update-leaderboard (#2)

- add toy results (caa5e2c7a15b22fffec51a9b2125523e08aba0b3)
- update column names (492f4350b7cca4b995eb44cf6af09ea22e5b8cd3)
- update about and env (d4cb50ad7a467c731f3363b755c3ab6264a03b75)
- add basic static leaderboard (bca96936ac7b6b6d1a21189155c05ca2340329da)

app.py

@@ -22,7 +22,7 @@ from src.display.utils import (
     ModelType,
     fields,
     WeightType,
-    Precision
+    Precision,
 )
 from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN
 from src.populate import get_evaluation_queue_df, get_leaderboard_df
@@ -32,24 +32,40 @@ from src.submission.submit import add_new_eval
 def restart_space():
     API.restart_space(repo_id=REPO_ID)
 
+
 ### Space initialisation
 try:
     print(EVAL_REQUESTS_PATH)
     snapshot_download(
-        repo_id=QUEUE_REPO, local_dir=EVAL_REQUESTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN
+        repo_id=QUEUE_REPO,
+        local_dir=EVAL_REQUESTS_PATH,
+        repo_type="dataset",
+        tqdm_class=None,
+        etag_timeout=30,
+        token=TOKEN,
     )
 except Exception:
     restart_space()
 try:
     print(EVAL_RESULTS_PATH)
     snapshot_download(
-        repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN
+        repo_id=RESULTS_REPO,
+        local_dir=EVAL_RESULTS_PATH,
+        repo_type="dataset",
+        tqdm_class=None,
+        etag_timeout=30,
+        token=TOKEN,
     )
 except Exception:
     restart_space()
 
 
-LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)
+LEADERBOARD_DF = get_leaderboard_df(
+    EVAL_RESULTS_PATH + "/" + "BOOM_leaderboard.csv", EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS
+)
+LEADERBOARD_DF_DOMAIN = get_leaderboard_df(
+    EVAL_RESULTS_PATH + "/" + "BOOM_leaderboard.csv", EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS
+)
 
 (
     finished_eval_queue_df,
@@ -57,7 +73,9 @@ LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS,
     pending_eval_queue_df,
 ) = get_evaluation_queue_df(EVAL_REQUESTS_PATH, EVAL_COLS)
 
+
 def init_leaderboard(dataframe):
+    # TODO: merge results df with model info df
     if dataframe is None or dataframe.empty:
         raise ValueError("Leaderboard DataFrame is empty or None.")
     return Leaderboard(
@@ -68,21 +86,10 @@ def init_leaderboard(dataframe):
             cant_deselect=[c.name for c in fields(AutoEvalColumn) if c.never_hidden],
             label="Select Columns to Display:",
         ),
-        search_columns=[AutoEvalColumn.model.name, AutoEvalColumn.license.name],
+        search_columns=[AutoEvalColumn.model.name],
         hide_columns=[c.name for c in fields(AutoEvalColumn) if c.hidden],
         filter_columns=[
             ColumnFilter(AutoEvalColumn.model_type.name, type="checkboxgroup", label="Model types"),
-            ColumnFilter(AutoEvalColumn.precision.name, type="checkboxgroup", label="Precision"),
-            ColumnFilter(
-                AutoEvalColumn.params.name,
-                type="slider",
-                min=0.01,
-                max=150,
-                label="Select the number of parameters (B)",
-            ),
-            ColumnFilter(
-                AutoEvalColumn.still_on_hub.name, type="boolean", label="Deleted/incomplete", default=True
-            ),
         ],
         bool_checkboxgroup_label="Hide models",
         interactive=False,
@@ -95,98 +102,15 @@ with demo:
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
-        with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
+        with gr.TabItem("🏅 Overall", elem_id="boom-benchmark-tab-table", id=0):
             leaderboard = init_leaderboard(LEADERBOARD_DF)
 
