modulift 0.0.2__tar.gz → 0.0.3__tar.gz

{modulift-0.0.2 → modulift-0.0.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: modulift
-Version: 0.0.2
+Version: 0.0.3
 Summary: A powerful search tool for discovering Python packages based on user queries
 Home-page: https://github.com/okanyenigun/modulift
 Author: Okan Yenigün
@@ -68,7 +68,7 @@ print(result)
     'package': 'numpy',
     'description': 'NumPy is the fundamental package for scientific computing with Python. It provides powerful N-dimensional arrays, sophisticated mathematical functions, tools for integrating C/C++ and Fortran code, and useful linear algebra, Fourier transform, and random number capabilities.',
     'keywords': 'scientific computing, array, linear algebra, Fourier transform, random number generation, data processing, numerical analysis',
-    'popularity': 'very popular'
+    'popularity': 5
 }
 ```
 
@@ -86,7 +86,7 @@ print(result)
     'package': 'nullsweep',
     'description': 'NullSweep is a Python library designed for detecting and handling patterns of missing data in pandas DataFrames. It provides a simple API to identify global missing data patterns across the entire dataset, patterns related to specific features within the dataset, and to impute missing values using various strategies.',
     'keywords': 'missing data, data imputation, pandas, pattern detection, feature-specific patterns',
-    'popularity': 'popular'
+    'popularity': 4
 }
 ```
 
@@ -101,7 +101,7 @@ result = mf.search_by_package_name("numpy", markdown=True)
 
 - **Description**: NumPy is the fundamental package for scientific computing with Python. It provides powerful N-dimensional arrays, sophisticated mathematical functions, tools for integrating C/C++ and Fortran code, and useful linear algebra, Fourier transform, and random number capabilities.
 - **Keywords**: scientific computing, array, linear algebra, Fourier transform, random number generation, data processing, numerical analysis
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ### 2. Search by Keywords
 
@@ -145,7 +145,7 @@ Markdown Output
 
 - **Description**: The SageMaker Python SDK is an open-source library designed for training and deploying machine learning models on Amazon SageMaker. It supports popular deep learning frameworks like Apache MXNet and TensorFlow, as well as Amazon's own scalable algorithms optimized for SageMaker. Users can also deploy custom algorithms in Docker containers, making it versatile for various machine learning tasks.
 - **Keywords**: machine learning, deep learning, model deployment, Amazon SageMaker, data science, AI, Docker
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ---
 
@@ -153,7 +153,7 @@ Markdown Output
 
 - **Description**: fastbook is a comprehensive resource for learning deep learning through practical coding examples using the fastai library and PyTorch. It is designed for coders of all backgrounds, providing accessible education on deep learning techniques and applications across various domains, including computer vision and natural language processing.
 - **Keywords**: deep learning, fastai, PyTorch, computer vision, natural language processing, machine learning, data science, Jupyter Notebooks
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ---
 
@@ -161,7 +161,7 @@ Markdown Output
 
 - **Description**: ProAiLab is a Python library designed for building and deploying AI models. It provides tools for data preprocessing, model training, and evaluation, making it suitable for both beginners and experienced machine learning practitioners.
 - **Keywords**: machine learning, AI, deep learning, data science, model development
-- **Popularity**: popular
+- **Popularity**: 4
 
 ### 3. Full-Text Search in Descriptions
 
@@ -177,7 +177,15 @@ Markdown Output
 
 - **Description**: The SageMaker Python SDK is an open-source library designed for training and deploying machine learning models on Amazon SageMaker. It supports popular deep learning frameworks like Apache MXNet and TensorFlow, as well as Amazon's own scalable algorithms optimized for SageMaker. Users can also deploy custom algorithms in Docker containers, making it versatile for various machine learning tasks.
 - **Keywords**: machine learning, deep learning, model deployment, Amazon SageMaker, data science, AI, Docker
-- **Popularity**: very popular
+- **Popularity**: 5
+
+### 4. Find Similar Packages
+
+Use cosine similarity to identify packages similar to a given package name.
+
+```python
+similar_packages = mf.find_similar_package("nullsweep")
+```
 
 ## License
 

modulift-0.0.3/modulift/__init__.py

@@ -0,0 +1,3 @@
+from .api import *
+
+__version__ = '0.0.3'

{modulift-0.0.2 → modulift-0.0.3}/modulift/api.py

@@ -1,5 +1,7 @@
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
 from IPython.display import display, Markdown
-from typing import List, Optional
+from typing import List, Optional, Dict, Any
 from .data import load_data
 
