upgini 1.2.24a2__tar.gz → 1.2.25__tar.gz

{upgini-1.2.24a2 → upgini-1.2.25}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: upgini
-Version: 1.2.24a2
+Version: 1.2.25
 Summary: Intelligent data search & enrichment for Machine Learning
 Project-URL: Bug Reports, https://github.com/upgini/upgini/issues
 Project-URL: Homepage, https://upgini.com/

{upgini-1.2.24a2 → upgini-1.2.25}/pyproject.toml

@@ -70,6 +70,11 @@ packages = [
     "src/upgini"
 ]
 
+[tool.hatch.build]
+include = [
+  "/src/utils/Roboto-Regular.ttf",
+]
+
 [tool.hatch.envs.default]
 type = "virtual"
 python = "3.10"

upgini-1.2.25/src/upgini/__about__.py

@@ -0,0 +1 @@
+__version__ = "1.2.25"

{upgini-1.2.24a2 → upgini-1.2.25}/src/upgini/features_enricher.py

@@ -3194,9 +3194,8 @@ class FeaturesEnricher(TransformerMixin):
 
         return df
 
-    @staticmethod
     def _add_current_date_as_key(
-        df: pd.DataFrame, search_keys: Dict[str, SearchKey], logger: logging.Logger, bundle: ResourceBundle
+        self, df: pd.DataFrame, search_keys: Dict[str, SearchKey], logger: logging.Logger, bundle: ResourceBundle
     ) -> pd.DataFrame:
         if (
             set(search_keys.values()) == {SearchKey.PHONE}
@@ -3204,9 +3203,7 @@ class FeaturesEnricher(TransformerMixin):
             or set(search_keys.values()) == {SearchKey.HEM}
             or set(search_keys.values()) == {SearchKey.COUNTRY, SearchKey.POSTAL_CODE}
         ):
-            msg = bundle.get("current_date_added")
-            print(msg)
-            logger.warning(msg)
+            self.__log_warning(bundle.get("current_date_added"))
             df[FeaturesEnricher.CURRENT_DATE] = datetime.date.today()
             search_keys[FeaturesEnricher.CURRENT_DATE] = SearchKey.DATE
             converter = DateTimeSearchKeyConverter(FeaturesEnricher.CURRENT_DATE)

{upgini-1.2.24a2 → upgini-1.2.25}/src/upgini/metrics.py

@@ -437,7 +437,8 @@ class EstimatorWrapper:
                                 f"Client cat_feature `{cat_feature}` not found in x columns: {x.columns.to_list()}"
                             )
                     estimator_copy.set_params(
-                        cat_features=[x.columns.get_loc(cat_feature) for cat_feature in cat_features]
+                        # cat_features=[x.columns.get_loc(cat_feature) for cat_feature in cat_features]
+                        cat_features=cat_features
                     )
                 estimator = CatBoostWrapper(**kwargs)
             else:

{upgini-1.2.24a2 → upgini-1.2.25}/src/upgini/resource_bundle/strings.properties

@@ -9,7 +9,7 @@ search_stopped=Search request stopped
 polling_search_task=\nRunning search request, search_id={}
 polling_unregister_information=We'll send email notification once it's completed, just use your personal api_key from profile.upgini.com
 ads_upload_finish=Thank you for your submission!\nWe'll check your data sharing proposal and get back to you
-demo_dataset_info=Demo training dataset detected. Registration for an API key is not required.
+demo_dataset_info=Demo training dataset detected. Registration for an API key is not required.\n
 transform_usage_info=You use Trial access to Upgini data enrichment. Limit for Trial: {} rows. You have already enriched: {} rows.
 transform_usage_warning=You are trying to launch enrichment for {} rows, which will exceed the rest limit {}.
 

{upgini-1.2.24a2 → upgini-1.2.25}/src/upgini/utils/display_utils.py

@@ -4,7 +4,7 @@ import textwrap
 import urllib.parse
 import uuid
 from datetime import datetime, timezone
-from io import BytesIO
+from io import StringIO
 from typing import Callable, List, Optional
 
 import pandas as pd
@@ -150,7 +150,7 @@ def make_html_report(
     search_id: str,
     email: Optional[str] = None,
     search_keys: Optional[List[str]] = None,
-):
+) -> str:
     # relevant_features_df = relevant_features_df.copy()
     # relevant_features_df["Feature name"] = relevant_features_df["Feature name"].apply(
     #     lambda x: "*" + x if x.contains("_autofe_") else x
@@ -161,9 +161,18 @@ def make_html_report(
         """<button type="button">Request a quote</button></a>"""
     )
     relevant_datasources_df.rename(columns={"action": "&nbsp;"}, inplace=True)
+
+    try:
+        from importlib.resources import files
+        font_path = files('upgini.utils').joinpath('Roboto-Regular.ttf')
+    except Exception:
+        from pkg_resources import resource_filename
+        font_path = resource_filename('upgini.utils', 'Roboto-Regular.ttf')
+
     return f"""<html>
         <head>
             <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
+            <meta charset="UTF-8">
             <style>
                 @page {{
                     size: a4 portrait;
@@ -184,12 +193,14 @@ def make_html_report(
                 }}
 
                 @font-face {{
-                    font-family: "Alice-Regular";
-                    src: url("/fonts/Alice-Regular.ttf") format("truetype");
+                    font-family: "Roboto";
+                    src: url("{font_path}") format("truetype");
                 }}
 
                 body {{
-                    font-family: "Alice-Regular", Arial, sans-serif;
+                    font-family: "Roboto", sans-serif;
+                    font-weight: 400;
+                    font-style: normal;
                 }}
 
                 #header_content {{
@@ -305,8 +316,13 @@ def show_button_download_pdf(
     from IPython.display import HTML, display
 
     file_name = f"upgini-report-{uuid.uuid4()}.pdf"
+
+    # from weasyprint import HTML
+
+    # html = HTML(string=source)
+    # html.write_pdf(file_name)
     with open(file_name, "wb") as output:
-        pisa.CreatePDF(src=BytesIO(source.encode("UTF-8")), dest=output)
+        pisa.CreatePDF(src=StringIO(source), dest=output, encoding="UTF-8")
 
     with open(file_name, "rb") as f:
         b64 = base64.b64encode(f.read())

upgini-1.2.24a2/src/upgini/__about__.py

@@ -1 +0,0 @@
-__version__ = "1.2.24a2"