Working on EMNIST Lines dataset

11 files changed, 133 insertions, 0 deletions

diff --git a/tests/support/__init__.py b/tests/support/__init__.py
new file mode 100644
index 0000000..a265ede
--- /dev/null
+++ b/tests/support/__init__.py
@@ -0,0 +1,2 @@
+"""Support file modules."""
+from .create_emnist_support_files import create_emnist_support_files
diff --git a/tests/support/create_emnist_lines_support_files.py b/tests/support/create_emnist_lines_support_files.py
new file mode 100644
index 0000000..9abe143
--- /dev/null
+++ b/tests/support/create_emnist_lines_support_files.py
@@ -0,0 +1,51 @@
+"""Module for creating EMNIST Lines test support files."""
+# flake8: noqa: S106
+
+from pathlib import Path
+import shutil
+
+import numpy as np
+
+from text_recognizer.datasets import EmnistLinesDataset
+import text_recognizer.util as util
+
+
+SUPPORT_DIRNAME = Path(__file__).parents[0].resolve() / "emnist_lines"
+
+
+def create_emnist_lines_support_files() -> None:
+    """Create EMNIST Lines test images."""
+    shutil.rmtree(SUPPORT_DIRNAME, ignore_errors=True)
+    SUPPORT_DIRNAME.mkdir()
+
+    # TODO: maybe have to add args to dataset.
+    dataset = EmnistLinesDataset(
+        init_token="<sos>",
+        pad_token="_",
+        eos_token="<eos>",
+        transform=[{"type": "ToTensor", "args": {}}],
+        target_transform=[
+            {
+                "type": "AddTokens",
+                "args": {"init_token": "<sos>", "pad_token": "_", "eos_token": "<eos>"},
+            }
+        ],
+    )  # nosec: S106
+    dataset.load_or_generate_data()
+
+    for index in [5, 7, 9]:
+        image, target = dataset[index]
+        if len(image.shape) == 3:
+            image = image.squeeze(0)
+        print(image.sum(), image.dtype)
+
+        label = "".join(dataset.mapper(label) for label in target[1:]).strip(
+            dataset.mapper.pad_token
+        )
+        print(label)
+        image = image.numpy()
+        util.write_image(image, str(SUPPORT_DIRNAME / f"{label}.png"))
+
+
+if __name__ == "__main__":
+    create_emnist_lines_support_files()
diff --git a/tests/support/create_emnist_support_files.py b/tests/support/create_emnist_support_files.py
new file mode 100644
index 0000000..f9ff030
--- /dev/null
+++ b/tests/support/create_emnist_support_files.py
@@ -0,0 +1,30 @@
+"""Module for creating EMNIST test support files."""
+from pathlib import Path
+import shutil
+
+from text_recognizer.datasets import EmnistDataset
+from text_recognizer.util import write_image
+
+SUPPORT_DIRNAME = Path(__file__).parents[0].resolve() / "emnist"
+
+
+def create_emnist_support_files() -> None:
+    """Create support images for test of CharacterPredictor class."""
+    shutil.rmtree(SUPPORT_DIRNAME, ignore_errors=True)
+    SUPPORT_DIRNAME.mkdir()
+
+    dataset = EmnistDataset(train=False)
+    dataset.load_or_generate_data()
+
+    for index in [5, 7, 9]:
+        image, label = dataset[index]
+        if len(image.shape) == 3:
+            image = image.squeeze(0)
+        image = image.numpy()
+        label = dataset.mapper(int(label))
+        print(index, label)
+        write_image(image, str(SUPPORT_DIRNAME / f"{label}.png"))
+
+
+if __name__ == "__main__":
+    create_emnist_support_files()
diff --git a/tests/support/create_iam_lines_support_files.py b/tests/support/create_iam_lines_support_files.py
new file mode 100644
index 0000000..50f9e3d
--- /dev/null
+++ b/tests/support/create_iam_lines_support_files.py
@@ -0,0 +1,50 @@
+"""Module for creating IAM Lines test support files."""
+# flake8: noqa
+from pathlib import Path
+import shutil
+
+import numpy as np
+
+from text_recognizer.datasets import IamLinesDataset
+import text_recognizer.util as util
+
+
+SUPPORT_DIRNAME = Path(__file__).parents[0].resolve() / "iam_lines"
+
+
+def create_emnist_lines_support_files() -> None:
+    """Create IAM Lines test images."""
+    shutil.rmtree(SUPPORT_DIRNAME, ignore_errors=True)
+    SUPPORT_DIRNAME.mkdir()
+
+    # TODO: maybe have to add args to dataset.
+    dataset = IamLinesDataset(
+        init_token="<sos>",
+        pad_token="_",
+        eos_token="<eos>",
+        transform=[{"type": "ToTensor", "args": {}}],
+        target_transform=[
+            {
+                "type": "AddTokens",
+                "args": {"init_token": "<sos>", "pad_token": "_", "eos_token": "<eos>"},
+            }
+        ],
+    )
+    dataset.load_or_generate_data()
+
+    for index in [0, 1, 3]:
+        image, target = dataset[index]
+        if len(image.shape) == 3:
+            image = image.squeeze(0)
+        print(image.sum(), image.dtype)
+
+        label = "".join(dataset.mapper(label) for label in target[1:]).strip(
+            dataset.mapper.pad_token
+        )
+        print(label)
+        image = image.numpy()
+        util.write_image(image, str(SUPPORT_DIRNAME / f"{label}.png"))
+
+
+if __name__ == "__main__":
+    create_emnist_lines_support_files()
diff --git a/tests/support/emnist_lines/Knox Ky<eos>.png b/tests/support/emnist_lines/Knox Ky<eos>.png
new file mode 100644
index 0000000..b7d0618
--- /dev/null
+++ b/tests/support/emnist_lines/Knox Ky<eos>.png
diff --git a/tests/support/emnist_lines/ancillary beliefs and<eos>.png b/tests/support/emnist_lines/ancillary beliefs and<eos>.png
new file mode 100644
index 0000000..14a8cf3
--- /dev/null
+++ b/tests/support/emnist_lines/ancillary beliefs and<eos>.png
diff --git a/tests/support/emnist_lines/they<eos>.png b/tests/support/emnist_lines/they<eos>.png
new file mode 100644
index 0000000..7f05951
--- /dev/null
+++ b/tests/support/emnist_lines/they<eos>.png
diff --git a/tests/support/iam_lines/He rose from his breakfast-nook bench<eos>.png b/tests/support/iam_lines/He rose from his breakfast-nook bench<eos>.png
new file mode 100644
index 0000000..6eeb642
--- /dev/null
+++ b/tests/support/iam_lines/He rose from his breakfast-nook bench<eos>.png
diff --git a/tests/support/iam_lines/and came into the livingroom, where<eos>.png b/tests/support/iam_lines/and came into the livingroom, where<eos>.png
new file mode 100644
index 0000000..4974cf8
--- /dev/null
+++ b/tests/support/iam_lines/and came into the livingroom, where<eos>.png
diff --git a/tests/support/iam_lines/his entrance. He came, almost falling<eos>.png b/tests/support/iam_lines/his entrance. He came, almost falling<eos>.png
new file mode 100644
index 0000000..a731245
--- /dev/null
+++ b/tests/support/iam_lines/his entrance. He came, almost falling<eos>.png
diff --git a/tests/support/iam_paragraphs/a01-000u.jpg b/tests/support/iam_paragraphs/a01-000u.jpg
new file mode 100644
index 0000000..d9753b6
--- /dev/null
+++ b/tests/support/iam_paragraphs/a01-000u.jpg