-        with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
-            gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
+        # TODO - add other tabs if needed
+        with gr.TabItem("🏅 By Domain - TODO", elem_id="boom-benchmark-tab-table", id=1):
+            leaderboard = init_leaderboard(LEADERBOARD_DF_DOMAIN)  # TODO - update table data
 
-        with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3):
-            with gr.Column():
-                with gr.Row():
-                    gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")
-
-                with gr.Column():
-                    with gr.Accordion(
-                        f"✅ Finished Evaluations ({len(finished_eval_queue_df)})",
-                        open=False,
-                    ):
-                        with gr.Row():
-                            finished_eval_table = gr.components.Dataframe(
-                                value=finished_eval_queue_df,
-                                headers=EVAL_COLS,
-                                datatype=EVAL_TYPES,
-                                row_count=5,
-                            )
-                    with gr.Accordion(
-                        f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})",
-                        open=False,
-                    ):
-                        with gr.Row():
-                            running_eval_table = gr.components.Dataframe(
-                                value=running_eval_queue_df,
-                                headers=EVAL_COLS,
-                                datatype=EVAL_TYPES,
-                                row_count=5,
-                            )
-
-                    with gr.Accordion(
-                        f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})",
-                        open=False,
-                    ):
-                        with gr.Row():
-                            pending_eval_table = gr.components.Dataframe(
-                                value=pending_eval_queue_df,
-                                headers=EVAL_COLS,
-                                datatype=EVAL_TYPES,
-                                row_count=5,
-                            )
-            with gr.Row():
-                gr.Markdown("# ✉️✨ Submit your model here!", elem_classes="markdown-text")
-
-            with gr.Row():
-                with gr.Column():
-                    model_name_textbox = gr.Textbox(label="Model name")
-                    revision_name_textbox = gr.Textbox(label="Revision commit", placeholder="main")
-                    model_type = gr.Dropdown(
-                        choices=[t.to_str(" : ") for t in ModelType if t != ModelType.Unknown],
-                        label="Model type",
-                        multiselect=False,
-                        value=None,
-                        interactive=True,
-                    )
-
-                with gr.Column():
-                    precision = gr.Dropdown(
-                        choices=[i.value.name for i in Precision if i != Precision.Unknown],
-                        label="Precision",
-                        multiselect=False,
-                        value="float16",
-                        interactive=True,
-                    )
-                    weight_type = gr.Dropdown(
-                        choices=[i.value.name for i in WeightType],
-                        label="Weights type",
-                        multiselect=False,
-                        value="Original",
-                        interactive=True,
-                    )
-                    base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)")
-
-            submit_button = gr.Button("Submit Eval")
-            submission_result = gr.Markdown()
-            submit_button.click(
-                add_new_eval,
-                [
-                    model_name_textbox,
-                    base_model_name_textbox,
-                    revision_name_textbox,
-                    precision,
-                    weight_type,
-                    model_type,
-                ],
-                submission_result,
-            )
+        with gr.TabItem("📝 About - TODO", elem_id="boom-benchmark-tab-table", id=2):
+            gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
 
     with gr.Row():
         with gr.Accordion("📙 Citation", open=False):
@@ -201,4 +125,4 @@ with demo:
 scheduler = BackgroundScheduler()
 scheduler.add_job(restart_space, "interval", seconds=1800)
 scheduler.start()
-demo.queue(default_concurrency_limit=40).launch()
+demo.queue(default_concurrency_limit=40).launch()

