{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "afe0f590",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T05:10:13.966028Z",
     "start_time": "2023-08-31T05:10:13.963957Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "a62a5a27",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T05:10:14.604311Z",
     "start_time": "2023-08-31T05:10:14.595363Z"
    }
   },
   "outputs": [],
   "source": [
    "import random\n",
    "import json\n",
    "import numpy as np\n",
    "import tqdm\n",
    "import torch\n",
    "import funcy\n",
    "import time\n",
    "import gc\n",
    "import tempfile\n",
    "import collections\n",
    "from joblib import Parallel, delayed\n",
    "\n",
    "from suno_utils.audio import Audio\n",
    "from suno_utils.tasks.data_loader import load_audio_mp\n",
    "from suno_utils.utils.text import write_jsonl, read_jsonl, write_json, read_json\n",
    "from suno_utils.utils.s3 import read_from_s3, check_s3_file_exists, open_from_s3\n",
    "from suno_utils.audio.conversion import convert_audio_files\n",
    "\n",
    "SAMPLE_RATE = 24_000\n",
    "EMBEDDING_RATE = 25\n",
    "N_CODEBOOKS = 8\n",
    "\n",
    "OUT_DATA_DIR = \"/app/suno/data/mert_25hz\"\n",
    "\n",
    "OUT_AUDIO_DIR = os.path.join(OUT_DATA_DIR, \"audio\")\n",
    "OUT_TSV_DIR = os.path.join(OUT_DATA_DIR, \"audio_tsv\")\n",
    "OUT_LABEL_DIR = os.path.join(OUT_DATA_DIR, \"label\")\n",
    "\n",
    "os.makedirs(OUT_DATA_DIR, exist_ok=True)\n",
    "os.makedirs(OUT_AUDIO_DIR, exist_ok=True)\n",
    "os.makedirs(OUT_TSV_DIR, exist_ok=True)\n",
    "os.makedirs(OUT_LABEL_DIR, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "49ff7b9a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T05:10:24.795358Z",
     "start_time": "2023-08-31T05:10:15.079288Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-31 05:10:15 | INFO | botocore.credentials | Found credentials in shared credentials file: ~/.aws/credentials\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25.0k hours of music\n"
     ]
    }
   ],
   "source": [
    "def load_metas_simple(filepath):\n",
    "    assert(filepath.startswith(\"s3://\"))\n",
    "    data = []\n",
    "    with open_from_s3(filepath) as f:\n",
    "        for line in f:\n",
    "            line = line.strip()\n",
    "            if len(line) == 0:\n",
    "                continue\n",
    "            m = json.loads(line)\n",
    "            _id = m[\"id\"]\n",
    "            duration_s = m[\"duration_s\"]\n",
    "            filepath = m.get(\"s3_filepath\", m.get(\"audio_filepath\", m.get(\"filepath\")))\n",
    "            assert(filepath is not None)\n",
    "            data.append({\n",
    "                \"id\": _id,\n",
    "                \"filepath\": filepath,\n",
    "                \"duration_s\": duration_s,\n",
    "            })\n",
    "    return data\n",
    "\n",
    "metas = load_metas_simple(\"s3://suno-data/datasets/bundles/v2/music_sample/metas.jsonl\")\n",
    "print(f\"{sum([m['duration_s'] for m in metas])/60/60/1e3:.1f}k hours of music\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a4b2e568",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.028821Z",
     "start_time": "2023-08-30T19:57:11.028811Z"
    }
   },
   "outputs": [],
   "source": [
    "metas[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6943cab3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.029757Z",
     "start_time": "2023-08-30T19:57:11.029749Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "CHUNKSIZE = 500\n",
    "MIN_DURATION_S = 5\n",
    "MAX_DURATION_S = 8*60\n",
    "\n",
    "# set up files\n",
    "for dset_type in (\"train\", \"valid\"):\n",
    "    with open(os.path.join(OUT_TSV_DIR, f\"{dset_type}.tsv\"), \"w\") as f:\n",
    "        f.write(OUT_AUDIO_DIR + \"\\n\")\n",
    "    for codebook in range(N_CODEBOOKS):\n",
    "        with open(os.path.join(OUT_LABEL_DIR, f\"{dset_type}.codec_{codebook}\"), \"w\") as f:\n",
    "            f.write(\"\")\n",
    "\n",
    "n_items = 0\n",
    "tot_duration_h = {\"valid\": 0, \"train\": 0}\n",
    "for n_iter, metas_chunk in enumerate(funcy.chunks(CHUNKSIZE, metas)):\n",
    "    # convert to wav\n",
    "    t0 = time.time()\n",
    "    out_rel_filepaths = [f\"{n_iter*CHUNKSIZE+n}.wav\" for n in range(len(metas_chunk))]\n",
    "    confirmed_list = convert_audio_files(\n",
    "        [m[\"filepath\"] for m in metas_chunk],\n",
    "        [os.path.join(OUT_AUDIO_DIR, out_rel_fp) for out_rel_fp in out_rel_filepaths],\n",
    "        n_cores=32,\n",
    "        force_threads=False,\n",
    "        sample_rate=SAMPLE_RATE,\n",
    "        byte_width=2,\n",
    "        n_channels=1,\n",
    "        debug=False,\n",
    "    )\n",
    "    td_convert = int(round(time.time() - t0))\n",
    "    # collect codec tokens and keep only valid ones then write tsv entries and codec\n",
    "    t0 = time.time()\n",
    "    codec_archive_fp = f\"s3://suno-data/datasets/bundles/v2/music_sample/dac_25_8/part_{n_iter}.npz\"\n",
    "    codec_archive = read_from_s3(codec_archive_fp, read_f=np.load)\n",
    "    # keep only valid ones in terms of duration and has both and then write tsv entries and codec\n",
    "    tsv_files = {\n",
    "        \"valid\": open(os.path.join(OUT_TSV_DIR, \"valid.tsv\"), \"a\"),\n",
    "        \"train\": open(os.path.join(OUT_TSV_DIR, \"train.tsv\"), \"a\")\n",
    "    }\n",
    "    codec_files = {\n",
    "        codebook: {\n",
    "            \"valid\": open(os.path.join(OUT_LABEL_DIR, f\"valid.codec_{codebook}\"), \"a\"),\n",
    "            \"train\": open(os.path.join(OUT_LABEL_DIR, f\"train.codec_{codebook}\"), \"a\")\n",
    "        } for codebook in range(N_CODEBOOKS)\n",
    "    }\n",
    "    chunk_duration_h = 0\n",
    "    n_segment = 0\n",
    "    for m, wav_fp, conv_success in zip(metas_chunk, out_rel_filepaths, confirmed_list):\n",
    "        if m[\"id\"] not in codec_archive or not conv_success:\n",
    "            continue\n",
    "        duration_s = Audio.get_duration_s(os.path.join(OUT_AUDIO_DIR, wav_fp))\n",
    "        codec_arr = codec_archive[m[\"id\"]]\n",
    "        codec_duration_s = codec_arr.shape[0] / EMBEDDING_RATE\n",
    "        if abs(duration_s - codec_duration_s) > 0.1:\n",
    "            continue\n",
    "        if duration_s < MIN_DURATION_S or duration_s > MAX_DURATION_S:\n",
    "            continue\n",
    "        dset_type = \"valid\" if n_segment % 20 == 0 else \"train\"\n",
    "        tsv_files[dset_type].write(wav_fp + \"\\t\" + str(int(round(duration_s*SAMPLE_RATE))) + \"\\n\")\n",
    "        for codebook in range(N_CODEBOOKS):\n",
    "            codec_files[codebook][dset_type].write(\" \".join(map(str, codec_arr[:,codebook])) + \"\\n\")\n",
    "        tot_duration_h[dset_type] += duration_s / 60 / 60\n",
    "        chunk_duration_h += duration_s / 60 / 60\n",
    "        n_segment += 1\n",
    "    for f in tsv_files.values():\n",
    "        f.close()\n",
    "    for codebook_files in codec_files.values():\n",
    "        for f in codebook_files.values():\n",
    "            f.close()\n",
    "    td_write = int(round(time.time() - t0))\n",
    "    print(\n",
    "        f\"{chunk_duration_h:,.1f} hours of data converted in {td_convert:,.1f}s\"\n",
    "        f\" and written in {td_write:,.1f}s\"\n",
    "    )\n",
    "    time.sleep(5) # make sure things close\n",
    "\n",
    "# total should take ~15h for prep\n",
    "# 30.4 hours of data converted in 62.0s and written in 33.0s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e0eeb87b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.030530Z",
     "start_time": "2023-08-30T19:57:11.030521Z"
    }
   },
   "outputs": [],
   "source": [
    "# only needed once\n",
    "for codebook in range(N_CODEBOOKS):\n",
    "    with open(os.path.join(OUT_LABEL_DIR, f\"dict.codec_{codebook}.txt\"), \"w\") as f:\n",
    "        for n in range(N_CODEBOOKS):\n",
    "            f.write(f\"{n} 1\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6b8bcd2d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.031049Z",
     "start_time": "2023-08-30T19:57:11.031041Z"
    }
   },
   "outputs": [],
   "source": [
    "# # (optional) overwrite train with valid\n",
    "# !cp /app/suno/data/mert_25hz/audio_tsv/valid.tsv /app/suno/data/mert_25hz/audio_tsv/train.tsv\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_0 /app/suno/data/mert_25hz/label/train.codec_0\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_1 /app/suno/data/mert_25hz/label/train.codec_1\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_2 /app/suno/data/mert_25hz/label/train.codec_2\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_3 /app/suno/data/mert_25hz/label/train.codec_3\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_4 /app/suno/data/mert_25hz/label/train.codec_4\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_5 /app/suno/data/mert_25hz/label/train.codec_5\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_6 /app/suno/data/mert_25hz/label/train.codec_6\n",
    "# !cp /app/suno/data/mert_25hz/label/valid.codec_7 /app/suno/data/mert_25hz/label/train.codec_7"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15eef138",
   "metadata": {},
   "source": [
    "## run training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b111cf17",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.031793Z",
     "start_time": "2023-08-30T19:57:11.031786Z"
    }
   },
   "outputs": [],
   "source": [
    "OMP_NUM_THREADS=24 python -u /home/georg/code/fairseq/fairseq_cli/hydra_train.py \\\n",
    "    --config-dir '/home/tony/Work/MERT/mert_fairseq/config/pretrain' \\\n",
    "    --config-name 'custom_95M' \\\n",
    "    common.user_dir='/home/tony/Work/MERT/mert_fairseq' \\\n",
    "    common.wandb_project='mert_test' \\\n",
    "    checkpoint.save_dir='/app/suno/checkpoints/mert_25hz_tony_try4' \\\n",
    "    distributed_training.distributed_rank=0 \\\n",
    "    distributed_training.distributed_world_size=4  \\\n",
    "    distributed_training.nprocs_per_node=4 \\\n",
    "    distributed_training.distributed_init_method=\"tcp://127.0.0.1:39683\" \\\n",
    "    task.data='/app/suno/data/mert_25hz/audio_tsv' \\\n",
    "    task.label_dir='/app/suno/data/mert_25hz/label' \\\n",
    "    task.labels='[\"codec_0\",\"codec_1\",\"codec_2\",\"codec_3\",\"codec_4\",\"codec_5\",\"codec_6\",\"codec_7\"]' \\\n",
    "    dataset.num_workers=24 \\\n",
    "    dataset.max_tokens=1800000 \\\n",
    "    dataset.disable_validation=false \\\n",
    "    model.label_rate=25\n",
    "\n",
    "# optimization.update_freq='[1]' \\\n",
    "# sample dataset: /app/suno/data/mert_25hz_test\n",
    "# max_tokens defines batch_size so might need to be lower?\n",
    "# need to change below because 4 instead of 8 GPUS?\n",
    "#     optimization.max_update=400000 \\\n",
    "#     lr_scheduler.warmup_updates=32000 \\\n",
    "# could also just do\n",
    "#     optimization.update_freq='[2]' \\"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "27a80368",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.032431Z",
     "start_time": "2023-08-30T19:57:11.032423Z"
    }
   },
   "outputs": [],
   "source": [
    "# ~7h for warmup steps -> 4x24h for full train"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ea023c5",
   "metadata": {},
   "source": [
    "## Playground"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "37d21a6a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.033161Z",
     "start_time": "2023-08-30T19:57:11.033154Z"
    }
   },
   "outputs": [],
   "source": [
    "MIN_DURATION_S = 3\n",
    "MAX_DURATION_S = 8*60\n",
    "\n",
    "def _convert_float_audio(sig):\n",
    "    dtype = np.int16\n",
    "    dtype_info = np.iinfo(dtype)\n",
    "    abs_max = 2 ** (dtype_info.bits - 1)\n",
    "    offset = dtype_info.min + abs_max\n",
    "    return (sig * abs_max + offset).clip(dtype_info.min, dtype_info.max).astype(dtype)\n",
    "\n",
    "out_mm_filepath_tr = os.path.join(OUT_DATA_DIR, f\"music_24khz_tr.bin\")\n",
    "out_mm_filepath_val = os.path.join(OUT_DATA_DIR, f\"music_24khz_val.bin\")\n",
    "out_mm_tr = np.memmap(out_mm_filepath_tr, dtype=np.int16, mode=\"w+\", shape=(1,))\n",
    "out_mm_val = np.memmap(out_mm_filepath_val, dtype=np.int16, mode=\"w+\", shape=(1,))\n",
    "n_offs_tr = 0\n",
    "n_offs_val = 0\n",
    "dset_duration_h = 0\n",
    "for n_iter, metas_chunk in enumerate(funcy.chunks(5000, metas)):\n",
    "#     t0 = time.time()\n",
    "#     audio_arr_list = load_audio_mp(\n",
    "#         [m[\"filepath\"] for m in metas_chunk],\n",
    "#         target_sample_rate=SAMPLE_RATE,\n",
    "#         min_duration_s=MIN_DURATION_S,\n",
    "#         max_duration_s=MAX_DURATION_S,\n",
    "#         num_workers=32,\n",
    "#         force_threads=False,\n",
    "#     #     debug=False,\n",
    "#         silent=True,\n",
    "#     )\n",
    "#     audio_arr_list = [_convert_float_audio(arr.numpy()[0]) for arr in audio_arr_list]\n",
    "#     td_fetch = int(round(time.time() - t0))\n",
    "#     chunk_duration_h = round(sum(arr.shape[-1] / SAMPLE_RATE for arr in audio_arr_list) / 60 / 60, 1)\n",
    "#     t0 = time.time()\n",
    "#     to_write_arr = np.concatenate(audio_arr_list, axis=0)\n",
    "#     to_write_len = to_write_arr.shape[-1]\n",
    "#     if n_iter == 0:\n",
    "#         out_mm = np.memmap(\n",
    "#             out_mm_filepath_val, dtype=np.int16, mode=\"r+\", shape=(n_offs_val+to_write_len,)\n",
    "#         )\n",
    "#         out_mm[n_offs_val:n_offs_val+to_write_len] = to_write_arr\n",
    "#         n_offs_val += to_write_len\n",
    "#         dset_type = \"val\"\n",
    "#     else:\n",
    "#         out_mm = np.memmap(\n",
    "#             out_mm_filepath_tr, dtype=np.int16, mode=\"r+\", shape=(n_offs_tr+to_write_len,)\n",
    "#         )\n",
    "#         out_mm[n_offs_tr:n_offs_tr+to_write_len] = to_write_arr\n",
    "#         n_offs_tr += to_write_len\n",
    "#         dset_type = \"tr\"\n",
    "#     out_mm.flush()\n",
    "#     del audio_arr_list, to_write_arr\n",
    "#     gc.collect();\n",
    "#     td_write = int(round(time.time() - t0))\n",
    "#     print(\n",
    "#         f\"{chunk_duration_h:,} hours of data fetched in {td_fetch:,}s\"\n",
    "#         f\" and written in {td_write:,}s as `{dset_type}`\"\n",
    "#     )\n",
    "#     dset_duration_h += chunk_duration_h\n",
    "#     time.sleep(5) # make sure things close\n",
    "# print(f\"done. collected total of {dset_duration_h:,.1f} hours\")\n",
    "# # should be ~4h for 5k hours\n",
    "# # 270.6 hours of data fetched in 373s and written in 214s as `tr`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cd111fdf",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.033997Z",
     "start_time": "2023-08-30T19:57:11.033989Z"
    }
   },
   "outputs": [],
   "source": [
    "# # test output\n",
    "# mm = np.memmap(\"/mnt/data/georg/data/raw_audio/music_24khz_val.bin\", dtype=np.int16, mode=\"r\")\n",
    "# idx = random.randint(0, len(mm)-1-24_000*10)\n",
    "# Audio.from_array(mm[idx:idx+24_000*10], sample_rate=24_000).play()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "141f9ab4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.034690Z",
     "start_time": "2023-08-30T19:57:11.034682Z"
    }
   },
   "outputs": [],
   "source": [
    "# !du -hs /mnt/data/georg/data/raw_audio/music_24khz_tr.bin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "47041ed8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.035384Z",
     "start_time": "2023-08-30T19:57:11.035375Z"
    }
   },
   "outputs": [],
   "source": [
    "# !aws s3 cp /mnt/data/georg/data/raw_audio/music_24khz_val.bin s3://suno-data/georg/data/raw_audio/\n",
    "# !aws s3 cp /mnt/data/georg/data/raw_audio/music_24khz_tr.bin s3://suno-data/georg/data/raw_audio/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3e67e882",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "9e0e0293",
   "metadata": {},
   "source": [
    "# some tests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "43ea4eeb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T13:54:00.397456Z",
     "start_time": "2023-08-31T13:54:00.395495Z"
    }
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6cfc261d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T13:54:01.022224Z",
     "start_time": "2023-08-31T13:54:01.019209Z"
    }
   },
   "outputs": [],
   "source": [
    "def clean_k(k):\n",
    "    \"\"\"Convert FairSeq Model keys to HF Model keys.\"\"\"\n",
    "    k = k.replace(\"mask_emb\", \"masked_spec_embed\")\n",
    "    k = k.replace(\".self_attn.\", \".attention.\")\n",
    "    k = k.replace(\".fc1.\", \".feed_forward.intermediate_dense.\")\n",
    "    k = k.replace(\".fc2.\", \".feed_forward.output_dense.\")\n",
    "    k = k.replace(\".self_attn_layer_norm.\", \".layer_norm.\")\n",
    "    for i in range(0, 7):\n",
    "        k = k.replace(f\"conv_layers.{i}.0\", \"conv_layers.{i}.conv\")\n",
    "    k = k.replace(\"encoder.pos_conv.0.\", \"encoder.pos_conv_embed.conv.\")\n",
    "    k = k.replace(\"post_extract_proj.\", \"feature_projection.projection.\")\n",
    "    if k == \"layer_norm.weight\":\n",
    "        k = \"feature_projection.layer_norm.weight\"\n",
    "    if k == \"layer_norm.bias\":\n",
    "        k = \"feature_projection.layer_norm.bias\"\n",
    "    k = k.replace(\"feature_extractor.conv_layers.0.2.\", \"feature_extractor.conv_layers.0.layer_norm.\")\n",
    "    return k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c5de053c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T13:54:13.554840Z",
     "start_time": "2023-08-31T13:54:01.690888Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> encoder.layer_norm.bias\n",
      "--> encoder.layer_norm.weight\n",
      "fair: encoder.layers.0.fc1.bias hf: encoder.layers.0.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.0.fc1.weight hf: encoder.layers.0.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.0.fc2.bias hf: encoder.layers.0.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.0.fc2.weight hf: encoder.layers.0.feed_forward.output_dense.weight\n",
      "--> encoder.layers.0.final_layer_norm.bias\n",
      "--> encoder.layers.0.final_layer_norm.weight\n",
      "fair: encoder.layers.0.self_attn.k_proj.bias hf: encoder.layers.0.attention.k_proj.bias\n",
      "fair: encoder.layers.0.self_attn.k_proj.weight hf: encoder.layers.0.attention.k_proj.weight\n",
      "fair: encoder.layers.0.self_attn.out_proj.bias hf: encoder.layers.0.attention.out_proj.bias\n",
      "fair: encoder.layers.0.self_attn.out_proj.weight hf: encoder.layers.0.attention.out_proj.weight\n",
      "fair: encoder.layers.0.self_attn.q_proj.bias hf: encoder.layers.0.attention.q_proj.bias\n",
      "fair: encoder.layers.0.self_attn.q_proj.weight hf: encoder.layers.0.attention.q_proj.weight\n",
      "fair: encoder.layers.0.self_attn.v_proj.bias hf: encoder.layers.0.attention.v_proj.bias\n",
      "fair: encoder.layers.0.self_attn.v_proj.weight hf: encoder.layers.0.attention.v_proj.weight\n",
      "fair: encoder.layers.0.self_attn_layer_norm.bias hf: encoder.layers.0.layer_norm.bias\n",
      "fair: encoder.layers.0.self_attn_layer_norm.weight hf: encoder.layers.0.layer_norm.weight\n",
      "fair: encoder.layers.1.fc1.bias hf: encoder.layers.1.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.1.fc1.weight hf: encoder.layers.1.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.1.fc2.bias hf: encoder.layers.1.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.1.fc2.weight hf: encoder.layers.1.feed_forward.output_dense.weight\n",
      "--> encoder.layers.1.final_layer_norm.bias\n",
      "--> encoder.layers.1.final_layer_norm.weight\n",
      "fair: encoder.layers.1.self_attn.k_proj.bias hf: encoder.layers.1.attention.k_proj.bias\n",
      "fair: encoder.layers.1.self_attn.k_proj.weight hf: encoder.layers.1.attention.k_proj.weight\n",
      "fair: encoder.layers.1.self_attn.out_proj.bias hf: encoder.layers.1.attention.out_proj.bias\n",
      "fair: encoder.layers.1.self_attn.out_proj.weight hf: encoder.layers.1.attention.out_proj.weight\n",
      "fair: encoder.layers.1.self_attn.q_proj.bias hf: encoder.layers.1.attention.q_proj.bias\n",
      "fair: encoder.layers.1.self_attn.q_proj.weight hf: encoder.layers.1.attention.q_proj.weight\n",
      "fair: encoder.layers.1.self_attn.v_proj.bias hf: encoder.layers.1.attention.v_proj.bias\n",
      "fair: encoder.layers.1.self_attn.v_proj.weight hf: encoder.layers.1.attention.v_proj.weight\n",
      "fair: encoder.layers.1.self_attn_layer_norm.bias hf: encoder.layers.1.layer_norm.bias\n",
      "fair: encoder.layers.1.self_attn_layer_norm.weight hf: encoder.layers.1.layer_norm.weight\n",
      "fair: encoder.layers.10.fc1.bias hf: encoder.layers.10.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.10.fc1.weight hf: encoder.layers.10.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.10.fc2.bias hf: encoder.layers.10.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.10.fc2.weight hf: encoder.layers.10.feed_forward.output_dense.weight\n",
      "--> encoder.layers.10.final_layer_norm.bias\n",
      "--> encoder.layers.10.final_layer_norm.weight\n",
      "fair: encoder.layers.10.self_attn.k_proj.bias hf: encoder.layers.10.attention.k_proj.bias\n",
      "fair: encoder.layers.10.self_attn.k_proj.weight hf: encoder.layers.10.attention.k_proj.weight\n",
      "fair: encoder.layers.10.self_attn.out_proj.bias hf: encoder.layers.10.attention.out_proj.bias\n",
      "fair: encoder.layers.10.self_attn.out_proj.weight hf: encoder.layers.10.attention.out_proj.weight\n",
      "fair: encoder.layers.10.self_attn.q_proj.bias hf: encoder.layers.10.attention.q_proj.bias\n",
      "fair: encoder.layers.10.self_attn.q_proj.weight hf: encoder.layers.10.attention.q_proj.weight\n",
      "fair: encoder.layers.10.self_attn.v_proj.bias hf: encoder.layers.10.attention.v_proj.bias\n",
      "fair: encoder.layers.10.self_attn.v_proj.weight hf: encoder.layers.10.attention.v_proj.weight\n",
      "fair: encoder.layers.10.self_attn_layer_norm.bias hf: encoder.layers.10.layer_norm.bias\n",
      "fair: encoder.layers.10.self_attn_layer_norm.weight hf: encoder.layers.10.layer_norm.weight\n",
      "fair: encoder.layers.11.fc1.bias hf: encoder.layers.11.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.11.fc1.weight hf: encoder.layers.11.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.11.fc2.bias hf: encoder.layers.11.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.11.fc2.weight hf: encoder.layers.11.feed_forward.output_dense.weight\n",
      "--> encoder.layers.11.final_layer_norm.bias\n",
      "--> encoder.layers.11.final_layer_norm.weight\n",
      "fair: encoder.layers.11.self_attn.k_proj.bias hf: encoder.layers.11.attention.k_proj.bias\n",
      "fair: encoder.layers.11.self_attn.k_proj.weight hf: encoder.layers.11.attention.k_proj.weight\n",
      "fair: encoder.layers.11.self_attn.out_proj.bias hf: encoder.layers.11.attention.out_proj.bias\n",
      "fair: encoder.layers.11.self_attn.out_proj.weight hf: encoder.layers.11.attention.out_proj.weight\n",
      "fair: encoder.layers.11.self_attn.q_proj.bias hf: encoder.layers.11.attention.q_proj.bias\n",
      "fair: encoder.layers.11.self_attn.q_proj.weight hf: encoder.layers.11.attention.q_proj.weight\n",
      "fair: encoder.layers.11.self_attn.v_proj.bias hf: encoder.layers.11.attention.v_proj.bias\n",
      "fair: encoder.layers.11.self_attn.v_proj.weight hf: encoder.layers.11.attention.v_proj.weight\n",
      "fair: encoder.layers.11.self_attn_layer_norm.bias hf: encoder.layers.11.layer_norm.bias\n",
      "fair: encoder.layers.11.self_attn_layer_norm.weight hf: encoder.layers.11.layer_norm.weight\n",
      "fair: encoder.layers.2.fc1.bias hf: encoder.layers.2.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.2.fc1.weight hf: encoder.layers.2.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.2.fc2.bias hf: encoder.layers.2.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.2.fc2.weight hf: encoder.layers.2.feed_forward.output_dense.weight\n",
      "--> encoder.layers.2.final_layer_norm.bias\n",
      "--> encoder.layers.2.final_layer_norm.weight\n",
      "fair: encoder.layers.2.self_attn.k_proj.bias hf: encoder.layers.2.attention.k_proj.bias\n",
      "fair: encoder.layers.2.self_attn.k_proj.weight hf: encoder.layers.2.attention.k_proj.weight\n",
      "fair: encoder.layers.2.self_attn.out_proj.bias hf: encoder.layers.2.attention.out_proj.bias\n",
      "fair: encoder.layers.2.self_attn.out_proj.weight hf: encoder.layers.2.attention.out_proj.weight\n",
      "fair: encoder.layers.2.self_attn.q_proj.bias hf: encoder.layers.2.attention.q_proj.bias\n",
      "fair: encoder.layers.2.self_attn.q_proj.weight hf: encoder.layers.2.attention.q_proj.weight\n",
      "fair: encoder.layers.2.self_attn.v_proj.bias hf: encoder.layers.2.attention.v_proj.bias\n",
      "fair: encoder.layers.2.self_attn.v_proj.weight hf: encoder.layers.2.attention.v_proj.weight\n",
      "fair: encoder.layers.2.self_attn_layer_norm.bias hf: encoder.layers.2.layer_norm.bias\n",
      "fair: encoder.layers.2.self_attn_layer_norm.weight hf: encoder.layers.2.layer_norm.weight\n",
      "fair: encoder.layers.3.fc1.bias hf: encoder.layers.3.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.3.fc1.weight hf: encoder.layers.3.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.3.fc2.bias hf: encoder.layers.3.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.3.fc2.weight hf: encoder.layers.3.feed_forward.output_dense.weight\n",
      "--> encoder.layers.3.final_layer_norm.bias\n",
      "--> encoder.layers.3.final_layer_norm.weight\n",
      "fair: encoder.layers.3.self_attn.k_proj.bias hf: encoder.layers.3.attention.k_proj.bias\n",
      "fair: encoder.layers.3.self_attn.k_proj.weight hf: encoder.layers.3.attention.k_proj.weight\n",
      "fair: encoder.layers.3.self_attn.out_proj.bias hf: encoder.layers.3.attention.out_proj.bias\n",
      "fair: encoder.layers.3.self_attn.out_proj.weight hf: encoder.layers.3.attention.out_proj.weight\n",
      "fair: encoder.layers.3.self_attn.q_proj.bias hf: encoder.layers.3.attention.q_proj.bias\n",
      "fair: encoder.layers.3.self_attn.q_proj.weight hf: encoder.layers.3.attention.q_proj.weight\n",
      "fair: encoder.layers.3.self_attn.v_proj.bias hf: encoder.layers.3.attention.v_proj.bias\n",
      "fair: encoder.layers.3.self_attn.v_proj.weight hf: encoder.layers.3.attention.v_proj.weight\n",
      "fair: encoder.layers.3.self_attn_layer_norm.bias hf: encoder.layers.3.layer_norm.bias\n",
      "fair: encoder.layers.3.self_attn_layer_norm.weight hf: encoder.layers.3.layer_norm.weight\n",
      "fair: encoder.layers.4.fc1.bias hf: encoder.layers.4.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.4.fc1.weight hf: encoder.layers.4.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.4.fc2.bias hf: encoder.layers.4.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.4.fc2.weight hf: encoder.layers.4.feed_forward.output_dense.weight\n",
      "--> encoder.layers.4.final_layer_norm.bias\n",
      "--> encoder.layers.4.final_layer_norm.weight\n",
      "fair: encoder.layers.4.self_attn.k_proj.bias hf: encoder.layers.4.attention.k_proj.bias\n",
      "fair: encoder.layers.4.self_attn.k_proj.weight hf: encoder.layers.4.attention.k_proj.weight\n",
      "fair: encoder.layers.4.self_attn.out_proj.bias hf: encoder.layers.4.attention.out_proj.bias\n",
      "fair: encoder.layers.4.self_attn.out_proj.weight hf: encoder.layers.4.attention.out_proj.weight\n",
      "fair: encoder.layers.4.self_attn.q_proj.bias hf: encoder.layers.4.attention.q_proj.bias\n",
      "fair: encoder.layers.4.self_attn.q_proj.weight hf: encoder.layers.4.attention.q_proj.weight\n",
      "fair: encoder.layers.4.self_attn.v_proj.bias hf: encoder.layers.4.attention.v_proj.bias\n",
      "fair: encoder.layers.4.self_attn.v_proj.weight hf: encoder.layers.4.attention.v_proj.weight\n",
      "fair: encoder.layers.4.self_attn_layer_norm.bias hf: encoder.layers.4.layer_norm.bias\n",
      "fair: encoder.layers.4.self_attn_layer_norm.weight hf: encoder.layers.4.layer_norm.weight\n",
      "fair: encoder.layers.5.fc1.bias hf: encoder.layers.5.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.5.fc1.weight hf: encoder.layers.5.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.5.fc2.bias hf: encoder.layers.5.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.5.fc2.weight hf: encoder.layers.5.feed_forward.output_dense.weight\n",
      "--> encoder.layers.5.final_layer_norm.bias\n",
      "--> encoder.layers.5.final_layer_norm.weight\n",
      "fair: encoder.layers.5.self_attn.k_proj.bias hf: encoder.layers.5.attention.k_proj.bias\n",
      "fair: encoder.layers.5.self_attn.k_proj.weight hf: encoder.layers.5.attention.k_proj.weight\n",
      "fair: encoder.layers.5.self_attn.out_proj.bias hf: encoder.layers.5.attention.out_proj.bias\n",
      "fair: encoder.layers.5.self_attn.out_proj.weight hf: encoder.layers.5.attention.out_proj.weight\n",
      "fair: encoder.layers.5.self_attn.q_proj.bias hf: encoder.layers.5.attention.q_proj.bias\n",
      "fair: encoder.layers.5.self_attn.q_proj.weight hf: encoder.layers.5.attention.q_proj.weight\n",
      "fair: encoder.layers.5.self_attn.v_proj.bias hf: encoder.layers.5.attention.v_proj.bias\n",
      "fair: encoder.layers.5.self_attn.v_proj.weight hf: encoder.layers.5.attention.v_proj.weight\n",
      "fair: encoder.layers.5.self_attn_layer_norm.bias hf: encoder.layers.5.layer_norm.bias\n",
      "fair: encoder.layers.5.self_attn_layer_norm.weight hf: encoder.layers.5.layer_norm.weight\n",
      "fair: encoder.layers.6.fc1.bias hf: encoder.layers.6.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.6.fc1.weight hf: encoder.layers.6.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.6.fc2.bias hf: encoder.layers.6.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.6.fc2.weight hf: encoder.layers.6.feed_forward.output_dense.weight\n",
      "--> encoder.layers.6.final_layer_norm.bias\n",
      "--> encoder.layers.6.final_layer_norm.weight\n",
      "fair: encoder.layers.6.self_attn.k_proj.bias hf: encoder.layers.6.attention.k_proj.bias\n",
      "fair: encoder.layers.6.self_attn.k_proj.weight hf: encoder.layers.6.attention.k_proj.weight\n",
      "fair: encoder.layers.6.self_attn.out_proj.bias hf: encoder.layers.6.attention.out_proj.bias\n",
      "fair: encoder.layers.6.self_attn.out_proj.weight hf: encoder.layers.6.attention.out_proj.weight\n",
      "fair: encoder.layers.6.self_attn.q_proj.bias hf: encoder.layers.6.attention.q_proj.bias\n",
      "fair: encoder.layers.6.self_attn.q_proj.weight hf: encoder.layers.6.attention.q_proj.weight\n",
      "fair: encoder.layers.6.self_attn.v_proj.bias hf: encoder.layers.6.attention.v_proj.bias\n",
      "fair: encoder.layers.6.self_attn.v_proj.weight hf: encoder.layers.6.attention.v_proj.weight\n",
      "fair: encoder.layers.6.self_attn_layer_norm.bias hf: encoder.layers.6.layer_norm.bias\n",
      "fair: encoder.layers.6.self_attn_layer_norm.weight hf: encoder.layers.6.layer_norm.weight\n",
      "fair: encoder.layers.7.fc1.bias hf: encoder.layers.7.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.7.fc1.weight hf: encoder.layers.7.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.7.fc2.bias hf: encoder.layers.7.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.7.fc2.weight hf: encoder.layers.7.feed_forward.output_dense.weight\n",
      "--> encoder.layers.7.final_layer_norm.bias\n",
      "--> encoder.layers.7.final_layer_norm.weight\n",
      "fair: encoder.layers.7.self_attn.k_proj.bias hf: encoder.layers.7.attention.k_proj.bias\n",
      "fair: encoder.layers.7.self_attn.k_proj.weight hf: encoder.layers.7.attention.k_proj.weight\n",
      "fair: encoder.layers.7.self_attn.out_proj.bias hf: encoder.layers.7.attention.out_proj.bias\n",
      "fair: encoder.layers.7.self_attn.out_proj.weight hf: encoder.layers.7.attention.out_proj.weight\n",
      "fair: encoder.layers.7.self_attn.q_proj.bias hf: encoder.layers.7.attention.q_proj.bias\n",
      "fair: encoder.layers.7.self_attn.q_proj.weight hf: encoder.layers.7.attention.q_proj.weight\n",
      "fair: encoder.layers.7.self_attn.v_proj.bias hf: encoder.layers.7.attention.v_proj.bias\n",
      "fair: encoder.layers.7.self_attn.v_proj.weight hf: encoder.layers.7.attention.v_proj.weight\n",
      "fair: encoder.layers.7.self_attn_layer_norm.bias hf: encoder.layers.7.layer_norm.bias\n",
      "fair: encoder.layers.7.self_attn_layer_norm.weight hf: encoder.layers.7.layer_norm.weight\n",
      "fair: encoder.layers.8.fc1.bias hf: encoder.layers.8.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.8.fc1.weight hf: encoder.layers.8.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.8.fc2.bias hf: encoder.layers.8.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.8.fc2.weight hf: encoder.layers.8.feed_forward.output_dense.weight\n",
      "--> encoder.layers.8.final_layer_norm.bias\n",
      "--> encoder.layers.8.final_layer_norm.weight\n",
      "fair: encoder.layers.8.self_attn.k_proj.bias hf: encoder.layers.8.attention.k_proj.bias\n",
      "fair: encoder.layers.8.self_attn.k_proj.weight hf: encoder.layers.8.attention.k_proj.weight\n",
      "fair: encoder.layers.8.self_attn.out_proj.bias hf: encoder.layers.8.attention.out_proj.bias\n",
      "fair: encoder.layers.8.self_attn.out_proj.weight hf: encoder.layers.8.attention.out_proj.weight\n",
      "fair: encoder.layers.8.self_attn.q_proj.bias hf: encoder.layers.8.attention.q_proj.bias\n",
      "fair: encoder.layers.8.self_attn.q_proj.weight hf: encoder.layers.8.attention.q_proj.weight\n",
      "fair: encoder.layers.8.self_attn.v_proj.bias hf: encoder.layers.8.attention.v_proj.bias\n",
      "fair: encoder.layers.8.self_attn.v_proj.weight hf: encoder.layers.8.attention.v_proj.weight\n",
      "fair: encoder.layers.8.self_attn_layer_norm.bias hf: encoder.layers.8.layer_norm.bias\n",
      "fair: encoder.layers.8.self_attn_layer_norm.weight hf: encoder.layers.8.layer_norm.weight\n",
      "fair: encoder.layers.9.fc1.bias hf: encoder.layers.9.feed_forward.intermediate_dense.bias\n",
      "fair: encoder.layers.9.fc1.weight hf: encoder.layers.9.feed_forward.intermediate_dense.weight\n",
      "fair: encoder.layers.9.fc2.bias hf: encoder.layers.9.feed_forward.output_dense.bias\n",
      "fair: encoder.layers.9.fc2.weight hf: encoder.layers.9.feed_forward.output_dense.weight\n",
      "--> encoder.layers.9.final_layer_norm.bias\n",
      "--> encoder.layers.9.final_layer_norm.weight\n",
      "fair: encoder.layers.9.self_attn.k_proj.bias hf: encoder.layers.9.attention.k_proj.bias\n",
      "fair: encoder.layers.9.self_attn.k_proj.weight hf: encoder.layers.9.attention.k_proj.weight\n",
      "fair: encoder.layers.9.self_attn.out_proj.bias hf: encoder.layers.9.attention.out_proj.bias\n",
      "fair: encoder.layers.9.self_attn.out_proj.weight hf: encoder.layers.9.attention.out_proj.weight\n",
      "fair: encoder.layers.9.self_attn.q_proj.bias hf: encoder.layers.9.attention.q_proj.bias\n",
      "fair: encoder.layers.9.self_attn.q_proj.weight hf: encoder.layers.9.attention.q_proj.weight\n",
      "fair: encoder.layers.9.self_attn.v_proj.bias hf: encoder.layers.9.attention.v_proj.bias\n",
      "fair: encoder.layers.9.self_attn.v_proj.weight hf: encoder.layers.9.attention.v_proj.weight\n",
      "fair: encoder.layers.9.self_attn_layer_norm.bias hf: encoder.layers.9.layer_norm.bias\n",
      "fair: encoder.layers.9.self_attn_layer_norm.weight hf: encoder.layers.9.layer_norm.weight\n",
      "fair: encoder.pos_conv.0.bias hf: encoder.pos_conv_embed.conv.bias\n",
      "fair: encoder.pos_conv.0.weight_g hf: encoder.pos_conv_embed.conv.weight_g\n",
      "fair: encoder.pos_conv.0.weight_v hf: encoder.pos_conv_embed.conv.weight_v\n",
      "--> encoder_cqt_model.fc.bias\n",
      "--> encoder_cqt_model.fc.weight\n",
      "--> encoder_cqt_model.spec_layer.cqt_kernels_imag\n",
      "--> encoder_cqt_model.spec_layer.cqt_kernels_real\n",
      "--> encoder_cqt_model.spec_layer.lenghts\n",
      "fair: feature_extractor.conv_layers.0.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.0.2.bias hf: feature_extractor.conv_layers.0.layer_norm.bias\n",
      "fair: feature_extractor.conv_layers.0.2.weight hf: feature_extractor.conv_layers.0.layer_norm.weight\n",
      "fair: feature_extractor.conv_layers.1.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.2.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.3.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.4.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.5.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "fair: feature_extractor.conv_layers.6.0.weight hf: feature_extractor.conv_layers.{i}.conv.weight\n",
      "--> feature_extractor.conv_layers.7.0.weight\n",
      "--> final_proj.bias\n",
      "--> final_proj.weight\n",
      "--> label_embs_concat\n",
      "fair: layer_norm.bias hf: feature_projection.layer_norm.bias\n",
      "fair: layer_norm.weight hf: feature_projection.layer_norm.weight\n",
      "fair: mask_emb hf: masked_spec_embed\n",
      "fair: post_extract_proj.bias hf: feature_projection.projection.bias\n",
      "fair: post_extract_proj.weight hf: feature_projection.projection.weight\n"
     ]
    }
   ],
   "source": [
    "# convert to HF checkpoint\n",
    "from_fp = \"/app/suno/checkpoints/mert_25hz/checkpoint_last.pt\"\n",
    "to_fp = \"/home/tony/Data/MERT/mert_test_25hz.pt\"\n",
    "sd = torch.load(\"/app/suno/checkpoints/mert_25hz/checkpoint_last.pt\", map_location=\"cpu\")\n",
    "new_sd = {}\n",
    "for k in sorted(sd[\"model\"].keys()):\n",
    "    v = sd[\"model\"][k]\n",
    "    orig_k = k\n",
    "    k = clean_k(k)\n",
    "    if k != orig_k:\n",
    "        print(\"fair:\", orig_k, \"hf:\", k)\n",
    "    else:\n",
    "        print(\"-->\", k)\n",
    "    new_sd[k] = v\n",
    "for k in [\n",
    "    \"label_embs_concat\", \"final_proj.bias\", \"final_proj.weight\", \"encoder_cqt_model.spec_layer.lenghts\", \n",
    "    \"encoder_cqt_model.spec_layer.cqt_kernels_real\", \"encoder_cqt_model.spec_layer.cqt_kernels_imag\", \n",
    "    \"encoder_cqt_model.fc.weight\", \"encoder_cqt_model.fc.bias\"\n",
    "]:\n",
    "    del new_sd[k]\n",
    "torch.save(new_sd, to_fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f1f7b1a4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T13:54:20.420167Z",
     "start_time": "2023-08-31T13:54:20.418552Z"
    }
   },
   "outputs": [],
   "source": [
    "# for k in sorted(sd[\"model\"].keys()):\n",
    "#     if \"conv_layers\" in k:\n",
    "#         print(\"fs:\", k, \"hf:\", clean_k(k), clean_k(k) in sd[\"model\"].keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c3ac0cc8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T13:54:22.027536Z",
     "start_time": "2023-08-31T13:54:22.025974Z"
    }
   },
   "outputs": [],
   "source": [
    "# from collections import Counter\n",
    "# c = Counter()\n",
    "# for k in sorted(sd[\"model\"].keys()):\n",
    "#     if \"layer_norm\" in k:\n",
    "#         print(\"fs:\", k, \"hf:\", clean_k(k), clean_k(k) in sd[\"model\"].keys())\n",
    "#         c[clean_k(k)] += 1\n",
    "#     elif \"layer_norm\" in clean_k(k):\n",
    "#         print(\" WTF fs:\", k, \"hf:\", clean_k(k), clean_k(k) in sd[\"model\"].keys())\n",
    "# for k in sorted(new_sd.keys()):\n",
    "#     if \"layer_norm\" in k and k not in c:\n",
    "#         print(\"-->hf:\", k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "89a6a3ae",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "a88ea965",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T21:43:31.146431Z",
     "start_time": "2023-08-30T21:43:28.333982Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import torch\n",
    "from transformers import Wav2Vec2FeatureExtractor\n",
    "from suno_utils.models.mert.modeling_MERT import MERTModel, MERTConfig\n",
    "from suno_utils.audio import Audio\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/preprocessor_config.json\") as f:\n",
    "    processor = Wav2Vec2FeatureExtractor(**json.load(f))\n",
    "\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/config_small_25hz.json\") as f:\n",
    "    cfg = MERTConfig(**json.load(f))\n",
    "model = MERTModel(cfg)\n",
    "sd = torch.load(\"/home/tony/Data/MERT/mert_test_25hz.pt\")\n",
    "model.load_state_dict(sd);\n",
    "model.eval();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1a333d01",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.038005Z",
     "start_time": "2023-08-30T19:57:11.037998Z"
    }
   },
   "outputs": [],
   "source": [
    "from transformers import AutoModel\n",
    "model_original = AutoModel.from_pretrained(\"m-a-p/MERT-v1-95M\", trust_remote_code=True)\n",
    "processor_original = Wav2Vec2FeatureExtractor.from_pretrained(\"m-a-p/MERT-v1-95M\",trust_remote_code=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d886b3e9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.038574Z",
     "start_time": "2023-08-30T19:57:11.038567Z"
    }
   },
   "outputs": [],
   "source": [
    "# validate_model(model, processor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "73f33cfc",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T19:57:11.039258Z",
     "start_time": "2023-08-30T19:57:11.039250Z"
    }
   },
   "outputs": [],
   "source": [
    "validate_model(model_original, processor_original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b5e3ca8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T18:59:43.633008Z",
     "start_time": "2023-08-30T18:59:43.606939Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4cb92308",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:15:31.806398Z",
     "start_time": "2023-08-30T20:15:31.802017Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c9340468",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:15:48.250872Z",
     "start_time": "2023-08-30T20:15:32.118550Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "load local MERT 25 HZ model succeed\n",
      "load local processor succeed\n"
     ]
    }
   ],
   "source": [
    "from suno_utils.tasks.mert_v2 import encode\n",
    "from suno_utils.audio import Audio\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "audio = Audio.from_file(\"audios/kitchen_sink_temperature_semantic_0.85_0.wav\").convert(24_000, 2, 1).get_segment(to_s=10)\n",
    "out = encode(audio, do_clustering=False)\n",
    "out_h = encode(audio.get_segment(to_s=5), do_clustering=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a011e2d6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:15:48.403959Z",
     "start_time": "2023-08-30T20:15:48.253143Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f615c444160>]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(out.T.mean(-1))\n",
    "plt.plot(out_h.T.mean(-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5971adeb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:16:38.670680Z",
     "start_time": "2023-08-30T20:16:36.477747Z"
    }
   },
   "outputs": [],
   "source": [
    "from transformers import AutoModel\n",
    "from transformers import Wav2Vec2FeatureExtractor\n",
    "import torch\n",
    "model_original = AutoModel.from_pretrained(\"m-a-p/MERT-v1-95M\", trust_remote_code=True, revision=\"8881df140a93e2e\")\n",
    "processor_original = Wav2Vec2FeatureExtractor.from_pretrained(\"m-a-p/MERT-v1-95M\",trust_remote_code=True, revision=\"8881df140a93e2e\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8c88ae93",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:41:49.407910Z",
     "start_time": "2023-08-30T20:41:49.402380Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "encoder.layer_norm.bias\n",
      "encoder.layer_norm.weight\n",
      "encoder.layers.0.attention.k_proj.bias\n",
      "encoder.layers.0.attention.k_proj.weight\n",
      "encoder.layers.0.attention.out_proj.bias\n",
      "encoder.layers.0.attention.out_proj.weight\n",
      "encoder.layers.0.attention.q_proj.bias\n",
      "encoder.layers.0.attention.q_proj.weight\n",
      "encoder.layers.0.attention.v_proj.bias\n",
      "encoder.layers.0.attention.v_proj.weight\n",
      "encoder.layers.0.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.0.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.0.feed_forward.output_dense.bias\n",
      "encoder.layers.0.feed_forward.output_dense.weight\n",
      "encoder.layers.0.final_layer_norm.bias\n",
      "encoder.layers.0.final_layer_norm.weight\n",
      "encoder.layers.0.layer_norm.bias\n",
      "encoder.layers.0.layer_norm.weight\n",
      "encoder.layers.1.attention.k_proj.bias\n",
      "encoder.layers.1.attention.k_proj.weight\n",
      "encoder.layers.1.attention.out_proj.bias\n",
      "encoder.layers.1.attention.out_proj.weight\n",
      "encoder.layers.1.attention.q_proj.bias\n",
      "encoder.layers.1.attention.q_proj.weight\n",
