🙈 fix typo in NaNMinMaxDownsampler

README.md

@@ -32,7 +32,7 @@ Extremely fast **time series downsampling 📈** for visualization, written in R
 - **Flexible**: works on any type of data
   - supported datatypes are
     - for `x`: `f32`, `f64`, `i16`, `i32`, `i64`, `u16`, `u32`, `u64`, `datetime64`, `timedelta64`
-    - for `y`: `f16`, `f32`, `f64`, `i8`, `i16`, `i32`, `i64`, `u8`, `u16`, `u32`, `u64`, `datetime64`, `timedelta64`, `bool`<sup>\*</sup>
+    - for `y`: `f16`, `f32`, `f64`, `i8`, `i16`, `i32`, `i64`, `u8`, `u16`, `u32`, `u64`, `datetime64`, `timedelta64`, `bool`
     <details>
       <summary><i>!! 🚀 <code>f16</code> <a href="https://github.com/jvdd/argminmax">argminmax</a> is 200-300x faster than numpy</i></summary>
       In contrast with all other data types above, <code>f16</code> is *not* hardware supported (i.e., no instructions for f16) by most modern CPUs!! <br>
@@ -42,8 +42,6 @@ Extremely fast **time series downsampling 📈** for visualization, written in R
     </details>
 - **Easy to use**: simple & flexible API
 
-<sup>\*</sup><i>When `NaN`s need to be retained in the downsampling, the only supported datatypes for `y` are `f16`, `f32` and `f64`</i>
-
 ## Install
 
 ```bash
@@ -126,11 +124,14 @@ This library supports two `NaN`-policies:
 | `MinMaxLTTBDownsampler` | `NaNMinMaxLTTBDownsampler` |
 |       `LTTBDownsampler` |                            |
 
+> Note that NaNs are not supported for `x`-data.
+
 ## Limitations & assumptions 🚨
 
 Assumes;
 
 1. `x`-data is (non-strictly) monotonic increasing (i.e., sorted)
+2. no `NaN`s in `x`-data
 
 ---
 

tests/test_algos_python_compliance.py

@@ -6,7 +6,7 @@
     M4Downsampler,
     MinMaxDownsampler,
     NaNM4Downsampler,
-    NanMinMaxDownsampler,
+    NaNMinMaxDownsampler,
 )
 from tsdownsample._python.downsamplers import (
     LTTB_py,
@@ -24,7 +24,7 @@
         (M4Downsampler(), M4_py()),
         (LTTBDownsampler(), LTTB_py()),
         # Include NaN downsamplers
-        (NanMinMaxDownsampler(), NaNMinMax_py()),
+        (NaNMinMaxDownsampler(), NaNMinMax_py()),
         (NaNM4Downsampler(), NaNM4_py()),
     ],
 )
@@ -48,7 +48,7 @@ def test_resampler_accordance(rust_python_pair, n, n_out):
 
 @pytest.mark.parametrize(
     "rust_python_pair",
-    [(NanMinMaxDownsampler(), NaNMinMax_py()), (NaNM4Downsampler(), NaNM4_py())],
+    [(NaNMinMaxDownsampler(), NaNMinMax_py()), (NaNM4Downsampler(), NaNM4_py())],
 )
 @pytest.mark.parametrize("n", [10_000, 10_032, 20_321, 23_489])
 @pytest.mark.parametrize("n_random_nans", [100, 200, 500, 2000, 5000])

tests/test_tsdownsample.py

@@ -11,7 +11,7 @@
     MinMaxDownsampler,
     MinMaxLTTBDownsampler,
     NaNM4Downsampler,
-    NanMinMaxDownsampler,
+    NaNMinMaxDownsampler,
     NaNMinMaxLTTBDownsampler,
 )
 from tsdownsample.downsampling_interface import (
@@ -31,7 +31,7 @@
 ]
 
 RUST_NAN_DOWNSAMPLERS = [
-    NanMinMaxDownsampler(),
+    NaNMinMaxDownsampler(),
     NaNM4Downsampler(),
     NaNMinMaxLTTBDownsampler(),
 ]
@@ -367,7 +367,7 @@ def test_nan_minmax_downsampler():
     """Test NaN downsamplers."""
     arr = np.random.randn(50_000)
     arr[::5] = np.nan
-    s_downsampled = NanMinMaxDownsampler().downsample(arr, n_out=100)
+    s_downsampled = NaNMinMaxDownsampler().downsample(arr, n_out=100)
     arr_downsampled = arr[s_downsampled]
     assert np.all(np.isnan(arr_downsampled))
 

tsdownsample/__init__.py

@@ -7,7 +7,7 @@
     MinMaxDownsampler,
     MinMaxLTTBDownsampler,
     NaNM4Downsampler,
-    NanMinMaxDownsampler,
+    NaNMinMaxDownsampler,
     NaNMinMaxLTTBDownsampler,
 )
 
@@ -20,7 +20,7 @@
     "M4Downsampler",
     "LTTBDownsampler",
     "MinMaxLTTBDownsampler",
-    "NanMinMaxDownsampler",
+    "NaNMinMaxDownsampler",
     "NaNM4Downsampler",
     "NaNMinMaxLTTBDownsampler",
 ]

tsdownsample/downsamplers.py

@@ -31,7 +31,7 @@ def _check_valid_n_out(n_out: int):
             raise ValueError("n_out must be even")
 
 
-class NanMinMaxDownsampler(AbstractRustNaNDownsampler):
+class NaNMinMaxDownsampler(AbstractRustNaNDownsampler):
     """Downsampler that uses the MinMax algorithm. If the y data contains NaNs, the
     indices of these NaNs are returned.