Switch to remove dataset storage

app.py

@@ -2,7 +2,7 @@ import json
 import requests
 from datasets import load_dataset
 import gradio as gr
-from huggingface_hub import HfApi, hf_hub_download
+from huggingface_hub import HfApi, hf_hub_download, Repository
 from huggingface_hub.repocard import metadata_load
 import pandas as pd
 from matchmaking import *
@@ -10,19 +10,24 @@ from background_task import init_matchmaking
 from apscheduler.schedulers.background import BackgroundScheduler
 
 
+DATASET_REPO_URL = "https://huggingface.co/datasets/CarlCochet/BotFightData"
+ELO_FILENAME = "elo.csv"
+ELO_DIR = "soccer_elo"
+ELO_FILE = os.path.join(ELO_DIR, ELO_FILENAME)
+HF_TOKEN = os.environ.get("HF_TOKEN")
+
 block = gr.Blocks()
-env = [
-    {
-        "name": "Soccer",
-        "global": None,
-    },
-]
 matchmaking = Matchmaking()
+api = HfApi()
 
 scheduler = BackgroundScheduler()
 scheduler.add_job(func=init_matchmaking, trigger="interval", seconds=15000)
 scheduler.start()
 
+repo = Repository(
+    local_dir=ELO_DIR, clone_from=DATASET_REPO_URL, use_auth_token=HF_TOKEN
+)
+
 
 def update_elos():
     matchmaking.read_history()
@@ -31,7 +36,9 @@ def update_elos():
 
 
 def get_elo_data() -> pd.DataFrame:
-    data = pd.read_csv(f"env_elos/elo.csv")
+    hf_hub_download(repo_id="CarlCochet/BotFightData", filename=ELO_FILENAME, subfolder=ELO_DIR)
+    with open(ELO_FILE, "r") as f:
+        data = pd.read_csv(f)
     return data
 
 

background_task.py

@@ -1,10 +1,15 @@
+import os
 import random
 import pandas as pd
 from datetime import datetime
 from huggingface_hub import HfApi
 
 
-env = "Soccer"
+DATASET_REPO_URL = "https://huggingface.co/datasets/CarlCochet/BotFightData"
+ELO_FILENAME = "elo.csv"
+ELO_DIR = "soccer_elo"
+ELO_FILE = os.path.join(ELO_DIR, ELO_FILENAME)
+HF_TOKEN = os.environ.get("HF_TOKEN")
 
 
 class Model:
@@ -62,7 +67,6 @@ class Matchmaking:
             self.matches["model2"].append(model2.name)
             self.matches["result"].append(result)
             self.matches["datetime"].append(datetime.now().strftime("%Y-%m-%d %H:%M:%S.%f"))
-            self.matches["env"].append(env)
 
     def compute_elo(self, model1, model2, result):
         """ Compute the new elo for each model based on a match result. """
@@ -106,7 +110,14 @@ class Matchmaking:
             data_dict["games_played"].append(model.games_played)
 
         df = pd.DataFrame(data_dict)
-        df.to_csv('env_elos/elo.csv', index=False)
+        fileobj = open('env_elos/elo.csv', 'w')
+        df.to_csv(fileobj, index=False)
+        api.upload_file(
+            fileobj,
+            "env_elos/elo.csv",
+            "CarlCochet/BotFights",
+            "Update elos",
+        )
         df_matches = pd.DataFrame(self.matches)
         date = datetime.now()
         df_matches.to_csv(f"match_history/{date.strftime('%Y-%m-%d_%H-%M-%S_%f')}.csv", index=False)

env_elos/elo.csv

@@ -1,5 +0,0 @@
-rank,author,model,elo,games_played
-1,CarlCochet,model3,1431,4
-2,CarlCochet,model1,1234,8
-3,CarlCochet,model4,1200,0
-4,CarlCochet,model2,1000,5