New estimators and data io fix (#8)

* actually change python version

* dummy classifiers and sklearn lower bound change

* test fix

* test fix

* dev

* early sklearn version fixes

* all interval classifiers

* dummy and conversion bugfix

* version

* test

* testing update

* version

* pandas

* stop all workflows failing

* copy check

* new estimators and data loading fix

* new ver

* ci

* gcc

* only macos

* codecov check

Author: MatthewMiddlehurst

.github/workflows/release.yml

@@ -73,10 +73,10 @@ jobs:
 
       - if: matrix.os == 'windows-latest'
         name: Windows install
-        run: python -m pip install "${env:WHEELNAME}[extras,dev]"
+        run: python -m pip install "${env:WHEELNAME}[dev,extras,unstable_extras]"
       - if: matrix.os != 'windows-latest'
         name: Unix install
-        run: python -m pip install "${{ env.WHEELNAME }}[extras,dev]"
+        run: python -m pip install "${{ env.WHEELNAME }}[dev,extras,unstable_extras]"
 
       - name: Tests
         run: python -m pytest

.github/workflows/tests.yml

@@ -43,7 +43,7 @@ jobs:
           python-version: ${{ matrix.python-version }}
 
       - name: Install
-        run: python -m pip install .[dev,extras]
+        run: python -m pip install .[dev,extras,unstable_extras]
 
       - name: Tests
         run: python -m pytest
@@ -63,7 +63,7 @@ jobs:
         run: echo "NUMBA_DISABLE_JIT=1" >> $GITHUB_ENV
 
       - name: Install
-        run: python -m pip install .[dev,extras]
+        run: python -m pip install .[dev,extras,unstable_extras]
 
       - name: Tests
         run: python -m pytest --cov=tsml --cov-report=xml

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tsml"
-version = "0.0.7"
+version = "0.1.0"
 description = "A toolkit for time series machine learning algorithms."
 authors = [
     {name = "Matthew Middlehurst", email = "m.middlehurst@uea.ac.uk"},
@@ -43,9 +43,13 @@ dependencies = [
 
 [project.optional-dependencies]
 extras = [
-    "pycatch22>=0.4.2",
     "pyfftw>=0.12.0",
     "statsmodels>=0.12.1",
+    "wildboar>=1.1.0",
+]
+unstable_extras = [
+    "mrsqm>=0.0.1 ; platform_system == 'Darwin'",  # requires gcc and fftw to be installed for Windows and some other OS (see http://www.fftw.org/index.html)
+    "pycatch22>=0.4.2",  # Known to fail installation on some setups
 ]
 dev = [
     "pre-commit",

tsml/__init__.py

@@ -1,4 +1,4 @@
 # -*- coding: utf-8 -*-
 """tsml."""
 
-__version__ = "0.0.7"
+__version__ = "0.1.0"

tsml/datasets/_data_io.py

@@ -30,9 +30,9 @@ def load_from_ts_file(
     data is unequal length, a list of 2D numpy arrays will be returned. If labels are
     present, they will be returned as well as the data.
 
-    The only mandatory tags in the loaded file are @equallength and one of
-    @targetlabels or @classlabels. Other details can be inferred, though some error
-    checking will be done if they are present.
+    The only mandatory tags in the loaded file are one of @targetlabels or @classlabels.
+    Other details can be inferred, though some error checking will be done if they are
+    present.
 
     Parameters
     ----------
@@ -304,11 +304,9 @@ def load_from_ts_file(
                     "required."
                 )
 
-            # Equal length tag is required.
+            # Assume equal length if no tag.
             if not equallength_tag:
-                raise IOError(
-                    "Unable to read .ts file. The @equallength tag is required."
-                )
+                equallength = True
 
             n_instances = len(lines) - data_start_line
             data_dims = len(first_line) - 1 if has_labels else len(first_line)
@@ -384,13 +382,19 @@ def load_from_ts_file(
                 )
 
             dimensions = line[:data_dims]
-            if not equallength:
-                data_length = len(dimensions[0].strip().split(","))
+            split = dimensions[0].strip().split(",")
+            length = len(split)
+            if equallength and length != data_length:
+                raise IOError(
+                    "Unable to read .ts file. Inconsistent number of channels."
+                    f"Expected {data_dims} but read {read_dims} on line {data_idx}."
+                )
 
             # Process the data for each channel
-            series = np.zeros((data_dims, data_length), dtype=X_dtype)
-            for i in range(data_dims):
-                series[i, :] = dimensions[i].strip().split(",")
+            series = np.zeros((data_dims, length), dtype=X_dtype)
+            series[0, :] = split
+            for n in range(1, data_dims):
+                series[n, :] = dimensions[n].strip().split(",")
 
             X[data_idx] = series
 

tsml/datasets/tests/__init__.py

@@ -0,0 +1,2 @@
+# -*- coding: utf-8 -*-
+"""Testing for data loaders."""