results/BOOM_leaderboard.csv

@@ -0,0 +1,15 @@
+model,model_type,MASE_6750_scaled,CRPS_6750_scaled,Rank_6750_scaled,MAE_663_unscaled,CRPS_663_unscaled,Rank_663_unscaled
+Toto-Open-Base-1.0,pretrained,0.617,0.375,2.351,0.001,0.025,7.549
+moirai_1.1_base,pretrained,0.710,0.428,4.278,0.000,0.003,5.644
+moirai_1.1_large,pretrained,0.720,0.436,4.499,0.001,0.005,6.707
+moirai_1.1_small,pretrained,0.738,0.447,4.796,0.001,0.009,7.404
+timesfm_2_0_500m,pretrained,0.725,0.447,5.153,0.014,0.091,10.029
+chronos_bolt_base,pretrained,0.726,0.451,5.446,0.003,0.019,7.682
+chronos_bolt_small,pretrained,0.733,0.455,5.793,0.003,0.022,8.140
+autoarima,statistical,0.824,0.736,9.171,0.000,0.001,5.496
+timer,pretrained,0.796,0.639,9.356,0.001,0.005,6.474
+time-moe,pretrained,0.806,0.649,9.369,0.001,0.005,8.505
+visionts,pretrained,0.991,0.675,10.336,0.001,0.009,8.538
+autoets,statistical,0.842,1.975,10.956,0.000,0.030,6.992
+autotheta,statistical,1.123,1.018,11.712,0.001,0.002,6.513
+naive,statistical,1.000,1.000,11.783,0.000,0.006,9.326

src/about.py

@@ -1,6 +1,7 @@
 from dataclasses import dataclass
 from enum import Enum
 
+
 @dataclass
 class Task:
     benchmark: str
@@ -11,24 +12,25 @@ class Task:
 # Select your tasks here
 # ---------------------------------------------------
 class Tasks(Enum):
-    # task_key in the json file, metric_key in the json file, name to display in the leaderboard 
+    # task_key in the json file, metric_key in the json file, name to display in the leaderboard
     task0 = Task("anli_r1", "acc", "ANLI")
     task1 = Task("logiqa", "acc_norm", "LogiQA")
 
-NUM_FEWSHOT = 0 # Change with your few shot
-# ---------------------------------------------------
 
+NUM_FEWSHOT = 0  # Change with your few shot
+# ---------------------------------------------------
 
 
 # Your leaderboard name
-TITLE = """<h1 align="center" id="space-title">Demo leaderboard</h1>"""
+TITLE = """<h1 align="center" id="space-title">BOOM 💥 Time Series Forecasting Leaderboard</h1>"""
 
 # What does your leaderboard evaluate?
 INTRODUCTION_TEXT = """
-Intro text
+BOOM (Benchmark of Observability Metrics) is a large-scale, real-world time series dataset designed for evaluating models on forecasting tasks in complex observability environments. Composed of real-world metrics data collected from Datadog, a leading observability platform, the benchmark captures the irregularity, structural complexity, and heavy-tailed statistics typical of production observability data. For more information, please refer to the [BOOM Dataset Card](https://huggingface.co/datasets/Datadog/BOOM) and the [BOOM GitHub repository](https://github.com/DataDog/toto?tab=readme-ov-file#boom-benchmark-of-observability-metrics)
 """
 
 # Which evaluations are you running? how can people reproduce what you have?
+# TODO
 LLM_BENCHMARKS_TEXT = f"""
 ## How it works
 
@@ -69,4 +71,12 @@ If everything is done, check you can launch the EleutherAIHarness on your model
 
 CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
 CITATION_BUTTON_TEXT = r"""
+@misc{toto2025,
+  title={This Time is Different: An Observability Perspective on Time Series Foundation Models},
+  author={TODO},
+  year={2025},
+  eprint={arXiv:TODO},
+  archivePrefix={arXiv},
+  primaryClass={cs.LG}
+}
 """

src/display/utils.py

@@ -5,6 +5,7 @@ import pandas as pd
 
 from src.about import Tasks
 
+
 def fields(raw_class):
     return [v for k, v in raw_class.__dict__.items() if k[:2] != "__" and k[-2:] != "__"]
 
@@ -20,29 +21,34 @@ class ColumnContent:
     hidden: bool = False
     never_hidden: bool = False
 