 
@@ -226,4 +228,86 @@ def search_by_description(text: str, limit: Optional[int] = None, markdown: bool
             display(Markdown(markdown_text))
     else:
         print(f"No matches found for the text '{text}'.")
-    return results
+    return results
+
+
+def find_similar_package(package_name: str, top_n: int=5, markdown: bool=False) -> List[Dict[str, Any]]:
+    """
+    Finds and returns a list of packages similar to the given package based on textual similarity.
+
+    This function uses TF-IDF (Term Frequency-Inverse Document Frequency) vectorization and cosine similarity
+    to compare package descriptions and keywords. It identifies the most relevant packages based on
+    text similarity, ranking them from highest to lowest similarity.
+
+    If `markdown=True`, the results will also be displayed in a structured Markdown format (suitable for Jupyter Notebook usage).
+
+    Args:
+        package_name (str): The name of the package for which similar packages should be retrieved.
+        top_n (int, optional): The number of similar packages to return. Defaults to 5.
+        markdown (bool, optional): Whether to display the results in a Markdown table (useful in Jupyter). Defaults to False.
+
+    Returns:
+        List[Dict[str, Any]]: A list of dictionaries, where each dictionary contains:
+            - "package" (str): The name of the similar package.
+            - "description" (str): The description of the similar package.
+            - "similarity_score" (float): A numerical score representing similarity (range: 0 to 1).
+
+    Raises:
+        ValueError: If the provided `package_name` is not found in the dataset.
+        TypeError: If `package_name` is not a string, `top_n` is not a positive integer, or `markdown` is not a boolean.
+
+    Example:
+        >>> find_similar_package("numpy", top_n=3, markdown=False)
+        [
+            {"package": "scipy", "description": "Scientific computing library", "similarity_score": 0.89},
+            {"package": "pandas", "description": "Data analysis library", "similarity_score": 0.75},
+            {"package": "matplotlib", "description": "Plotting library", "similarity_score": 0.72}
+        ]
+
+        # If markdown=True in a Jupyter Notebook, it will also display a formatted table.
+    """
+    if not isinstance(package_name, str):
+        raise TypeError(f"Expected 'package_name' to be a string, got {type(package_name).__name__}.")
+    if not isinstance(top_n, int) or top_n <= 0:
+        raise TypeError(f"'top_n' should be a positive integer, got {top_n}.")
+    if not isinstance(markdown, bool):
+        raise TypeError(f"'markdown' should be a boolean, got {type(markdown).__name__}.")
+    
+    df = load_data()
+
+    package_row = df.query("package == @package_name")
+    
+    if package_row.empty:
+        raise ValueError(f"No package found with the name '{package_name}'. Ensure the name is correct.")
+
+    package_idx = package_row.index[0]
+    
+    df["combined_text"] = df["description"].fillna("") + " " + df["keywords"].fillna("")
+
+    tfidf_vectorizer = TfidfVectorizer(stop_words="english")
+    tfidf_matrix = tfidf_vectorizer.fit_transform(df["combined_text"])
+
+    cosine_similarities = cosine_similarity(tfidf_matrix[package_idx], tfidf_matrix).flatten()
+
+    similar_indices = cosine_similarities.argsort()[::-1][1 : top_n + 1]
+
+    similar_packages = [
+        {
+            "package": df.iloc[i]["package"],
+            "description": df.iloc[i]["description"],
+            "similarity_score": round(cosine_similarities[i], 4),
+        }
+        for i in similar_indices
+    ]
+
+    if markdown:
+        markdown_text = f"### Similar Packages to '{package_name}'\n"
+        for package in similar_packages:
+            markdown_text += f"""
+**{package['package']}**
+- **Description**: {package['description']}
+- **Similarity Score**: {package['similarity_score']}
+"""
+        display(Markdown(markdown_text))
+
+    return similar_packages

{modulift-0.0.2 → modulift-0.0.3}/modulift.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: modulift
-Version: 0.0.2
+Version: 0.0.3
 Summary: A powerful search tool for discovering Python packages based on user queries
 Home-page: https://github.com/okanyenigun/modulift
 Author: Okan Yenigün
@@ -68,7 +68,7 @@ print(result)
     'package': 'numpy',
     'description': 'NumPy is the fundamental package for scientific computing with Python. It provides powerful N-dimensional arrays, sophisticated mathematical functions, tools for integrating C/C++ and Fortran code, and useful linear algebra, Fourier transform, and random number capabilities.',
     'keywords': 'scientific computing, array, linear algebra, Fourier transform, random number generation, data processing, numerical analysis',
-    'popularity': 'very popular'
+    'popularity': 5
 }
 ```
 
@@ -86,7 +86,7 @@ print(result)
     'package': 'nullsweep',
     'description': 'NullSweep is a Python library designed for detecting and handling patterns of missing data in pandas DataFrames. It provides a simple API to identify global missing data patterns across the entire dataset, patterns related to specific features within the dataset, and to impute missing values using various strategies.',
     'keywords': 'missing data, data imputation, pandas, pattern detection, feature-specific patterns',
-    'popularity': 'popular'
+    'popularity': 4
 }
 ```
 
