Merge pull request #76 from cvs-health/release-branch/v0.3.0

Release PR: v0.3.0

.gitignore

@@ -137,3 +137,6 @@ dmypy.json
 
 #Example generated files
 examples/evaluations/text_generation/final_metrics.txt
+
+#Data generated by langfair data_loader module
+langfair/data/*

README.md

@@ -50,8 +50,8 @@ We can use `ResponseGenerator.generate_responses` to generate 25 responses for e
 from langfair.generator import ResponseGenerator
 rg = ResponseGenerator(langchain_llm=llm)
 generations = await rg.generate_responses(prompts=prompts, count=25)
-responses = [str(r) for r in generations["data"]["response"]]
-duplicated_prompts = [str(r) for r in generations["data"]["prompt"]] # so prompts correspond to responses
+responses = generations["data"]["response"]
+duplicated_prompts = generations["data"]["prompt"] # so prompts correspond to responses
 ```
 
 ##### Compute toxicity metrics
@@ -96,8 +96,8 @@ cg = CounterfactualGenerator(langchain_llm=llm)
 cf_generations = await cg.generate_responses(
     prompts=prompts, attribute='gender', count=25
 )
-male_responses = [str(r) for r in cf_generations['data']['male_response']]
-female_responses = [str(r) for r in cf_generations['data']['female_response']]
+male_responses = cf_generations['data']['male_response']
+female_responses = cf_generations['data']['female_response']
 ```
 
 Counterfactual metrics can be easily computed with `CounterfactualMetrics`.
@@ -109,7 +109,7 @@ cf_result = cm.evaluate(
     texts2=female_responses,
     attribute='gender'
 )
-cf_result
+cf_result['metrics']
 # # Output is below
 # {'Cosine Similarity': 0.8318708,
 # 'RougeL Similarity': 0.5195852482361165,
@@ -127,7 +127,7 @@ auto_object = AutoEval(
     # toxicity_device=device # uncomment if GPU is available
 )
 results = await auto_object.evaluate()
-results
+results['metrics']
 # Output is below
 # {'Toxicity': {'Toxic Fraction': 0.0004,
 #   'Expected Maximum Toxicity': 0.013845130120171235,

examples/evaluations/classification/classification_metrics_demo.ipynb

@@ -1,5 +1,13 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "ead0356d",
+   "metadata": {},
+   "source": [
+    "# Classification Metrics Demo"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 1,
@@ -10,7 +18,8 @@
    "outputs": [],
    "source": [
     "import numpy as np\n",
-    "from langfair.metrics.classification import ClassificationMetrics"
+    "\n",
+    "from langfair.metrics.classification import ClassificationMetrics\n"
    ]
   },
   {
@@ -171,9 +180,9 @@
    "uri": "us-docker.pkg.dev/deeplearning-platform-release/gcr.io/workbench-notebooks:m125"
   },
   "kernelspec": {
-   "display_name": "langfair",
+   "display_name": "langfair-ZgpfWZGz-py3.9",
    "language": "python",
-   "name": "langfair"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -185,7 +194,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.20"
+   "version": "3.9.4"
   }
  },
  "nbformat": 4,

examples/evaluations/recommendation/recommendation_metrics_demo.ipynb

@@ -1,5 +1,13 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "3c377f48",
+   "metadata": {},
+   "source": [
+    "# Recommendation Metrics Demo"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "eab56f24-c606-4b0c-a259-f125d5a7e227",