fixed model

Dockerfile

@@ -1,14 +0,0 @@
-FROM python:3.12-slim
-# Set the working directory
-# All subsequent commands will run from here.
-WORKDIR /app
-RUN apt-get update && apt-get upgrade -y && rm -rf /var/lib/apt/lists/*
-# Copy project files into the container.
-# This copies everything from your local directory into the container's /app directory.
-COPY . .
-# Install all the Python packages listed in your requirements.txt.
-# The --no-cache-dir flag keeps the final image size smaller.
-RUN pip install --no-cache-dir --upgrade -r requirements.txt
-# Define the default command to run when the Space starts.
-# This launches a JupyterLab server that is accessible from the web.
-CMD ["jupyter", "lab", "--ip=0.0.0.0", "--port=7860", "--allow-root", "--NotebookApp.token=''"]

README.md

@@ -14,3 +14,9 @@ pip install git+https://github.com/ccss17/toy-duration-predictor.git
 ```python
 import toy_duration_predictor as tdp
 ```
+
+## Hugging Face 
+
+- Model: https://huggingface.co/ccss17/toy-duration-predictor
+- Dataset: https://huggingface.co/datasets/ccss17/note-duration-dataset
+- Spaces: https://huggingface.co/spaces/ccss17/toy-duration-predictor

requirements-dev.txt

@@ -1,162 +0,0 @@
--e .
-aiohappyeyeballs==2.6.1
-aiohttp==3.12.13
-aiohttp-cors==0.8.1
-aiosignal==1.3.2
-annotated-types==0.7.0
-anyio==4.9.0
-appnope==0.1.4 ; sys_platform == 'darwin'
-asttokens==3.0.0
-async-timeout==5.0.1 ; python_full_version < '3.11'
-attrs==25.3.0
-black==25.1.0
-cachetools==5.5.2
-certifi==2025.6.15
-cffi==1.17.1 ; implementation_name == 'pypy'
-charset-normalizer==3.4.2
-click==8.1.8 ; python_full_version < '3.10'
-click==8.2.1 ; python_full_version >= '3.10'
-colorama==0.4.6 ; sys_platform == 'win32'
-colorful==0.5.6
-comm==0.2.2
-datasets==3.6.0
-debugpy==1.8.14
-decorator==5.2.1
-dill==0.3.8
-distlib==0.3.9
-exceptiongroup==1.3.0 ; python_full_version < '3.11'
-executing==2.2.0
-fastapi==0.115.13
-filelock==3.18.0
-frozenlist==1.7.0
-fsspec==2025.3.0
-google-api-core==2.25.1
-google-auth==2.40.3
-googleapis-common-protos==1.70.0
-grpcio==1.73.0
-h11==0.16.0
-hf-xet==1.1.5 ; platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'
-httptools==0.6.4
-huggingface-hub==0.33.0
-idna==3.10
-importlib-metadata==8.7.0
-ipykernel==6.29.5
-ipython==8.18.1 ; python_full_version < '3.10'
-ipython==8.37.0 ; python_full_version == '3.10.*'
-ipython==9.3.0 ; python_full_version >= '3.11'
-ipython-pygments-lexers==1.1.1 ; python_full_version >= '3.11'
-ipywidgets==8.1.7
-jedi==0.19.2
-jinja2==3.1.6
-jsonschema==4.24.0
-jsonschema-specifications==2025.4.1
-jupyter-client==8.6.3
-jupyter-core==5.8.1
-jupyterlab-widgets==3.0.15
-llvmlite==0.44.0 ; python_full_version >= '3.10'
-markdown-it-py==3.0.0
-markupsafe==2.1.5 ; python_full_version < '3.10'
-markupsafe==3.0.2 ; python_full_version >= '3.10'
-matplotlib-inline==0.1.7
-mdurl==0.1.2
-midii==0.1.19 ; python_full_version < '3.10'
-midii==0.1.41 ; python_full_version >= '3.10'
-mido==1.3.3
-mpmath==1.3.0
-msgpack==1.1.1
-multidict==6.5.0
-multiprocess==0.70.16
-mypy-extensions==1.1.0
-nest-asyncio==1.6.0
-networkx==3.2.1 ; python_full_version < '3.10'
-networkx==3.4.2 ; python_full_version == '3.10.*'
-networkx==3.5 ; python_full_version >= '3.11'