+
 ## Leaderboard columns
 auto_eval_column_dict = []
 # Init
 auto_eval_column_dict.append(["model_type_symbol", ColumnContent, ColumnContent("T", "str", True, never_hidden=True)])
 auto_eval_column_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
-#Scores
-auto_eval_column_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
-for task in Tasks:
-    auto_eval_column_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
+# Scores
+auto_eval_column_dict.append(["MASE_6750_scaled", ColumnContent, ColumnContent("MASE_scaled", "number", True)])
+auto_eval_column_dict.append(["CRPS_6750_scaled", ColumnContent, ColumnContent("CRPS_scaled", "number", True)])
+auto_eval_column_dict.append(["Rank_6750_scaled", ColumnContent, ColumnContent("Rank_scaled", "number", True)])
+auto_eval_column_dict.append(["MAE_663_unscaled", ColumnContent, ColumnContent("MAE[0.5]_unscaled", "number", True)])
+auto_eval_column_dict.append(["CRPS_663_unscaled", ColumnContent, ColumnContent("CRPS_unscaled", "number", True)])
+auto_eval_column_dict.append(["Rank_663_unscaled", ColumnContent, ColumnContent("Rank_unscaled", "number", True)])
 # Model information
-auto_eval_column_dict.append(["model_type", ColumnContent, ColumnContent("Type", "str", False)])
-auto_eval_column_dict.append(["architecture", ColumnContent, ColumnContent("Architecture", "str", False)])
-auto_eval_column_dict.append(["weight_type", ColumnContent, ColumnContent("Weight type", "str", False, True)])
-auto_eval_column_dict.append(["precision", ColumnContent, ColumnContent("Precision", "str", False)])
-auto_eval_column_dict.append(["license", ColumnContent, ColumnContent("Hub License", "str", False)])
-auto_eval_column_dict.append(["params", ColumnContent, ColumnContent("#Params (B)", "number", False)])
-auto_eval_column_dict.append(["likes", ColumnContent, ColumnContent("Hub ❤️", "number", False)])
-auto_eval_column_dict.append(["still_on_hub", ColumnContent, ColumnContent("Available on the hub", "bool", False)])
-auto_eval_column_dict.append(["revision", ColumnContent, ColumnContent("Model sha", "str", False, False)])
+auto_eval_column_dict.append(["model_type", ColumnContent, ColumnContent("Type", "str", False, hidden=True)])
+# auto_eval_column_dict.append(["architecture", ColumnContent, ColumnContent("Architecture", "str", False)])
+# auto_eval_column_dict.append(["weight_type", ColumnContent, ColumnContent("Weight type", "str", False, True)])
+# auto_eval_column_dict.append(["precision", ColumnContent, ColumnContent("Precision", "str", False)])
+# auto_eval_column_dict.append(["license", ColumnContent, ColumnContent("Hub License", "str", False)])
+# auto_eval_column_dict.append(["params", ColumnContent, ColumnContent("#Params (B)", "number", False)])
+# auto_eval_column_dict.append(["likes", ColumnContent, ColumnContent("Hub ❤️", "number", False)])
+# auto_eval_column_dict.append(["still_on_hub", ColumnContent, ColumnContent("Available on the hub", "bool", False)])
+# auto_eval_column_dict.append(["revision", ColumnContent, ColumnContent("Model sha", "str", False, False)])
 
 # We use make dataclass to dynamically fill the scores from Tasks
 AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
 
+
 ## For the queue columns in the submission tab
 @dataclass(frozen=True)
 class EvalQueueColumn:  # Queue column
@@ -53,19 +59,21 @@ class EvalQueueColumn:  # Queue column
     weight_type = ColumnContent("weight_type", "str", "Original")
     status = ColumnContent("status", "str", True)
 
