Merge pull request #21 from epinzur/update_build

updated build and analysis

.github/workflows/publish-to-pypi.yml

@@ -36,6 +36,7 @@ jobs:
         run: |
           echo "pyroject.toml:"
           cat pyproject.toml
+          poetry self add poetry-plugin-export
           poetry install
           poetry build
           poetry export -f requirements.txt --output requirements.txt

README.md

@@ -2,14 +2,14 @@
 
 A tool for evaluating RAG pipelines
 
-![ragulate_logo](images/logo_smaller.png)
+![ragulate_logo](https://raw.githubusercontent.com/epinzur/ragulate/main/images/logo_smaller.png)
 
 ## The Metrics
 
 The RAGulate currently reports 4 relevancy metrics: Answer Correctness, Answer Relevance, Context Relevance, and Groundedness.
 
 
-![metrics_diagram](images/metrics.png)
+![metrics_diagram](https://raw.githubusercontent.com/epinzur/ragulate/main/images/metrics.png)
 
 * Answer Correctness
   * How well does the generated answer match the ground-truth answer?
@@ -28,7 +28,7 @@ The RAGulate currently reports 4 relevancy metrics: Answer Correctness, Answer R
 
 The tool outputs results as images like this:
 
-![example_output](images/example.png)
+![example_output](https://raw.githubusercontent.com/epinzur/ragulate/main/images/example.png)
 
 These images show distribution box plots of the metrics for different test runs.
 

colbert_chunk_size_and_k.py

@@ -27,14 +27,14 @@
 keyspace = "colbert"
 
 import logging
+
 logging.basicConfig(level=logging.INFO)
 logging.getLogger("unstructured").setLevel(logging.ERROR)
 logging.getLogger("cassandra").setLevel(logging.ERROR)
 logging.getLogger("http").setLevel(logging.ERROR)
 logging.getLogger("httpx").setLevel(logging.ERROR)
 
 
-
 def get_embedding_model(chunk_size: int) -> ColbertEmbeddingModel:
     return ColbertEmbeddingModel(doc_maxlen=chunk_size, batch_size=batch_size)
 

pyproject.toml

@@ -21,6 +21,7 @@ tqdm = ">=4.66.1"
 pyyaml = "^6.0.1"
 cerberus = "^1.3.5"
 pydantic = "^2.7.3"
+setuptools = "^70.0.0"
 
 [tool.poetry.group.dev.dependencies]
 black = "^24.4.2"

ragulate/analysis.py

@@ -53,6 +53,8 @@ def get_all_data(self, recipes: List[str]) -> DataFrame:
     def output_plots_by_dataset(self, df: DataFrame, metrics: List[str]):
         recipes = sorted(df["recipe"].unique(), key=lambda x: x.lower())
         datasets = sorted(df["dataset"].unique(), key=lambda x: x.lower())
+        metrics = sorted(metrics)
+        metrics.reverse()
 
         # generate an array of rainbow colors by fixing the saturation and lightness of the HSL
         # representation of color and marching around the hue.

scripts/test_integration_runner.py

@@ -1,5 +1,7 @@
-import pytest
 import sys
 
+import pytest
+
+
 def main():
     sys.exit(pytest.main(["tests/integration_tests"]))

scripts/test_unit_runner.py

@@ -1,5 +1,7 @@
-import pytest
 import sys
 
+import pytest
+
+
 def main():
     sys.exit(pytest.main(["tests/unit_tests"]))