@@ -101,7 +101,7 @@ result = mf.search_by_package_name("numpy", markdown=True)
 
 - **Description**: NumPy is the fundamental package for scientific computing with Python. It provides powerful N-dimensional arrays, sophisticated mathematical functions, tools for integrating C/C++ and Fortran code, and useful linear algebra, Fourier transform, and random number capabilities.
 - **Keywords**: scientific computing, array, linear algebra, Fourier transform, random number generation, data processing, numerical analysis
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ### 2. Search by Keywords
 
@@ -145,7 +145,7 @@ Markdown Output
 
 - **Description**: The SageMaker Python SDK is an open-source library designed for training and deploying machine learning models on Amazon SageMaker. It supports popular deep learning frameworks like Apache MXNet and TensorFlow, as well as Amazon's own scalable algorithms optimized for SageMaker. Users can also deploy custom algorithms in Docker containers, making it versatile for various machine learning tasks.
 - **Keywords**: machine learning, deep learning, model deployment, Amazon SageMaker, data science, AI, Docker
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ---
 
@@ -153,7 +153,7 @@ Markdown Output
 
 - **Description**: fastbook is a comprehensive resource for learning deep learning through practical coding examples using the fastai library and PyTorch. It is designed for coders of all backgrounds, providing accessible education on deep learning techniques and applications across various domains, including computer vision and natural language processing.
 - **Keywords**: deep learning, fastai, PyTorch, computer vision, natural language processing, machine learning, data science, Jupyter Notebooks
-- **Popularity**: very popular
+- **Popularity**: 5
 
 ---
 
@@ -161,7 +161,7 @@ Markdown Output
 
 - **Description**: ProAiLab is a Python library designed for building and deploying AI models. It provides tools for data preprocessing, model training, and evaluation, making it suitable for both beginners and experienced machine learning practitioners.
 - **Keywords**: machine learning, AI, deep learning, data science, model development
-- **Popularity**: popular
+- **Popularity**: 4
 
 ### 3. Full-Text Search in Descriptions
 
@@ -177,7 +177,15 @@ Markdown Output
 
 - **Description**: The SageMaker Python SDK is an open-source library designed for training and deploying machine learning models on Amazon SageMaker. It supports popular deep learning frameworks like Apache MXNet and TensorFlow, as well as Amazon's own scalable algorithms optimized for SageMaker. Users can also deploy custom algorithms in Docker containers, making it versatile for various machine learning tasks.
 - **Keywords**: machine learning, deep learning, model deployment, Amazon SageMaker, data science, AI, Docker
-- **Popularity**: very popular
+- **Popularity**: 5
+
+### 4. Find Similar Packages
+
+Use cosine similarity to identify packages similar to a given package name.
+
+```python
+similar_packages = mf.find_similar_package("nullsweep")
+```
 
 ## License
 

{modulift-0.0.2 → modulift-0.0.3}/modulift.egg-info/requires.txt

@@ -1,4 +1,6 @@
 datasets==3.1.0
+scikit-learn==1.6.1
+ipywidgets==8.1.5
 
 [dev]
 twine==5.1.1

{modulift-0.0.2 → modulift-0.0.3}/setup.py

@@ -7,7 +7,7 @@ with open("Readme.md", "r") as f:
 
 setup(
     name="modulift",
-    version="0.0.2",
+    version="0.0.3",
     description="A powerful search tool for discovering Python packages based on user queries",
     packages=find_packages(),
     long_description=long_description,
@@ -35,7 +35,9 @@ setup(
         'Operating System :: OS Independent'
     ],
     install_requires=[
-        'datasets==3.1.0'
+        'datasets==3.1.0',
+        'scikit-learn==1.6.1',
+        'ipywidgets==8.1.5',
         ],
     extras_require={
         'dev': ['twine==5.1.1'],

modulift-0.0.2/modulift/__init__.py

@@ -1,3 +0,0 @@
-from .api import *
-
-__version__ = '0.0.2'