Spaces:
Sleeping
Sleeping
Victoria Oberascher
commited on
Commit
·
2278280
1
Parent(s):
f3b0d73
fix typos in readme
Browse files
README.md
CHANGED
@@ -34,12 +34,12 @@ To get started with horizon-metrics, make sure you have the necessary dependenci
|
|
34 |
This is how you can quickly evaluate your horizon prediction models using SEA-AI/horizon-metrics:
|
35 |
|
36 |
> [!IMPORTANT]
|
37 |
-
> The vertical_fov_degrees and
|
38 |
|
39 |
> [!IMPORTANT]
|
40 |
> The horizon metric should be calculated per sequence. Make sure that the vertical_fov and height are consistent across the inputs and do not change.
|
41 |
|
42 |
-
##### Use
|
43 |
|
44 |
```python
|
45 |
ground_truth_points = [[[0.0, 0.5384765625], [1.0, 0.4931640625]],
|
@@ -90,7 +90,7 @@ module.compute()
|
|
90 |
|
91 |
```
|
92 |
|
93 |
-
This is output the
|
94 |
|
95 |
```console
|
96 |
{'average_slope_error': 0.39394822776758726,
|
|
|
34 |
This is how you can quickly evaluate your horizon prediction models using SEA-AI/horizon-metrics:
|
35 |
|
36 |
> [!IMPORTANT]
|
37 |
+
> The vertical_fov_degrees and height parameters are required. The default value for roll_threshold is 0.5 and for pitch_threshold it is 0.1.
|
38 |
|
39 |
> [!IMPORTANT]
|
40 |
> The horizon metric should be calculated per sequence. Make sure that the vertical_fov and height are consistent across the inputs and do not change.
|
41 |
|
42 |
+
##### Use artificial data for testing
|
43 |
|
44 |
```python
|
45 |
ground_truth_points = [[[0.0, 0.5384765625], [1.0, 0.4931640625]],
|
|
|
90 |
|
91 |
```
|
92 |
|
93 |
+
This is output the evaluation metrics for your horizon prediction model:
|
94 |
|
95 |
```console
|
96 |
{'average_slope_error': 0.39394822776758726,
|