-numba==0.61.2 ; python_full_version >= '3.10'
-numpy==2.0.2 ; python_full_version < '3.10'
-numpy==2.2.6 ; python_full_version >= '3.10'
-nvidia-cublas-cu12==12.6.4.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cuda-cupti-cu12==12.6.80 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cuda-nvrtc-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cuda-runtime-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cudnn-cu12==9.5.1.17 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cufft-cu12==11.3.0.4 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cufile-cu12==1.11.1.6 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-curand-cu12==10.3.7.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cusolver-cu12==11.7.1.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cusparse-cu12==12.5.4.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-cusparselt-cu12==0.6.3 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-nccl-cu12==2.26.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-nvjitlink-cu12==12.6.85 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-nvidia-nvtx-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-opencensus==0.11.4
-opencensus-context==0.1.3
-opentelemetry-api==1.34.1
-opentelemetry-exporter-prometheus==0.55b1
-opentelemetry-proto==1.34.1
-opentelemetry-sdk==1.34.1
-opentelemetry-semantic-conventions==0.55b1
-packaging==25.0
-pandas==2.3.0
-parso==0.8.4
-pathspec==0.12.1
-pexpect==4.9.0 ; (python_full_version < '3.10' and sys_platform == 'emscripten') or (sys_platform != 'emscripten' and sys_platform != 'win32')
-platformdirs==4.3.8
-prometheus-client==0.22.1
-prompt-toolkit==3.0.51
-propcache==0.3.2
-proto-plus==1.26.1
-protobuf==5.29.5
-psutil==7.0.0
-ptyprocess==0.7.0 ; (python_full_version < '3.10' and sys_platform == 'emscripten') or (sys_platform != 'emscripten' and sys_platform != 'win32')
-pure-eval==0.2.3
-py-spy==0.4.0
-pyarrow==17.0.0
-pyasn1==0.6.1
-pyasn1-modules==0.4.2
-pycparser==2.22 ; implementation_name == 'pypy'
-pydantic==2.11.7
-pydantic-core==2.33.2
-pygments==2.19.2
-python-dateutil==2.9.0.post0
-python-dotenv==1.1.1
-pytz==2025.2
-pywin32==310 ; platform_python_implementation != 'PyPy' and sys_platform == 'win32'
-pyyaml==6.0.2
-pyzmq==27.0.0
-ray==2.47.1
-referencing==0.36.2
-requests==2.32.4
-rich==14.0.0
-rpds-py==0.25.1
-rsa==4.9.1
-ruff==0.12.0
-setuptools==80.9.0 ; (python_full_version >= '3.12' and platform_machine != 'x86_64') or (python_full_version >= '3.12' and sys_platform != 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux')
-six==1.17.0
-smart-open==7.1.0
-sniffio==1.3.1
-stack-data==0.6.3
-starlette==0.46.2
-sympy==1.14.0
-tensorboardx==2.6.4
-tomli==2.2.1 ; python_full_version < '3.11'
-torch==2.7.1
-tornado==6.5.1
-tqdm==4.67.1
-traitlets==5.14.3
-triton==3.3.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
-typing-extensions==4.14.0
-typing-inspection==0.4.1
-tzdata==2025.2
-urllib3==2.5.0
-uvicorn==0.34.3
-uvloop==0.21.0 ; platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'
-virtualenv==20.31.2
-watchfiles==1.1.0
-wcwidth==0.2.13
-websockets==12.0 ; python_full_version < '3.10'
-websockets==15.0.1 ; python_full_version >= '3.10'
-widgetsnbextension==4.0.14
-wrapt==1.17.2
-xxhash==3.5.0
-yarl==1.20.1
-zipp==3.23.0

