Merge pull request kubeagi#656 from Lanture1064/dev

chore: fix ragas-once readme typo & bug
bjwswang · Jan 26, 2024 · b047b43 · b047b43
2 parents 0fbd6cf + ec14529
commit b047b43
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 6 deletions.
diff --git a/pypi/ragas_once/README.md b/pypi/ragas_once/README.md
@@ -18,9 +18,9 @@ pip install ragas_once
 
 - `--model`: Specifies the model to use for evaluation.
     - Default value is "gpt-3.5-turbo". Langchain compatible.
-- `--api_base`: Specifies the base URL for the API.
+- `--apibase`: Specifies the base URL for the API.
     - Default value is "https://api.openai.com/v1".
-- `--api_key`: Specifies the API key to authenticate requests. 
+- `--apikey`: Specifies the API key to authenticate requests. 
     - Not required if using psuedo-openai API server, e.g. vLLM, Fastchat, etc.
 - `--embeddings`: Specifies the Huggingface embeddings model to use for evaluation. 
     - Embeddings will run **locally**.
@@ -39,7 +39,7 @@ pip install ragas_once
 ### Fiqa dataset demo:
 
 ```bash
-python3 -m ragas_once.cli --api_key "YOUR_OPENAI_API_KEY"
+python3 -m ragas_once.cli --apikey "YOUR_OPENAI_API_KEY"
 ```
 
 ### Evaluate with GPT-4 and `BAAI/bge-small-en` embeddings
@@ -52,7 +52,7 @@ pip install sentence-transformers
 Then run:
 
 ```bash
-python3 -m ragas_once.cli --model "gpt-4" --api_key "YOUR_OPENAI_API_KEY" --embeddings "BAAI/bge-small-en" --dataset "path/to/dataset.csv"
+python3 -m ragas_once.cli --model "gpt-4" --apikey "YOUR_OPENAI_API_KEY" --embeddings "BAAI/bge-small-en" --dataset "path/to/dataset.csv"
 ```
 
 ### Prepare Dataset

diff --git a/pypi/ragas_once/ragas_once/cli.py b/pypi/ragas_once/ragas_once/cli.py
@@ -66,8 +66,10 @@ def main():
 
     if dataset:
         data = pd.read_csv(dataset)
-        data["ground_truths"] = data["ground_truths"].apply(lambda x: x.split(";"))
-        data["contexts"] = data["contexts"].apply(lambda x: x.split(";"))
+        if "ground_truths" in data.columns & data["ground_truths"].dtype == 'string':
+            data["ground_truths"] = data["ground_truths"].apply(lambda x: x.split(";"))
+        if "contexts" in data.columns & data["contexts"].dtype == 'string':
+            data["contexts"] = data["contexts"].apply(lambda x: x.split(";"))
         test_set = Dataset.from_pandas(data)
     else:
         print("test_set not provided, using fiqa dataset")