diff --git a/tabby/tasks/update_dataset.yaml b/tabby/tasks/update_dataset.yaml
index 90ef220..ba0ccd5 100644
--- a/tabby/tasks/update_dataset.yaml
+++ b/tabby/tasks/update_dataset.yaml
@@ -7,12 +7,12 @@ env:
   - GIT_REPOSITORIES_DIR: "$GIT_REPOSITORIES_DIR"
   - DATASET_DIR: "$DATASET_DIR"
 steps:
-  - name: Update repositories
+  - name: update repositories
     dir: $APP_DIR
     command: python -m tabby.tools.repository.updater --data_dir=$GIT_REPOSITORIES_DIR --config_file=$CONFIG_FILE
 
-  - name: Generate dataset
+  - name: generate dataset
     dir: $APP_DIR
-    command: python -m tabby.tools.preprocess.build_dataset --project_dir=$GIT_REPOSITORIES_DIR --output_dir=$DATASET_DIR
+    command: python -m tabby.tools.build_dataset --project_dir=$GIT_REPOSITORIES_DIR --output_dir=$DATASET_DIR
     depends:
-      - Update repositories
+      - update repositories
diff --git a/tabby/tools/preprocess/build_dataset.py b/tabby/tools/build_dataset/__main__.py
similarity index 83%
rename from tabby/tools/preprocess/build_dataset.py
rename to tabby/tools/build_dataset/__main__.py
index e9a52a2..36645c3 100644
--- a/tabby/tools/preprocess/build_dataset.py
+++ b/tabby/tools/build_dataset/__main__.py
@@ -2,10 +2,11 @@ import glob
 import json
 import os
 
+import pandas as pd
 from datasets import Dataset
 from transformers import HfArgumentParser
 
-from . import filters, metrics
+from . import metrics
 from .args import PreprocessProjectArgs
 
 
@@ -61,6 +62,17 @@ def dataset_iter(files):
     return gen
 
 
+def count_by_language(dataset):
+    key = "language"
+    df = (
+        pd.DataFrame(dataset[key], columns=[key])
+        .groupby([key])
+        .size()
+        .to_frame("count")
+    )
+    return df
+
+
 if __name__ == "__main__":
     valid_extensions = read_valid_extensions()
 
@@ -80,5 +92,8 @@ if __name__ == "__main__":
     )
 
     ds = Dataset.from_generator(dataset_iter(files))
-    ds = ds.filter(filters.basic_filters(args))
     ds.save_to_disk(args.output_dir)
+
+    print("\n## Summary")
+    print("Number of source files", len(ds))
+    print("Number of source files by languages", count_by_language(ds).to_json())
diff --git a/tabby/tools/build_dataset/args.py b/tabby/tools/build_dataset/args.py
new file mode 100644
index 0000000..99c4af3
--- /dev/null
+++ b/tabby/tools/build_dataset/args.py
@@ -0,0 +1,14 @@
+from dataclasses import dataclass, field
+from typing import Optional
+
+
+@dataclass
+class PreprocessProjectArgs:
+    # add arguments in the following format
+    project_dir: Optional[str] = field(
+        metadata={"help": "Project directory."},
+    )
+
+    output_dir: Optional[str] = field(
+        metadata={"help": "Output save path directory."},
+    )
diff --git a/tabby/tools/preprocess/metrics.py b/tabby/tools/build_dataset/metrics.py
similarity index 100%
rename from tabby/tools/preprocess/metrics.py
rename to tabby/tools/build_dataset/metrics.py
diff --git a/tabby/tools/preprocess/programming-languages-to-file-extensions.json b/tabby/tools/build_dataset/programming-languages-to-file-extensions.json
similarity index 100%
rename from tabby/tools/preprocess/programming-languages-to-file-extensions.json
rename to tabby/tools/build_dataset/programming-languages-to-file-extensions.json
diff --git a/tabby/tools/preprocess/args.py b/tabby/tools/preprocess/args.py
deleted file mode 100644
index f55b113..0000000
--- a/tabby/tools/preprocess/args.py
+++ /dev/null
@@ -1,30 +0,0 @@
-from dataclasses import dataclass, field
-from typing import Optional
-
-
-@dataclass(kw_only=True)
-class FilterArgs:
-    line_max: Optional[int] = field(
-        default=1000,
-        metadata={"help": "Max line length allowed"},
-    )
-    line_mean: Optional[int] = field(
-        default=100,
-        metadata={"help": "Mean line length allowed"},
-    )
-    alpha_frac: Optional[float] = field(
-        default=0.25,
-        metadata={"help": "Minimum fraction of alphanumeric characters allowed."},
-    )
-
-
-@dataclass
-class PreprocessProjectArgs(FilterArgs):
-    # add arguments in the following format
-    project_dir: Optional[str] = field(
-        metadata={"help": "Project directory."},
-    )
-
-    output_dir: Optional[str] = field(
-        metadata={"help": "Output save path directory."},
-    )
diff --git a/tabby/tools/preprocess/filters.py b/tabby/tools/preprocess/filters.py
deleted file mode 100644
index 11c7fe3..0000000
--- a/tabby/tools/preprocess/filters.py
+++ /dev/null
@@ -1,15 +0,0 @@
-from .args import FilterArgs
-
-
-def basic_filters(args: FilterArgs):
-    def fn(example):
-        """Filter files based on line length and % alphanumeric characters"""
-        if example["max_line_length"] > args.line_max:
-            return False
-        elif example["avg_line_length"] > args.line_mean:
-            return False
-        elif example["alphanum_fraction"] < args.alpha_frac:
-            return False
-        return True
-
-    return fn