requirements.txt

@@ -1,23 +1,30 @@
--e .
 aiohappyeyeballs==2.6.1
 aiohttp==3.12.13
 aiohttp-cors==0.8.1
 aiosignal==1.3.2
 annotated-types==0.7.0
 anyio==4.9.0
+appnope==0.1.4 ; sys_platform == 'darwin'
+asttokens==3.0.0
 async-timeout==5.0.1 ; python_full_version < '3.11'
 attrs==25.3.0
+black==25.1.0
 cachetools==5.5.2
 certifi==2025.6.15
+cffi==1.17.1 ; implementation_name == 'pypy'
 charset-normalizer==3.4.2
 click==8.1.8 ; python_full_version < '3.10'
 click==8.2.1 ; python_full_version >= '3.10'
 colorama==0.4.6 ; sys_platform == 'win32'
 colorful==0.5.6
+comm==0.2.2
 datasets==3.6.0
+debugpy==1.8.14
+decorator==5.2.1
 dill==0.3.8
 distlib==0.3.9
 exceptiongroup==1.3.0 ; python_full_version < '3.11'
+executing==2.2.0
 fastapi==0.115.13
 filelock==3.18.0
 frozenlist==1.7.0
@@ -32,13 +39,24 @@ httptools==0.6.4
 huggingface-hub==0.33.0
 idna==3.10
 importlib-metadata==8.7.0
+ipykernel==6.29.5
+ipython==8.18.1 ; python_full_version < '3.10'
+ipython==8.37.0 ; python_full_version == '3.10.*'
+ipython==9.3.0 ; python_full_version >= '3.11'
+ipython-pygments-lexers==1.1.1 ; python_full_version >= '3.11'
+ipywidgets==8.1.7
+jedi==0.19.2
 jinja2==3.1.6
 jsonschema==4.24.0
 jsonschema-specifications==2025.4.1
+jupyter-client==8.6.3
+jupyter-core==5.8.1
+jupyterlab-widgets==3.0.15
 llvmlite==0.44.0 ; python_full_version >= '3.10'
 markdown-it-py==3.0.0
 markupsafe==2.1.5 ; python_full_version < '3.10'
 markupsafe==3.0.2 ; python_full_version >= '3.10'
+matplotlib-inline==0.1.7
 mdurl==0.1.2
 midii==0.1.19 ; python_full_version < '3.10'
 midii==0.1.41 ; python_full_version >= '3.10'
@@ -47,6 +65,8 @@ mpmath==1.3.0
 msgpack==1.1.1
 multidict==6.5.0
 multiprocess==0.70.16
+mypy-extensions==1.1.0
+nest-asyncio==1.6.0
 networkx==3.2.1 ; python_full_version < '3.10'
 networkx==3.4.2 ; python_full_version == '3.10.*'
 networkx==3.5 ; python_full_version >= '3.11'
@@ -76,37 +96,52 @@ opentelemetry-sdk==1.34.1
 opentelemetry-semantic-conventions==0.55b1
 packaging==25.0
 pandas==2.3.0
+parso==0.8.4
+pathspec==0.12.1
+pexpect==4.9.0 ; (python_full_version < '3.10' and sys_platform == 'emscripten') or (sys_platform != 'emscripten' and sys_platform != 'win32')
 platformdirs==4.3.8
 prometheus-client==0.22.1
+prompt-toolkit==3.0.51
 propcache==0.3.2
 proto-plus==1.26.1
 protobuf==5.29.5
+psutil==7.0.0
+ptyprocess==0.7.0 ; (python_full_version < '3.10' and sys_platform == 'emscripten') or (sys_platform != 'emscripten' and sys_platform != 'win32')
+pure-eval==0.2.3
 py-spy==0.4.0
 pyarrow==17.0.0
 pyasn1==0.6.1
 pyasn1-modules==0.4.2
+pycparser==2.22 ; implementation_name == 'pypy'
 pydantic==2.11.7
 pydantic-core==2.33.2
 pygments==2.19.2
 python-dateutil==2.9.0.post0
 python-dotenv==1.1.1
 pytz==2025.2
+pywin32==310 ; platform_python_implementation != 'PyPy' and sys_platform == 'win32'
 pyyaml==6.0.2
+pyzmq==27.0.0
 ray==2.47.1
 referencing==0.36.2
 requests==2.32.4
 rich==14.0.0
 rpds-py==0.25.1
 rsa==4.9.1
+ruff==0.12.0
 setuptools==80.9.0 ; (python_full_version >= '3.12' and platform_machine != 'x86_64') or (python_full_version >= '3.12' and sys_platform != 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux')
 six==1.17.0
 smart-open==7.1.0
 sniffio==1.3.1
+stack-data==0.6.3
 starlette==0.46.2
 sympy==1.14.0
 tensorboardx==2.6.4
+tomli==2.2.1 ; python_full_version < '3.11'
 torch==2.7.1
+tornado==6.5.1
 tqdm==4.67.1
+traitlets==5.14.3
 triton==3.3.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
 typing-extensions==4.14.0
 typing-inspection==0.4.1
@@ -116,9 +151,14 @@ uvicorn==0.34.3
 uvloop==0.21.0 ; platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'
 virtualenv==20.31.2
 watchfiles==1.1.0
+wcwidth==0.2.13
 websockets==12.0 ; python_full_version < '3.10'
 websockets==15.0.1 ; python_full_version >= '3.10'
+widgetsnbextension==4.0.14
 wrapt==1.17.2
 xxhash==3.5.0
 yarl==1.20.1
 zipp==3.23.0
+
+jupyterlab
+git+https://github.com/ccss17/toy-duration-predictor.git

src/toy_duration_predictor/__init__.py

@@ -1,2 +1,3 @@
-# from .preprocess import mssv
-# from .preprocess import utils
+from .train import *
+from .upload import *
+from .load import *

src/toy_duration_predictor/preprocess/__init__.py

@@ -0,0 +1 @@
+from .mssv import *

src/toy_duration_predictor/preprocess/mssv.py

@@ -137,7 +137,7 @@ def process_midi_flat_map(row: Dict[str, Any]) -> List[Dict[str, Any]]:
 
 
 def preprocess_dataset(midi_file_directory, output_parquet_path):
-    context = ray.init()
+    context = ray.init(ignore_reinit_error=True)
     print(context.dashboard_url)
 
     all_midi_paths = Path(midi_file_directory).rglob("*.mid")

src/toy_duration_predictor/train.py

@@ -12,10 +12,10 @@ BATCH_SIZE = 64
 
 MODEL_CONFIG = {
     "num_singers": 18,
-    "singer_embedding_dim": 64,
-    "hidden_size": 1024,
-    "num_layers": 4,
-    "dropout": 0.5,
+    "singer_embedding_dim": 32,
+    "hidden_size": 256,
+    "num_layers": 3,
+    "dropout": 0.4,
 }
 
 LEARNING_RATE = 1e-4

test.ipynb

@@ -10,17 +10,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "id": "d05b06c2",
    "metadata": {},
    "outputs": [],
    "source": [
-    "from toy_duration_predictor.preprocess import mssv"
+    "from toy_duration_predictor import preprocess"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 2,
    "id": "77c0576d",
    "metadata": {},
    "outputs": [],
@@ -37,7 +37,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "id": "ce7e5029",
    "metadata": {},
    "outputs": [
@@ -45,138 +45,27 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "2025-06-23 15:44:45,964\tINFO worker.py:1908 -- Started a local Ray instance. View the dashboard at \u001b[1m\u001b[32mhttp://127.0.0.1:8265 \u001b[39m\u001b[22m\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "127.0.0.1:8265\n",
-      "Ray cluster started: {'CPU': 12.0, 'node:__internal_head__': 1.0, 'object_store_memory': 918235545.0, 'node:172.20.119.165': 1.0, 'memory': 2142549607.0}\n",
-      "Launching parallel processing tasks...\n",
-      "\u001b[36m(process_midi_to_dict pid=30235)\u001b[0m Error processing /mnt/d/dataset/004.다화자 가창 데이터/01.데이터/1.Training/라벨링데이터/01.발라드R&B/A. 남성/01. 20대/가창자_s02/ba_05688_-4_a_s02_m_02.mid: 'quantized duration'\n",
-      "\u001b[36m(process_midi_to_dict pid=30235)\u001b[0m Error processing /mnt/d/dataset/004.다화자 가창 데이터/01.데이터/1.Training/라벨링데이터/01.발라드R&B/A. 남성/02. 30대/가창자_s09/ba_14343_-1_a_s09_m_03.mid: 'quantized duration'\u001b[32m [repeated 208x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/user-guides/configure-logging.html#log-deduplication for more options.)\u001b[0m\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2025-06-23 15:44:51,225\tWARNING arrow.py:165 -- Failed to convert column 'item' into pyarrow array due to: Error converting data to Arrow: [ObjectRef(c8ef45ccd0112571ffffffffffffffffffffffff0100000001000000), ObjectRef(16310a0f0a45af5cffffffffffffffffffffffff0100000001000000), ObjectRef(c2668a65bda616c1ffffffffffffffffffffffff01000000010...; falling back to serialize as pickled python objects\n",
-      "Traceback (most recent call last):\n",
-      "  File \"/home/ccsss/repo/toy-duration-predictor/.venv/lib/python3.12/site-packages/ray/air/util/tensor_extensions/arrow.py\", line 204, in _convert_to_pyarrow_native_array\n",
-      "    pa_type = _infer_pyarrow_type(column_values)\n",
-      "              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
-      "  File \"/home/ccsss/repo/toy-duration-predictor/.venv/lib/python3.12/site-packages/ray/air/util/tensor_extensions/arrow.py\", line 317, in _infer_pyarrow_type\n",
-      "    inferred_pa_dtype = pa.infer_type(column_values)\n",
-      "                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
-      "  File \"pyarrow/array.pxi\", line 564, in pyarrow.lib.infer_type\n",
-      "  File \"pyarrow/error.pxi\", line 155, in pyarrow.lib.pyarrow_internal_check_status\n",
-      "  File \"pyarrow/error.pxi\", line 92, in pyarrow.lib.check_status\n",
-      "pyarrow.lib.ArrowInvalid: Could not convert ObjectRef(c8ef45ccd0112571ffffffffffffffffffffffff0100000001000000) with type ray._raylet.ObjectRef: did not recognize Python value type when inferring an Arrow data type\n",
-      "\n",
-      "The above exception was the direct cause of the following exception:\n",
-      "\n",
-      "Traceback (most recent call last):\n",
-      "  File \"/home/ccsss/repo/toy-duration-predictor/.venv/lib/python3.12/site-packages/ray/air/util/tensor_extensions/arrow.py\", line 117, in convert_to_pyarrow_array\n",
-      "    return _convert_to_pyarrow_native_array(column_values, column_name)\n",
-      "           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
-      "  File \"/home/ccsss/repo/toy-duration-predictor/.venv/lib/python3.12/site-packages/ray/air/util/tensor_extensions/arrow.py\", line 231, in _convert_to_pyarrow_native_array\n",
-      "    raise ArrowConversionError(str(column_values)) from e\n",
-      "ray.air.util.tensor_extensions.arrow.ArrowConversionError: Error converting data to Arrow: [ObjectRef(c8ef45ccd0112571ffffffffffffffffffffffff0100000001000000), ObjectRef(16310a0f0a45af5cffffffffffffffffffffffff0100000001000000), ObjectRef(c2668a65bda616c1ffffffffffffffffffffffff01000000010...\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Creating a Ray Dataset from results...\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2025-06-23 15:44:53,181\tINFO dataset.py:3046 -- Tip: Use `take_batch()` instead of `take() / show()` to return records in pandas or numpy batch format.\n",
-      "2025-06-23 15:44:53,357\tINFO logging.py:295 -- Registered dataset logger for dataset dataset_1_0\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "Dataset schema:\n",
-      "\n",
-      "First 5 rows:\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2025-06-23 15:44:53,393\tINFO streaming_executor.py:117 -- Starting execution of Dataset dataset_1_0. Full logs are in /tmp/ray/session_2025-06-23_15-44-44_311677_29614/logs/ray-data\n",
-      "2025-06-23 15:44:53,395\tINFO streaming_executor.py:118 -- Execution plan of Dataset dataset_1_0: InputDataBuffer[Input] -> LimitOperator[limit=5]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[dataset]: Run `pip install tqdm` to enable progress reporting.\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2025-06-23 15:44:53,647\tINFO streaming_executor.py:227 -- ✔️  Dataset dataset_1_0 execution finished in 0.25 seconds\n",
-      "2025-06-23 15:44:53,798\tWARNING plan.py:472 -- Warning: The Ray cluster currently does not have any available CPUs. The Dataset job will hang unless more CPUs are freed up. A common reason is that cluster resources are used by Actors or Tune trials; see the following link for more details: https://docs.ray.io/en/latest/data/data-internals.html#ray-data-and-tune\n",
-      "2025-06-23 15:44:53,799\tINFO logging.py:295 -- Registered dataset logger for dataset dataset_4_0\n",
-      "2025-06-23 15:44:53,806\tINFO streaming_executor.py:117 -- Starting execution of Dataset dataset_4_0. Full logs are in /tmp/ray/session_2025-06-23_15-44-44_311677_29614/logs/ray-data\n",
-      "2025-06-23 15:44:53,809\tINFO streaming_executor.py:118 -- Execution plan of Dataset dataset_4_0: InputDataBuffer[Input] -> AllToAllOperator[Repartition] -> TaskPoolMapOperator[Write]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{'item': ObjectRef(c8ef45ccd0112571ffffffffffffffffffffffff0100000001000000)}\n",
-      "{'item': ObjectRef(16310a0f0a45af5cffffffffffffffffffffffff0100000001000000)}\n",
-      "{'item': ObjectRef(c2668a65bda616c1ffffffffffffffffffffffff0100000001000000)}\n",
-      "{'item': ObjectRef(32d950ec0ccf9d2affffffffffffffffffffffff0100000001000000)}\n",
-      "{'item': ObjectRef(e0dc174c83599034ffffffffffffffffffffffff0100000001000000)}\n",
-      "Repartitioning dataset to control output file size...\n",
-      "\n",
-      "Writing dataset to Parquet format at: /mnt/d/dataset/mssv_preprocessed_duration\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2025-06-23 15:44:55,806\tINFO streaming_executor.py:227 -- ✔️  Dataset dataset_4_0 execution finished in 2.00 seconds\n",
-      "2025-06-23 15:45:01,968\tINFO dataset.py:4601 -- Data sink Parquet finished. 4205 rows and 839.9KB data written.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "Processing complete! Your dataset is ready.\n",
-      "\u001b[36m(process_midi_to_dict pid=30232)\u001b[0m Error processing /mnt/d/dataset/004.다화자 가창 데이터/01.데이터/1.Training/라벨링데이터/01.발라드R&B/A. 남성/03. 40대 이상/가창자_s13/ba_06120_-2_a_s13_m_04.mid: 'quantized duration'\u001b[32m [repeated 344x across cluster]\u001b[0m\n"
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:32:29,958 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.749 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:32:39,971 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.749 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:32:49,985 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.749 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:32:59,997 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:33:10,008 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:33:20,020 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:33:30,034 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:33:40,048 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:33:50,061 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:34:00,074 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n",
+      "\u001b[33m(raylet)\u001b[0m [2025-06-26 14:34:10,087 E 2007289 2007315] (raylet) file_system_monitor.cc:116: /tmp/ray/session_2025-06-26_14-31-47_408952_2006798 is over 95% full, available space: 108.748 GB; capacity: 3651.19 GB. Object creation will fail if spilling is required.\n"
      ]
     }
    ],
    "source": [
-    "mssv.preprocess_dataset(mssv_path, mssv_preprocessed_path)"
+    "preprocess.preprocess_dataset(mssv_path, mssv_preprocessed_path)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "id": "2d4c561c",
    "metadata": {},
    "outputs": [
@@ -3140,19 +3029,19 @@
        "  'duration': 480}]"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "mssv_sample_list, ticks_per_beat = mssv.midi_to_note_list(mssv_sample_midi)\n",
+    "mssv_sample_list, ticks_per_beat = preprocess.midi_to_note_list(mssv_sample_midi)\n",
     "mssv_sample_list "
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "0bac9307",
    "metadata": {},
    "outputs": [
@@ -3313,13 +3202,13 @@
     }
    ],
    "source": [
-    "df = mssv.preprocess_notes(mssv_sample_list, ticks_per_beat=ticks_per_beat)\n",
+    "df = preprocess.preprocess_notes(mssv_sample_list, ticks_per_beat=ticks_per_beat)\n",
     "df"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "id": "c0b7965b",
    "metadata": {},
    "outputs": [
@@ -3335,7 +3224,7 @@
     }
    ],
    "source": [
-    "singer_id = mssv.singer_id_from_filepath(mssv_sample_midi)\n",
+    "singer_id = preprocess.singer_id_from_filepath(mssv_sample_midi)\n",
     "singer_id "
    ]
   },
@@ -3349,7 +3238,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 6,
    "id": "0e0ef1e4",
    "metadata": {},
    "outputs": [],
@@ -3360,7 +3249,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 7,
    "id": "c519f235",
    "metadata": {},
    "outputs": [
@@ -3373,7 +3262,7 @@
        "})"
       ]
      },
-     "execution_count": 2,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3387,7 +3276,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 8,
    "id": "6f62d4d5",
    "metadata": {},
    "outputs": [
@@ -3397,7 +3286,7 @@
        "(4204, 4204)"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3408,7 +3297,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 9,
    "id": "ff7ec906",
    "metadata": {},
    "outputs": [
@@ -3418,7 +3307,7 @@
        "(4204, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18}, 18)"
       ]
      },
-     "execution_count": 4,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3429,7 +3318,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 10,
    "id": "a625ec7c",
    "metadata": {},
    "outputs": [
@@ -3449,7 +3338,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 11,
    "id": "e22dc42e",
    "metadata": {},
    "outputs": [
@@ -3459,7 +3348,7 @@
        "True"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3470,7 +3359,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 12,
    "id": "aa41450d",
    "metadata": {},
    "outputs": [
@@ -3480,7 +3369,7 @@
        "(817, 817, 2)"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -3489,265 +3378,6 @@
     "len(train_dataset[1]['durations']), len(train_dataset[1]['quantized_durations']), train_dataset[1]['singer_id']"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "id": "cbcad2ca",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Chunking dataset into fixed-length sequences...\n"
-     ]
-    }
-   ],
-   "source": [
-    "# --- Step 2: Define the Preprocessing Function ---\n",
-    "# This function will take a batch of songs and chop their long duration lists\n",
-    "# into fixed-length chunks that the model can handle.\n",
-    "\n",
-    "SEQUENCE_LENGTH = 128 \n",
-    "\n",
-    "def chunk_examples(examples):\n",
-    "    # Dictionaries to hold the new, chunked data\n",
-    "    chunked_inputs = {\n",
-    "        'original_durations': [],\n",
-    "        'quantized_durations': [],\n",
-    "        'singer_id': [],\n",
-    "    }\n",
-    "    \n",
-    "    # Iterate through each song in the batch\n",
-    "    for i in range(len(examples[\"durations\"])):\n",
-    "        durs = examples[\"durations\"][i]\n",
-    "        q_durs = examples[\"quantized_durations\"][i]\n",
-    "        sid = examples[\"singer_id\"][i]\n",
-    "        \n",
-    "        # Chop the long lists into smaller, fixed-length chunks\n",
-    "        for j in range(0, len(durs) - SEQUENCE_LENGTH, SEQUENCE_LENGTH):\n",
-    "            chunked_inputs['original_durations'].append(durs[j : j + SEQUENCE_LENGTH])\n",
-    "            chunked_inputs['quantized_durations'].append(q_durs[j : j + SEQUENCE_LENGTH])\n",
-    "            chunked_inputs['singer_id'].append(sid)\n",
-    "            \n",
-    "    return chunked_inputs\n",
-    "\n",
-    "\n",
-    "# --- Step 3: Apply the Preprocessing ---\n",
-    "# We use .map() to apply our function to the entire dataset.\n",
-    "# `batched=True` sends multiple rows at a time to our function for efficiency.\n",
-    "# `remove_columns` gets rid of the old, variable-length columns.\n",
-    "print(\"Chunking dataset into fixed-length sequences...\")\n",
-    "processed_dataset = train_dataset.map(\n",
-    "    chunk_examples,\n",
-    "    batched=True,\n",
-    "    remove_columns=train_dataset.column_names\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "id": "93b51c05",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Dataset({\n",
-       "    features: ['quantized_durations', 'singer_id', 'original_durations'],\n",
-       "    num_rows: 17941\n",
-       "})"
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "processed_dataset"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "id": "8d312061",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "17941"
-      ]
-     },
-     "execution_count": 10,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "len(processed_dataset['quantized_durations'])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "id": "a38b02c3",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "(128, 128, 2)"
-      ]
-     },
-     "execution_count": 11,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "len(processed_dataset['original_durations'][0]), len(processed_dataset['quantized_durations'][0]), processed_dataset['singer_id'][0]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "id": "07583462",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "45ae421c3aa0414386598a1213673229",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Map:   0%|          | 0/17941 [00:00<?, ? examples/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "DataLoader for Control Model (original -> original) created.\n",
-      "DataLoader for Your Method's Model (quantized -> original) created.\n"
-     ]
-    }
-   ],
-   "source": [
-    "# --- Step 4: Set Format and Create DataLoaders ---\n",
-    "# Now we format the dataset to output PyTorch tensors and create the DataLoaders\n",
-    "# which will handle batching and shuffling for us during training.\n",
-    "\n",
-    "BATCH_SIZE = 32\n",
-    "\n",
-    "# --- Dataloader for Model A (Control) ---\n",
-    "# Goal: original -> original\n",
-    "# We need to duplicate the 'original_durations' column to use it for both input and labels.\n",
-    "model_A_dataset = processed_dataset.map(\n",
-    "    lambda batch: {\"labels\": batch[\"original_durations\"]},\n",
-    "    batched=True\n",
-    ")\n",
-    "# Now we have: 'original_durations', 'quantized_durations', 'singer_id', 'labels'\n",
-    "\n",
-    "# Rename 'original_durations' to 'input_ids' for the model's input\n",
-    "model_A_dataset = model_A_dataset.rename_column(\"original_durations\", \"input_ids\")\n",
-    "model_A_dataset.set_format(type='torch', columns=['input_ids', 'labels', 'singer_id'])\n",
-    "model_A_dataloader = DataLoader(model_A_dataset, batch_size=BATCH_SIZE, shuffle=True)\n",
-    "print(\"DataLoader for Control Model (original -> original) created.\")\n",
-    "\n",
-    "\n",
-    "# --- Dataloader for Model B (Your Method) ---\n",
-    "# Goal: quantized -> original\n",
-    "model_B_dataset = processed_dataset.map(\n",
-    "    lambda batch: {\"labels\": batch[\"original_durations\"]},\n",
-    "    batched=True\n",
-    ")\n",
-    "\n",
-    "# Rename 'quantized_durations' to 'input_ids' for the model's input\n",
-    "model_B_dataset = model_B_dataset.rename_column(\"quantized_durations\", \"input_ids\")\n",
-    "model_B_dataset.set_format(type='torch', columns=['input_ids', 'labels', 'singer_id'])\n",
-    "model_B_dataloader = DataLoader(model_B_dataset, batch_size=BATCH_SIZE, shuffle=True)\n",
-    "print(\"DataLoader for Your Method's Model (quantized -> original) created.\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "id": "65c78eb3",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "--- Verifying one batch for Model B (Your Method) ---\n",
-      "Shape of the input tensor: torch.Size([32, 128])\n",
-      "Shape of the label tensor: torch.Size([32, 128])\n",
-      "Shape of the singer_id tensor: torch.Size([32])\n",
-      "\n",
-      "First input sequence in the batch (quantized):\n",
-      "tensor([ 240,   60,  360,   60,  300,    0,  780, 1560,  300,    0,  480,    0,\n",
-      "         540,  240,  240,   60,  420,    0,  420,    0,  420,  300,  300,    0,\n",
-      "         180,   60,  300,   60,  540,    0,  360,    0,  180,   60, 1080,  720,\n",
-      "         540,   60,  300,    0,  300,    0,  180,    0,  180,    0,  360,   60,\n",
-      "         540,    0,  300,    0,  420,    0,  300,  360,  540,   60,  360,    0,\n",
-      "         300,    0,  180,   60,  180,   60,  240,    0, 1680,  180,  480,    0,\n",
-      "         420,   60,  360,    0,  480,    0, 1860,  300,  360,   60,  240,    0,\n",
-      "         480,   60,  300,    0,  300,   60, 1260, 1020,  240,  120,  360,   60,\n",
-      "         240,    0,  240,    0,  240,    0,  300,    0,  300,    0,  240,    0,\n",
-      "         480,  120,  300,    0,  240,   60, 2160, 1440,  360,    0,  240,   60,\n",
-      "         360,   60,  480,    0,  300,    0, 1140,  780])\n",
-      "\n",
-      "Corresponding label sequence in the batch (original):\n",
-      "tensor([ 263,   40,  380,   33,  292,   14,  773, 1553,  278,   15,  483,   19,\n",
-      "         511,  253,  264,   40,  419,   13,  398,   15,  399,  278,  279,   13,\n",
-      "         198,   33,  277,   34,  539,   13,  347,    1,  205,   39, 1097,  712,\n",
-      "         564,   40,  293,   11,  271,   12,  178,   15,  177,   12,  375,   73,\n",
-      "         539,   13,  307,   24,  397,   16,  318,  353,  560,   34,  371,   24,\n",
