updated readmes with refs to colab notebook

README.md

@@ -37,7 +37,7 @@ Table of contents:
 > Compatibility is only maintained with **Linux OS**.
 > 
 > If you don't have access to a Linux machine we advise using the free Google 
-> Colab service ([example Colab notebook here](https://colab.research.google.com/drive/1OElgST6yLpcgsC2i4BE_dJxVny_bLN6F?usp=sharing)).
+> Colab service ([example Colab notebook here](https://colab.research.google.com/github/AndreFCruz/fairgbm-fork/blob/add-colab-example/examples/FairGBM-python-notebooks/FairGBM_example_for_equalized_odds_%5Bgoogle_colab%5D.ipynb)).
 
 FairGBM can be installed from [PyPI](https://pypi.org/project/fairgbm/):
 
@@ -59,7 +59,7 @@ pip install fairgbm/python-package/
 
 ## Getting started
 
-> **Recommended** Python notebook example [here](examples/FairGBM-python-notebooks/UCI-Adult-example-with-hyperparameter-tuning.ipynb) (or see the example Google Colab notebook [here](https://colab.research.google.com/drive/1OElgST6yLpcgsC2i4BE_dJxVny_bLN6F?usp=sharing)).
+> **Recommended** Python notebook example [here](examples/FairGBM-python-notebooks/FairGBM_example_for_equalized_odds_[google_colab].ipynb) (Google Colab link [here](https://colab.research.google.com/github/AndreFCruz/fairgbm-fork/blob/add-colab-example/examples/FairGBM-python-notebooks/FairGBM_example_for_equalized_odds_%5Bgoogle_colab%5D.ipynb)).
 
 You can get FairGBM up and running in just a few lines of Python code:
 
@@ -84,7 +84,7 @@ Y_test_pred = fairgbm_clf.predict_proba(X_test)[:, -1]  # Compute continuous cla
 
 **For Python examples see the [_notebooks folder_](/examples/FairGBM-python-notebooks).**
 
-A more in-depth explanation and other usage examples can be found in the [**_examples folder_**](/examples).
+A more in-depth explanation and other usage examples (with python package or compiled binary) can be found in the [**_examples folder_**](/examples).
 
 > **Note** 
 > FairGBM is a research project, so its default hyperparameters (key-word arguments) 
@@ -161,7 +161,7 @@ constraints** (_e.g._, equal opportunity).
 
 Namely, you can target equality of performance metrics (FPR, FNR, or both) across instances from _two or more_ different
 protected groups (see [fairness constraints](#fairness-constraints) section).
-Simultaneously (and optionally), you can add global constraints on specific metrics (see [global constraints](#global-constraints) section).
+Optionally, you can simultaneously add global constraints on specific metrics (see [global constraints](#global-constraints) section).
 
 ### Fairness constraints
 

examples/FairGBM-python-notebooks/README.md

@@ -4,13 +4,13 @@ In this folder you can find several FairGBM usage examples.
 
 | _File name_             | _Dataset (size)_ | _Description_ | _Fairness criterion_ |
 |-------------------------|------------------|---------------|----------------------|
-| [UCI-Adult-example.ipynb](UCI-Adult-example.ipynb) | [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | Simple example on the popular UCI Adult dataset. | Equal TPR (_equal opportunity_) |
-| [UCI-Adult-example-with-hyperparameter-tuning.ipynb](UCI-Adult-example-with-hyperparameter-tuning.ipynb) | [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | **Recommended**: run hyperparameter-tuning to obtain optimal `multiplier_learning_rate` parameter value. | Equal TPR and FPR (_equalized odds_) |
-| [FairGBM hyperparameter search on google colab.ipynb](https://colab.research.google.com/drive/1OElgST6yLpcgsC2i4BE_dJxVny_bLN6F?usp=sharing) | [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | Easy **Google Colab** example to get up and running with FairGBM. | Equal TPR and FPR (_equalized odds_) |
+| [FairGBM_example_for_equalized_odds_[google_colab].ipynb](FairGBM_example_for_equalized_odds_[google_colab].ipynb) ([colab link](https://colab.research.google.com/github/AndreFCruz/fairgbm-fork/blob/add-colab-example/examples/FairGBM-python-notebooks/FairGBM_example_for_equalized_odds_%5Bgoogle_colab%5D.ipynb))| [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | Simple [*Colab* example](https://colab.research.google.com/github/AndreFCruz/fairgbm-fork/blob/add-colab-example/examples/FairGBM-python-notebooks/FairGBM_example_for_equalized_odds_%5Bgoogle_colab%5D.ipynb) to get up and running with FairGBM. | Equal TPR and FPR (_equalized odds_) |
+| [UCI-Adult-example.ipynb](UCI-Adult-example.ipynb) | [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | Simple local example on the popular UCI Adult dataset. | Equal TPR (_equal opportunity_) |
+| [UCI-Adult-example-with-hyperparameter-tuning.ipynb](UCI-Adult-example-with-hyperparameter-tuning.ipynb) | [UCI Adult](https://archive.ics.uci.edu/ml/datasets/adult) (49K) | Example with hyperparameter-tuning to obtain optimal `multiplier_learning_rate` parameter value. | Equal TPR and FPR (_equalized odds_) |
 
 <!--
 | [credit-card-fraud-example.ipynb](credit-card-fraud-example.ipynb) | [Credit Card Fraud](https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud) (285K) | Highly class imbalanced dataset in which a trivial classifier achieves 99.8% accuracy. Show-cases the use of both fairness and global constraints! | Equal FPR (predictive equality) |
 | [ACSIncome-example.ipynb](ACSIncome-example.ipynb) | [ACSIncome](https://github.com/zykls/folktables) (1.7M) | Large dataset for comparing run-times of popular fairness-aware algorithms. _Requires a sizeable amount of RAM memory (+32GB)._ | Equal FNR (equal opportunity) |
 -->
 
-Examples require extra requirements listed in the `requirements.txt` file. Please run `pip install -r requirements.txt` to install them.
+Examples may require extra requirements listed in the `requirements.txt` file. Please run `pip install -r requirements.txt` to install them.