Replace deprecated use of 'Repository' when pushing to hub with HfApi

CHANGELOG.md

@@ -6,6 +6,8 @@
 ### New Features
 
 ### Bug fixes
+- Replaced deprecated use of `Repository` when pushing to huggingface hub by the
+  recommended `HfApi` (see https://huggingface.co/docs/huggingface_hub/concepts/git_vs_http).
 
 ### Documentation
 
@@ -92,6 +94,7 @@
 - Huggingface push to hub now accepts a `--n-timesteps` argument to adjust the length of the video
 - Fixed `record_video` steps (before it was stepping in a closed env)
 
+
 ## Release 1.8.0 (2023-04-07)
 
 **New Documentation, Multi-Env HerReplayBuffer**

rl_zoo3/push_to_hub.py

@@ -195,14 +195,10 @@ def package_to_hub(
         private=False,
         exist_ok=True,
     )
-
-    # Git pull
     repo_local_path = Path(local_repo_path) / repo_name
-    repo = Repository(repo_local_path, clone_from=repo_url)
-    repo.git_pull(rebase=True)
-
-    repo.lfs_track(["*.mp4"])
 
+    # Retrieve current repo state
+    api.hf_hub_download(repo_id=repo_id, local_dir=local_repo_path)
     # Step 1: Save the model
     print("Saving model to:", repo_local_path / model_name)
     model.save(repo_local_path / model_name)
@@ -269,8 +265,7 @@ def package_to_hub(
     save_model_card(repo_local_path, generated_model_card, metadata)
 
     msg.info(f"Pushing repo {repo_name} to the Hugging Face Hub")
-    repo.push_to_hub(commit_message=commit_message)
-
+    api.upload_folder(repo_id=repo_id, folder_path=repo_local_path, commit_message=commit_message)
     msg.info(f"Your model is pushed to the hub. You can view your model here: {repo_url}")
     return repo_url
 
@@ -378,6 +373,12 @@ def package_to_hub(
         env_kwargs=env_kwargs,
     )
 
+    if is_atari:
+        # Patch Atari for rendering
+        # see https://github.com/mgbellemare/Arcade-Learning-Environment/pull/476
+        eval_env.unwrapped.render_mode = env_kwargs.get("render_mode")
+        eval_env.render_mode = env_kwargs.get("render_mode")
+
     kwargs = dict(seed=args.seed)
     if algo in off_policy_algos:
         # Dummy buffer size as we don't need memory to enjoy the trained agent

rl_zoo3/record_video.py

@@ -137,6 +137,12 @@
     if video_folder is None:
         video_folder = os.path.join(log_path, "videos")
 
+    if is_atari:
+        # Patch Atari for rendering
+        # see https://github.com/mgbellemare/Arcade-Learning-Environment/pull/476
+        env.unwrapped.render_mode = env_kwargs.get("render_mode")
+        env.render_mode = env_kwargs.get("render_mode")
+
     # Note: apparently it renders by default
     env = VecVideoRecorder(
         env,