+
 ## All the model information that we might need
 @dataclass
 class ModelDetails:
     name: str
     display_name: str = ""
-    symbol: str = "" # emoji
+    symbol: str = ""  # emoji
 
 
 class ModelType(Enum):
-    PT = ModelDetails(name="pretrained", symbol="🟢")
-    FT = ModelDetails(name="fine-tuned", symbol="🔶")
-    IFT = ModelDetails(name="instruction-tuned", symbol="⭕")
-    RL = ModelDetails(name="RL-tuned", symbol="🟦")
+    PT = ModelDetails(name="🟢 pretrained", symbol="🟢")
+    FT = ModelDetails(name="🔶 fine-tuned", symbol="🔶")
+    DL = ModelDetails(name="🔷 deep-learning", symbol="🔷")
+    ST = ModelDetails(name="🟣 statistical", symbol="🟣")
+
     Unknown = ModelDetails(name="", symbol="?")
 
     def to_str(self, separator=" "):
@@ -77,17 +85,19 @@ class ModelType(Enum):
             return ModelType.FT
         if "pretrained" in type or "🟢" in type:
             return ModelType.PT
-        if "RL-tuned" in type or "🟦" in type:
-            return ModelType.RL
-        if "instruction-tuned" in type or "⭕" in type:
-            return ModelType.IFT
+        if "deep-learning" in type or "🟦" in type:
+            return ModelType.DL
+        if "statistical" in type or "🟣" in type:
+            return ModelType.ST
         return ModelType.Unknown
 
+
 class WeightType(Enum):
     Adapter = ModelDetails("Adapter")
     Original = ModelDetails("Original")
     Delta = ModelDetails("Delta")
 
+
 class Precision(Enum):
     float16 = ModelDetails("float16")
     bfloat16 = ModelDetails("bfloat16")
@@ -100,6 +110,7 @@ class Precision(Enum):
             return Precision.bfloat16
         return Precision.Unknown
 
+
 # Column selection
 COLS = [c.name for c in fields(AutoEvalColumn) if not c.hidden]
 
@@ -107,4 +118,3 @@ EVAL_COLS = [c.name for c in fields(EvalQueueColumn)]
 EVAL_TYPES = [c.type for c in fields(EvalQueueColumn)]
 
 BENCHMARK_COLS = [t.value.col_name for t in Tasks]
-

src/envs.py

@@ -4,21 +4,24 @@ from huggingface_hub import HfApi
 
 # Info to change for your repository
 # ----------------------------------
-TOKEN = os.environ.get("HF_TOKEN") # A read/write token for your org
+TOKEN = os.environ.get("HF_TOKEN")  # A read/write token for your org
 
-OWNER = "demo-leaderboard-backend" # Change to your org - don't forget to create a results and request dataset, with the correct format!
+OWNER = (
+    "Datadog"  # Change to your org - don't forget to create a results and request dataset, with the correct format!
+)
 # ----------------------------------
 
-REPO_ID = f"{OWNER}/leaderboard"
+REPO_ID = f"{OWNER}/BOOM-Leaderboard"  # The repo id of your space
 QUEUE_REPO = f"{OWNER}/requests"
 RESULTS_REPO = f"{OWNER}/results"
 
 # If you setup a cache later, just change HF_HOME
-CACHE_PATH=os.getenv("HF_HOME", ".")
+CACHE_PATH = os.getenv("HF_HOME", ".")
 
 # Local caches
 EVAL_REQUESTS_PATH = os.path.join(CACHE_PATH, "eval-queue")
-EVAL_RESULTS_PATH = os.path.join(CACHE_PATH, "eval-results")
+# EVAL_RESULTS_PATH = os.path.join(CACHE_PATH, "eval-results")
+EVAL_RESULTS_PATH = os.path.join(CACHE_PATH, "results")
 EVAL_REQUESTS_PATH_BACKEND = os.path.join(CACHE_PATH, "eval-queue-bk")
 EVAL_RESULTS_PATH_BACKEND = os.path.join(CACHE_PATH, "eval-results-bk")
 

