Upload folder using huggingface_hub

README.md

@@ -32,11 +32,6 @@ from datasets import load_dataset
 ds = load_dataset("Ekgren/swedish_skolprov", "all")
 ```
 
-You can find an evaluation script that uses OpenRouter here: [src/evaluate_data.py](src/evaluate_data.py)  
-
-```bash
-python evaluate_data.py --model_name "mistralai/mistral-small-3.1-24b-instruct" --eval_subset "all" --output_path "./results"
-```
 
 ## Högskoleprovet
 - https://www.studera.nu/hogskoleprov/forbered/tidigare-hogskoleprov/

src/combine.py

@@ -0,0 +1,24 @@
+import pandas as pd
+import json
+import glob
+import os
+
+json_folder = "results"
+output_csv = "results/combined.csv"
+
+data_rows = []
+
+for file in glob.glob(os.path.join(json_folder, "*.json")):
+    with open(file, 'r') as f:
+        content = json.load(f)
+        flat_row = {'sampled_model_name': content['sampled_model_name']}
+        
+        for key, value in content.items():
+            if key != 'sampled_model_name':
+                flat_row[f"{key}_points"] = value['points']
+                flat_row[f"{key}_total"] = value['total']
+        
+        data_rows.append(flat_row)
+
+df = pd.DataFrame(data_rows)
+df.to_csv(output_csv, index=False)

src/run_tests.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+
+# List of model names
+models=(
+    "openai/gpt-4o-mini"
+    "openai/o3-mini-high"
+    "google/gemini-2.0-flash-001"
+)
+
+# Loop through each model and run evaluation
+for model in "${models[@]}"; do
+    echo "Evaluating model: ${model}"
+    python evaluate_data.py --model_name "${model}" --eval_subset "all" --output_path "./results"
+done
+