-      "         289,    4,  199,   40,  205,   40,  237,   13, 1680,  172,  466,    3,\n",
-      "         433,   33,  358,   13,  458,   15, 1880,  312,  384,   42,  226,    0,\n",
-      "         457,   53,  280,   15,  325,   37, 1234, 1012,  265,  122,  385,   39,\n",
-      "         221,   12,  224,   15,  250,   25,  277,   16,  270,   14,  219,   14,\n",
-      "         473,   92,  278,   16,  256,   33, 2135, 1413,  337,   16,  267,   43,\n",
-      "         346,   74,  472,    7,  286,    4, 1123,  779])\n",
-      "\n",
-      "Corresponding singer_id:\n",
-      "tensor(4)\n"
-     ]
-    }
-   ],
-   "source": [
-    "# --- Step 5: Verify One Batch ---\n",
-    "# Let's pull one batch from the dataloader for Model B to see what it looks like.\n",
-    "# This is the data in its final form, \"right before being input into model\".\n",
-    "\n",
-    "print(\"\\n--- Verifying one batch for Model B (Your Method) ---\")\n",
-    "one_batch = next(iter(model_B_dataloader))\n",
-    "\n",
-    "# The dataloader gives us a dictionary of tensors\n",
-    "input_tensor = one_batch['input_ids']\n",
-    "label_tensor = one_batch['labels']\n",
-    "sid_tensor = one_batch['singer_id']\n",
-    "\n",
-    "print(f\"Shape of the input tensor: {input_tensor.shape}\")\n",
-    "print(f\"Shape of the label tensor: {label_tensor.shape}\")\n",
-    "print(f\"Shape of the singer_id tensor: {sid_tensor.shape}\")\n",
-    "print(\"\\nFirst input sequence in the batch (quantized):\")\n",
-    "print(input_tensor[0])\n",
-    "print(\"\\nCorresponding label sequence in the batch (original):\")\n",
-    "print(label_tensor[0])\n",
-    "print(\"\\nCorresponding singer_id:\")\n",
-    "print(sid_tensor[0])"
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "3828f8dd",
@@ -3763,7 +3393,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import toy_duration_predictor.train as tt"
+    "import toy_duration_predictor as tdp"
    ]
   },
   {
@@ -3779,83 +3409,15 @@
       "\n",
       "==================== FINAL MODEL COMPARISON ====================\n",
       "Using device for evaluation: cuda:0\n",
-      "--- Preparing data for Model B ---\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3007639f434040c8aa28cef9039d10c1",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Map:   0%|          | 0/4204 [00:00<?, ? examples/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Calculated training set stats: Mean=128.66, Std=201.76\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "036eb8eb8a8f42ca9fc4223114c47566",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Map:   0%|          | 0/3363 [00:00<?, ? examples/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "418d0c402c41422db24d5f18dea4ecc4",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Map:   0%|          | 0/420 [00:00<?, ? examples/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "44c6f6c9620543f3baed8f09f66514d8",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Map:   0%|          | 0/421 [00:00<?, ? examples/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
+      "--- Preparing test data ---\n",
+      "Calculated training set stats: Mean=128.66, Std=201.76\n",
       "Evaluating models on the test set...\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "f989de925f684cefbd2e1bd6af6cc554",
+       "model_id": "de7de013364a40fa8d9af567b9140063",
        "version_major": 2,
        "version_minor": 0
       },
@@ -3879,7 +3441,7 @@
    "source": [
     "model_a_path = 'model_stable/model_A.pth'\n",
     "model_b_path = 'model_stable/model_B.pth'\n",
-    "tt.evaluate_and_compare(model_a_path,  model_b_path,  gpu_id=0)"
+    "tdp.evaluate_and_compare(model_a_path,  model_b_path,  gpu_id=0)"
    ]
   },
   {
@@ -3892,17 +3454,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "7a54b60d",
    "metadata": {},
    "outputs": [],
    "source": [
-    "import toy_duration_predictor.upload as up"
+    "import toy_duration_predictor as tdp"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "fa70b5bf",
    "metadata": {},
    "outputs": [
@@ -3989,7 +3551,7 @@
     }
    ],
    "source": [
-    "up.upload_models_to_hub()"
+    "tdp.upload_models_to_hub()"
    ]
   },
   {
@@ -4007,7 +3569,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from toy_duration_predictor.load import load_and_test"
+    "import toy_duration_predictor as tdp"
    ]
   },
   {
@@ -4036,7 +3598,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "71f3f4ce46d343768862a8daf9aa6413",
+       "model_id": "65b38f6fa7724d73b91fc332eda22893",
        "version_major": 2,
        "version_minor": 0
       },
@@ -4058,7 +3620,7 @@
     }
    ],
    "source": [
-    "load_and_test()"
+    "tdp.load_and_test()"
    ]
   },
   {