src/populate.py

@@ -2,23 +2,65 @@ import json
 import os
 
 import pandas as pd
-
+from dataclasses import fields
 from src.display.formatting import has_no_nan_values, make_clickable_model
 from src.display.utils import AutoEvalColumn, EvalQueueColumn
 from src.leaderboard.read_evals import get_raw_eval_results
+from src.display.utils import ModelType
+
+
+# def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
+#     """Creates a dataframe from all the individual experiment results"""
+#     raw_data = get_raw_eval_results(results_path, requests_path)
+#     all_data_json = [v.to_dict() for v in raw_data]
+
+#     df = pd.DataFrame.from_records(all_data_json)
+#     df = df.sort_values(by=[AutoEvalColumn.average.name], ascending=False)
+#     df = df[cols].round(decimals=2)
+
+#     # filter out if any of the benchmarks have not been produced
+#     df = df[has_no_nan_values(df, benchmark_cols)]
+#     return df
 
 
 def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
-    """Creates a dataframe from all the individual experiment results"""
-    raw_data = get_raw_eval_results(results_path, requests_path)
-    all_data_json = [v.to_dict() for v in raw_data]
+    """
+    Processes a STATIC results CSV file to generate a leaderboard DataFrame with formatted columns and sorted values.
+    Args:
+        results_path (str): The file path to the results CSV file.
+    Returns:
+        pd.DataFrame: A processed DataFrame with renamed columns, additional formatting, and sorted values.
+    Notes:
+        - The function reads a CSV file from the given `results_path`.
+        - Internal column names are mapped to display names using `AutoEvalColumn`.
+        - A new column for model type symbols is created by parsing the `model_type` column.
+        - The `model_type` column is updated to prepend the model type symbol.
+        - The DataFrame is sorted by the `Rank_6750_scaled` column in ascending order.
+    """
+
+    df = pd.read_csv(results_path)
+    # Create the mapping from internal column name to display name
+
+    column_mapping = {field.name: getattr(AutoEvalColumn, field.name).name for field in fields(AutoEvalColumn)}
+    # Assuming `df` is your DataFrame:
+    df.rename(columns=column_mapping, inplace=True)
+
+    # Create a new column for model type symbol by parsing the model_type column
+    df[AutoEvalColumn.model_type_symbol.name] = df[AutoEvalColumn.model_type.name].apply(
+        lambda x: ModelType.from_str(x).value.symbol
+    )
+    # Prepend the value of model_type_symbol to the value of model_type
+    df[AutoEvalColumn.model_type.name] = (
+        df[AutoEvalColumn.model_type_symbol.name] + " " + df[AutoEvalColumn.model_type.name]
+    )
 
-    df = pd.DataFrame.from_records(all_data_json)
-    df = df.sort_values(by=[AutoEvalColumn.average.name], ascending=False)
-    df = df[cols].round(decimals=2)
+    # Move the model_type_symbol column to the beginning
+    cols = [AutoEvalColumn.model_type_symbol.name] + [
+        col for col in df.columns if col != AutoEvalColumn.model_type_symbol.name
+    ]
+    df = df[cols]
 
-    # filter out if any of the benchmarks have not been produced
-    df = df[has_no_nan_values(df, benchmark_cols)]
+    df = df.sort_values(by=[AutoEvalColumn.Rank_6750_scaled.name], ascending=True)
     return df
 
 
@@ -39,7 +81,9 @@ def get_evaluation_queue_df(save_path: str, cols: list) -> list[pd.DataFrame]:
             all_evals.append(data)
         elif ".md" not in entry:
             # this is a folder
-            sub_entries = [e for e in os.listdir(f"{save_path}/{entry}") if os.path.isfile(e) and not e.startswith(".")]
+            sub_entries = [
+                e for e in os.listdir(f"{save_path}/{entry}") if os.path.isfile(e) and not e.startswith(".")
+            ]
             for sub_entry in sub_entries:
                 file_path = os.path.join(save_path, entry, sub_entry)
                 with open(file_path) as fp: