aggregate component (#26)

Dask transform component to aggregate the evaluation results. Might have
to be adapted to new dataset format.

src/components/aggregate_eval_results/Dockerfile

@@ -0,0 +1,18 @@
+FROM --platform=linux/amd64 python:3.8-slim as base
+
+# System dependencies
+RUN apt-get update && \
+    apt-get upgrade -y && \
+    apt-get install git -y
+
+# Install requirements
+COPY requirements.txt /
+RUN pip3 install --no-cache-dir -r requirements.txt
+
+# Set the working directory to the component folder
+WORKDIR /component/src
+
+# Copy over src-files
+COPY src/ .
+
+ENTRYPOINT ["fondant", "execute", "main"]

src/components/aggregate_eval_results/fondant_component.yaml

@@ -0,0 +1,20 @@
+#metadata: to be matched w/ docker image
+name: aggregate_eval_results
+description: Component that aggregates results of the evaluation of the retriever
+image: ghcr.io/ml6team/aggregate_eval_results:dev
+
+consumes:
+  text: #TODO: add/retrieve metrics to consider
+    fields:
+      context+precision:
+        type: float32
+      context+relevancy:
+        type: float32
+
+produces:
+  text:
+    fields:
+      metric:
+        type: string
+      score:
+        type: float32

src/components/aggregate_eval_results/requirements.txt

@@ -0,0 +1 @@
+fondant[component]==0.7.0

src/components/aggregate_eval_results/src/main.py

@@ -0,0 +1,19 @@
+import dask.dataframe as dd
+from fondant.component import DaskTransformComponent
+
+
+class AggregateResults(DaskTransformComponent):
+    def __init__(self, *_) -> None:
+        return None
+
+    def transform(self, dataframe: dd.DataFrame) -> dd.DataFrame:
+        metrics = list(dataframe.select_dtypes(["float", "int"]).columns)
+        agg = dataframe[metrics].mean()
+        agg_df = agg.to_frame(name="score")
+        agg_df["metric"] = agg.index
+        agg_results_df = agg_df[["metric", "score"]]
+        agg_results_df = agg_results_df.reset_index(drop=True).rename(
+            columns={"metric": "text_metric", "score": "text_score"}
+        )
+
+        return agg_results_df