Merge pull request pycaret#2052 from pycaret/master

update create_app

CHANGELOG.md

@@ -2,6 +2,27 @@
 All notable changes to this project will be documented in this file.
 <br/><br/>
 
+#### Release: PyCaret 2.3.6 | Release Date: January 11th, 2022 (NEW FEATURES, BUG FIXES)
+- Added new function `create_docker` (https://github.com/pycaret/pycaret/pull/2005)
+- Added new function `create_api` (https://github.com/pycaret/pycaret/pull/2000)
+- Added new function `check_fairness` (https://github.com/pycaret/pycaret/pull/1997)
+- Added new function `eda` (https://github.com/pycaret/pycaret/pull/1983)
+- Added new function `convert_model` (https://github.com/pycaret/pycaret/pull/1959)
+- Added an ability to pass kwargs to plots in `plot_model` (https://github.com/pycaret/pycaret/pull/19400)
+- Added `drift_report` functionality to `predict_model` (https://github.com/pycaret/pycaret/pull/1935)
+- Added new function `create_dashboard` (https://github.com/pycaret/pycaret/pull/1925)
+- Added `grid_interval` parameter to `optimize_threshold` - thanks to @wolfryu (https://github.com/pycaret/pycaret/pull/1938)
+- Made logging level configurable by environment variable (https://github.com/pycaret/pycaret/pull/2026)
+- Made the optional path in AWS configurable (https://github.com/pycaret/pycaret/pull/2045)
+- Fixed TSNE plot with PCA (https://github.com/pycaret/pycaret/pull/2032)
+- Fixed rendering of streamlit plots (https://github.com/pycaret/pycaret/pull/2008)
+- Fixed class names in `tree` plot - thanks to @yamasakih (https://github.com/pycaret/pycaret/pull/1982)
+- Fixed NearZeroVariance preprocessor not being configurable - thanks to @Flyfoxs (https://github.com/pycaret/pycaret/pull/1952)
+- Removed duplicated code - thanks to @Flyfoxs (https://github.com/pycaret/pycaret/pull/1882)
+- Documentation improvements - thanks to @harsh204016, @khrapovs (https://github.com/pycaret/pycaret/pull/1931/files, https://github.com/pycaret/pycaret/pull/1956, https://github.com/pycaret/pycaret/pull/1946, https://github.com/pycaret/pycaret/pull/1949)
+- Pinned `pyyaml<6.0.0` to fix issues with Google Colab
+<br/><br/><br/>
+
 #### Release: PyCaret 2.3.5 | Release Date: November 19th, 2021 (NEW FEATURES, BUG FIXES)
 - Fixed an issue where `Fix_multicollinearity` would fail if the target was a float (https://github.com/pycaret/pycaret/pull/1640)
 - MLFlow runs are now nested - thanks to @jfagn (https://github.com/pycaret/pycaret/pull/1660)

pycaret/anomaly.py

@@ -1046,7 +1046,7 @@ def load_model(
         dictionary of applicable authentication tokens.
 
         when platform = 'aws':
-        {'bucket' : 'S3-bucket-name'}
+        {'bucket' : 'Name of Bucket on S3', 'path': (optional) folder name under the bucket}
 
         when platform = 'gcp':
         {'project': 'gcp-project-name', 'bucket' : 'gcp-bucket-name'}

pycaret/classification.py

@@ -1883,7 +1883,7 @@ def optimize_threshold(
     optimize: str = "Accuracy",
     grid_interval: float = 0.1,
     return_data: bool = False, 
-    plot_kwargs: dict = {},
+    plot_kwargs: Optional[dict] = None,
 ):
 
     """
@@ -2265,7 +2265,7 @@ def load_model(
         dictionary of applicable authentication tokens.
 
         when platform = 'aws':
-        {'bucket' : 'S3-bucket-name'}
+        {'bucket' : 'Name of Bucket on S3', 'path': (optional) folder name under the bucket}
 
         when platform = 'gcp':
         {'project': 'gcp-project-name', 'bucket' : 'gcp-bucket-name'}

pycaret/clustering.py

@@ -1078,7 +1078,7 @@ def load_model(
         dictionary of applicable authentication tokens.
 
         when platform = 'aws':
-        {'bucket' : 'S3-bucket-name'}
+        {'bucket' : 'Name of Bucket on S3', 'path': (optional) folder name under the bucket}
 
         when platform = 'gcp':
         {'project': 'gcp-project-name', 'bucket' : 'gcp-bucket-name'}

pycaret/internal/persistence.py

@@ -348,7 +348,7 @@ def load_model(
         dictionary of applicable authentication tokens.
 
