Try fixing the Tensorflow example

comet-ml · Jul 9, 2024 · cf71668 · cf71668
1 parent 8e448a0
commit cf71668
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 3 deletions.
diff --git a/integrations/workflow-orchestration/metaflow/metaflow-model-evaluation/README.md b/integrations/workflow-orchestration/metaflow/metaflow-model-evaluation/README.md
@@ -32,7 +32,7 @@ export COMET_WORKSPACE=<Your Comet Workspace>
 In this guide, we will demonstrate how to use Comet's Metaflow integration to build a simple model evaluation flow.
 
 ```shell
-python metaflow_model_evaluation.py run --max-workers 1 --n_samples 100
+python metaflow-model-evaluation.py run --max-workers 1 --n_samples 100
 ```
 
 Our flow consists of two steps.

diff --git a/...ns/workflow-orchestration/metaflow/metaflow-model-evaluation/metaflow-model-evaluation.py b/...ns/workflow-orchestration/metaflow/metaflow-model-evaluation/metaflow-model-evaluation.py
@@ -215,8 +215,7 @@ def evaluate_classification_metrics(self):
         )
         accuracy = accuracy_score(labels, torch.argmax(predictions, dim=1))
 
-        self.comet_experiment.log_metrics(clf_metrics["micro avg"], prefix="micro_avg")
-        self.comet_experiment.log_metrics(clf_metrics["macro avg"], prefix="macro_avg")
+        self.comet_experiment.log_metrics(clf_metrics)
         self.comet_experiment.log_metrics({"accuracy": accuracy})
 
         log_model(self.comet_experiment, model, self.input)