673
|
|
20
|
+
xrtm_eval-0.1.2.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
|
|
21
|
+
xrtm_eval-0.1.2.dist-info/top_level.txt,sha256=Jz-i0a9P8GVrIR9KJTT-9wT95E1brww6U5o2QViAt20,5
|
|
22
|
+
xrtm_eval-0.1.2.dist-info/RECORD,,
|
|
File without changes
|
|
File without changes
|
|
File without changes
|