         When platform = 'aws':
-        {'bucket' : 'Name of Bucket on S3'}
+        {'bucket' : 'Name of Bucket on S3', 'path': (optional) folder name under the bucket}
 
         When platform = 'gcp':
         {'project': 'gcp_pycaret', 'bucket' : 'pycaret-test'}
@@ -415,11 +415,17 @@ def load_model(
             raise ValueError('S3 bucket name missing. Provide `bucket` name as part of authentication parameter')
 
         filename = f"{model_name}.pkl"
+
+        if "path" in authentication:
+            key = os.path.join(authentication.get("path"), filename)
+        else:
+            key = filename
+
         index = filename.rfind("/")
         s3 = boto3.resource("s3")
 
         if index == -1:
-            s3.Bucket(bucketname).download_file(filename, filename)
+            s3.Bucket(bucketname).download_file(key, filename)
         else:
             path, key = filename[: index + 1], filename[index + 1:]
             if not os.path.exists(path):

pycaret/internal/tabular.py

@@ -8305,7 +8305,7 @@ def optimize_threshold(
     optimize: str = "Accuracy",
     grid_interval: float = 0.1,
     return_data: bool = False, 
-    plot_kwargs: dict = {},
+    plot_kwargs: Optional[dict] = None,
 ):
 
     """
@@ -8396,11 +8396,12 @@ def optimize_threshold(
 
     if isinstance(optimize, str):
         # checking optimize parameter
-        optimize = _get_metric(optimize).display_name
+        optimize = _get_metric(optimize)
         if optimize is None:
             raise ValueError(
                 "Optimize method not supported. See docstring for list of available parameters."
             )
+        optimize = optimize.display_name
 
     """
     ERROR HANDLING ENDS HERE
@@ -8436,11 +8437,12 @@ def optimize_threshold(
     optimized_metric_index = np.array(results_concat_melted[results_concat_melted['variable'] == optimize]['value']).argmax()
     best_model_by_metric = models_by_threshold[optimized_metric_index]
 
-    logger.info("plotting optimizartion threshold using plotly")
+    logger.info("plotting optimization threshold using plotly")
 
     # plotting threshold
     import plotly.express as px
-    title = str(model_name) + " Probability Threshold Optimization (default = 0.5) "
+    title = f"{model_name} Probability Threshold Optimization (default = 0.5)"
+    plot_kwargs = plot_kwargs or {}
     fig = px.line(results_concat_melted, x="probability_threshold", y="value", title = title,\
                     color='variable', **plot_kwargs)
     logger.info("Figure ready for render")

pycaret/regression.py

@@ -1998,7 +1998,7 @@ def load_model(
         dictionary of applicable authentication tokens.
 
         when platform = 'aws':
-        {'bucket' : 'S3-bucket-name'}
+        {'bucket' : 'Name of Bucket on S3', 'path': (optional) folder name under the bucket}
 
         when platform = 'gcp':
         {'project': 'gcp-project-name', 'bucket' : 'gcp-bucket-name'}

pycaret/tests/test_persistence.py

@@ -2,7 +2,7 @@
 import pytest
 import boto3
 from moto import mock_s3
-from pycaret.internal.persistence import deploy_model
+from pycaret.internal.persistence import deploy_model, load_model
 
 
 @pytest.fixture(scope='function')
@@ -29,5 +29,8 @@ def test_deploy_model(s3):
     model_name = 'test'
 
     deploy_model(model, model_name=model_name, platform='aws', authentication=authentication)
+
     s3.head_object(Bucket=authentication.get("bucket"),
                    Key=os.path.join(authentication.get("path"), f"{model_name}.pkl"))
+
+    _ = load_model(model_name=model_name, platform='aws', authentication=authentication, verbose=True)

pycaret/utils.py

@@ -5,8 +5,8 @@
 import pandas as pd
 import functools
 
-version_ = "2.3.5"
-nightly_version_ = "2.3.5"
+version_ = "2.3.6"
+nightly_version_ = "2.3.6"
 
 __version__ = version_
 

requirements.txt

@@ -1,6 +1,5 @@
 pandas
 scipy<=1.5.4
-numpy==1.19.5
 seaborn
 matplotlib
 IPython
@@ -27,4 +26,4 @@ imbalanced-learn==0.7.0
 scikit-plot #for lift and gain charts
 Boruta
 pyyaml<6.0.0
-#numba<0.54
+#numba<0.54

setup.py

@@ -21,7 +21,7 @@ def readme():
 
 setup(
     name="pycaret",
-    version="2.3.5",
+    version="2.3.6",
     description="PyCaret - An open source, low-code machine learning library in Python.",
     long_description=readme(),
     long_description_content_type="text/markdown",

setup_nightly.py

@@ -4,7 +4,7 @@
 from setuptools import setup, find_packages
 import time
 
-nightly_version = "2.3.5"
+nightly_version = "2.3.6"
 
 nightly_readme = f"This is a nightly version of the [PyCaret](https://pypi.org/project/pycaret/) library, intended as a preview of the upcoming {nightly_version} version. It may contain unstable and untested code.\n"