From c6c0f98bb03fe49b11bcb4e23d77817038e7c2cb Mon Sep 17 00:00:00 2001
From: Eric Chen <kych@berkeley.edu>
Date: Wed, 24 Apr 2024 02:24:06 +0000
Subject: [PATCH] fix path double slash and os dir

---
 README.md                          | 6 +++---
 fog_x/database/polars_connector.py | 2 +-
 fog_x/dataset.py                   | 9 +++++++--
 requirements.txt                   | 5 ++++-
 4 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/README.md b/README.md
index 6b93567..f38be4c 100644
--- a/README.md
+++ b/README.md
@@ -1,11 +1,11 @@
-# 🦊 Fog-X
+# 🦊 Fog-RT-X
 
 [![codecov](https://codecov.io/gh/KeplerC/fog_x/branch/main/graph/badge.svg?token=fog_x_token_here)](https://codecov.io/gh/KeplerC/fog_x)
 [![CI](https://github.com/KeplerC/fog_x/actions/workflows/main.yml/badge.svg)](https://github.com/KeplerC/fog_x/actions/workflows/main.yml)
 
-🦊 Fog-X: An Efficient and Scalable Data Collection and Management Framework For Robotics Learning. Support [Open-X-Embodiment](https://robotics-transformer-x.github.io/), 🤗[HuggingFace](https://huggingface.co/). 
+🦊 Fog-RT-X: An Efficient and Scalable Data Collection and Management Framework For Robotics Learning. Support [Open-X-Embodiment](https://robotics-transformer-x.github.io/), 🤗[HuggingFace](https://huggingface.co/). 
 
-🦊 Fog-X considers both speed 🚀 and memory efficiency 📈 with active metadata and lazily-loaded trajectory data. It supports flexible and distributed dataset partitioning. 
+🦊 Fog-RT-X considers both speed 🚀 and memory efficiency 📈 with active metadata and lazily-loaded trajectory data. It supports flexible and distributed dataset partitioning. 
 
 ## Install 
 
diff --git a/fog_x/database/polars_connector.py b/fog_x/database/polars_connector.py
index 051cd0d..c9bef50 100644
--- a/fog_x/database/polars_connector.py
+++ b/fog_x/database/polars_connector.py
@@ -139,7 +139,7 @@ def load_tables(self, table_names: List[str]):
         
         # load tables from the path
         for table_name in table_names:
-            path = self.path.strip("/")
+            path = self.path.removesuffix("/")
             path = f"{path}/{table_name}.parquet"
             logger.info(f"Prepare to load table {table_name} loaded from {path}.")
 
diff --git a/fog_x/dataset.py b/fog_x/dataset.py
index eb99de0..1f9f987 100644
--- a/fog_x/dataset.py
+++ b/fog_x/dataset.py
@@ -100,7 +100,8 @@ def __init__(
             * is replace_existing actually used anywhere?
         """
         self.name = name
-        path = os.path.expanduser(path).strip("/")
+        path = os.path.expanduser(path).removesuffix("/")
+        logger.info(f"Dataset path: {path}")
         self.path = path
         if path is None:
             raise ValueError("Path is required")
@@ -112,7 +113,7 @@ def __init__(
         self.features = features
         self.enable_feature_inference = enable_feature_inference
         if episode_info_connector is None:
-                episode_info_connector = DataFrameConnector(f"{path}/")
+                episode_info_connector = DataFrameConnector(f"{path}")
         
         if step_data_connector is None:
             if not os.path.exists(f"{path}/{name}"):
@@ -437,6 +438,10 @@ def _prepare_rtx_metadata(
                     feature_name = r["feature"]
                     if "image" in feature_name and "depth" not in feature_name:
                         image = np.load(io.BytesIO(r["value"]))
+
+                        # convert from RGB to BGR
+                        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+                
                         if feature_name not in video_writers:
                             
                             output_filename = f"{self.name}_{counter}_{feature_name}"
diff --git a/requirements.txt b/requirements.txt
index c65802b..3293b2f 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -11,4 +11,7 @@ torch
 tensorflow
 tensorflow_datasets
 envlogger
-opencv-python
\ No newline at end of file
+opencv-python
+# cloud
+boto3
+smart_open
\ No newline at end of file