bump version to 4.0.0, update changelog and citation files

Author: cwmeijer

.zenodo.json

@@ -1,58 +1,71 @@
-{
-    "creators": [
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "van Kuppevelt, Dafne"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Meijer, Christiaan"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Huber, Florian",
-            "orcid": "0000-0002-3535-9406"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "van Hees, Vincent",
-            "orcid": "0000-0003-0182-9008"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Solino Fernandez, Breixo"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Bos, Patrick"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Spaaks, Jurriaan"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Kuzak, Mateusz",
-            "orcid": "0000-0003-0087-6021"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "Hidding, Johan"
-        },
-        {
-            "affiliation": "Netherlands eScience Center",
-            "name": "van der Ploeg, Atze"
-        }
-    ],
-    "description": "The goal of mcfly is to ease the use of deep learning technology for time series classification. The advantage of deep learning is that it can handle raw data directly, without the need to compute signal features. Deep learning does not require expert domain knowledge about the data, and has been shown to be competitive with conventional machine learning techniques. As an example, you can apply mcfly on accelerometer data for activity classification.",
-    "keywords": [
-        "machine learning",
-        "deep learning",
-        "time series",
-        "automated machine learning"
-    ],
-    "license": {
-        "id": "Apache-2.0"
-    },
-    "title": "mcfly: deep learning for time series"
-}
+{
+  "creators": [
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "van Kuppevelt, Dafne"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Meijer, Christiaan",
+      "orcid": "0000-0002-5529-5761"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Huber, Florian",
+      "orcid": "0000-0002-3535-9406"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "van Hees, Vincent",
+      "orcid": "0000-0003-0182-9008"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Solino Fernandez, Breixo"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Bos, Patrick"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Spaaks, Jurriaan"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Kuzak, Mateusz",
+      "orcid": "0000-0003-0087-6021"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Hidding, Johan"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "van der Ploeg, Atze"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Lüken, Malte",
+      "orcid": "0000-0001-7095-203X"
+    },
+    {
+      "affiliation": "Netherlands eScience Center",
+      "name": "Lyashevska, Olga",
+      "orcid": "0000-0002-8686-8550"
+    }
+  ],
+  "description": "The goal of mcfly is to ease the use of deep learning technology for time series classification. The advantage of deep learning is that it can handle raw data directly, without the need to compute signal features. Deep learning does not require expert domain knowledge about the data, and has been shown to be competitive with conventional machine learning techniques. As an example, you can apply mcfly on accelerometer data for activity classification.",
+  "keywords": [
+    "machine learning",
+    "deep learning",
+    "time series",
+    "automated machine learning"
+  ],
+  "license": {
+    "id": "Apache-2.0"
+  },
+  "publication_date": "2022-12-21",
+  "title": "mcfly: deep learning for time series",
+  "version": "4.0.0"
+}

CHANGELOG.md

@@ -1,5 +1,10 @@
 # Change Log
 
+## v4.0.0
+- Add regression functionality
+- Fix Dataset/data generator support
+- Dropped support for python 3.6, added support for python 3.9 and 3.10
+
 ## v3.1.0
 - Added support for training with Datasets, generators, and other data types supported by Keras [#211](https://github.com/NLeSC/mcfly/issues/211)
 - Added separate classes for each model type, also allowing easier extension by own custom models [#239](https://github.com/NLeSC/mcfly/pull/239)

CITATION.cff

@@ -60,7 +60,7 @@ authors:
     given-names: Olga
     orcid: "https://orcid.org/0000-0002-8686-8550"
 cff-version: "1.0.3"
-date-released: 2019-11-05
+date-released: 2022-12-21
 doi: "10.5281/zenodo.596127"
 keywords:
   - "machine learning"
@@ -70,7 +70,7 @@ keywords:
 license: "Apache-2.0"
 message: "If you use this software, please cite it using these metadata."
 title: "mcfly: deep learning for time series"
-version: "3.0.0"
+version: "4.0.0"
 references:
   - type: article
     title: "Mcfly: Automated deep learning on time series"

mcfly/_version.py

@@ -15,4 +15,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 #
-__version__ = '3.1.0'
+__version__ = '4.0.0'

setup.py

@@ -22,7 +22,7 @@ def read(fname):
         'Programming Language :: Python',
     ],
     test_suite="tests",
-    python_requires='>=3.6',
+    python_requires='>=3.7',
     install_requires=[
         "numpy",
         "scikit-learn>=0.15.0",