      "encoder.layers.1.attention.v_proj.bias\n",
      "encoder.layers.1.attention.v_proj.weight\n",
      "encoder.layers.1.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.1.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.1.feed_forward.output_dense.bias\n",
      "encoder.layers.1.feed_forward.output_dense.weight\n",
      "encoder.layers.1.final_layer_norm.bias\n",
      "encoder.layers.1.final_layer_norm.weight\n",
      "encoder.layers.1.layer_norm.bias\n",
      "encoder.layers.1.layer_norm.weight\n",
      "encoder.layers.10.attention.k_proj.bias\n",
      "encoder.layers.10.attention.k_proj.weight\n",
      "encoder.layers.10.attention.out_proj.bias\n",
      "encoder.layers.10.attention.out_proj.weight\n",
      "encoder.layers.10.attention.q_proj.bias\n",
      "encoder.layers.10.attention.q_proj.weight\n",
      "encoder.layers.10.attention.v_proj.bias\n",
      "encoder.layers.10.attention.v_proj.weight\n",
      "encoder.layers.10.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.10.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.10.feed_forward.output_dense.bias\n",
      "encoder.layers.10.feed_forward.output_dense.weight\n",
      "encoder.layers.10.final_layer_norm.bias\n",
      "encoder.layers.10.final_layer_norm.weight\n",
      "encoder.layers.10.layer_norm.bias\n",
      "encoder.layers.10.layer_norm.weight\n",
      "encoder.layers.11.attention.k_proj.bias\n",
      "encoder.layers.11.attention.k_proj.weight\n",
      "encoder.layers.11.attention.out_proj.bias\n",
      "encoder.layers.11.attention.out_proj.weight\n",
      "encoder.layers.11.attention.q_proj.bias\n",
      "encoder.layers.11.attention.q_proj.weight\n",
      "encoder.layers.11.attention.v_proj.bias\n",
      "encoder.layers.11.attention.v_proj.weight\n",
      "encoder.layers.11.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.11.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.11.feed_forward.output_dense.bias\n",
      "encoder.layers.11.feed_forward.output_dense.weight\n",
      "encoder.layers.11.final_layer_norm.bias\n",
      "encoder.layers.11.final_layer_norm.weight\n",
      "encoder.layers.11.layer_norm.bias\n",
      "encoder.layers.11.layer_norm.weight\n",
      "encoder.layers.2.attention.k_proj.bias\n",
      "encoder.layers.2.attention.k_proj.weight\n",
      "encoder.layers.2.attention.out_proj.bias\n",
      "encoder.layers.2.attention.out_proj.weight\n",
      "encoder.layers.2.attention.q_proj.bias\n",
      "encoder.layers.2.attention.q_proj.weight\n",
      "encoder.layers.2.attention.v_proj.bias\n",
      "encoder.layers.2.attention.v_proj.weight\n",
      "encoder.layers.2.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.2.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.2.feed_forward.output_dense.bias\n",
      "encoder.layers.2.feed_forward.output_dense.weight\n",
      "encoder.layers.2.final_layer_norm.bias\n",
      "encoder.layers.2.final_layer_norm.weight\n",
      "encoder.layers.2.layer_norm.bias\n",
      "encoder.layers.2.layer_norm.weight\n",
      "encoder.layers.3.attention.k_proj.bias\n",
      "encoder.layers.3.attention.k_proj.weight\n",
      "encoder.layers.3.attention.out_proj.bias\n",
      "encoder.layers.3.attention.out_proj.weight\n",
      "encoder.layers.3.attention.q_proj.bias\n",
      "encoder.layers.3.attention.q_proj.weight\n",
      "encoder.layers.3.attention.v_proj.bias\n",
      "encoder.layers.3.attention.v_proj.weight\n",
      "encoder.layers.3.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.3.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.3.feed_forward.output_dense.bias\n",
      "encoder.layers.3.feed_forward.output_dense.weight\n",
      "encoder.layers.3.final_layer_norm.bias\n",
      "encoder.layers.3.final_layer_norm.weight\n",
      "encoder.layers.3.layer_norm.bias\n",
      "encoder.layers.3.layer_norm.weight\n",
      "encoder.layers.4.attention.k_proj.bias\n",
      "encoder.layers.4.attention.k_proj.weight\n",
      "encoder.layers.4.attention.out_proj.bias\n",
      "encoder.layers.4.attention.out_proj.weight\n",
      "encoder.layers.4.attention.q_proj.bias\n",
      "encoder.layers.4.attention.q_proj.weight\n",
      "encoder.layers.4.attention.v_proj.bias\n",
      "encoder.layers.4.attention.v_proj.weight\n",
      "encoder.layers.4.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.4.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.4.feed_forward.output_dense.bias\n",
      "encoder.layers.4.feed_forward.output_dense.weight\n",
      "encoder.layers.4.final_layer_norm.bias\n",
      "encoder.layers.4.final_layer_norm.weight\n",
      "encoder.layers.4.layer_norm.bias\n",
      "encoder.layers.4.layer_norm.weight\n",
      "encoder.layers.5.attention.k_proj.bias\n",
      "encoder.layers.5.attention.k_proj.weight\n",
      "encoder.layers.5.attention.out_proj.bias\n",
      "encoder.layers.5.attention.out_proj.weight\n",
      "encoder.layers.5.attention.q_proj.bias\n",
      "encoder.layers.5.attention.q_proj.weight\n",
      "encoder.layers.5.attention.v_proj.bias\n",
      "encoder.layers.5.attention.v_proj.weight\n",
      "encoder.layers.5.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.5.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.5.feed_forward.output_dense.bias\n",
      "encoder.layers.5.feed_forward.output_dense.weight\n",
      "encoder.layers.5.final_layer_norm.bias\n",
      "encoder.layers.5.final_layer_norm.weight\n",
      "encoder.layers.5.layer_norm.bias\n",
      "encoder.layers.5.layer_norm.weight\n",
      "encoder.layers.6.attention.k_proj.bias\n",
      "encoder.layers.6.attention.k_proj.weight\n",
      "encoder.layers.6.attention.out_proj.bias\n",
      "encoder.layers.6.attention.out_proj.weight\n",
      "encoder.layers.6.attention.q_proj.bias\n",
      "encoder.layers.6.attention.q_proj.weight\n",
      "encoder.layers.6.attention.v_proj.bias\n",
      "encoder.layers.6.attention.v_proj.weight\n",
      "encoder.layers.6.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.6.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.6.feed_forward.output_dense.bias\n",
      "encoder.layers.6.feed_forward.output_dense.weight\n",
      "encoder.layers.6.final_layer_norm.bias\n",
      "encoder.layers.6.final_layer_norm.weight\n",
      "encoder.layers.6.layer_norm.bias\n",
      "encoder.layers.6.layer_norm.weight\n",
      "encoder.layers.7.attention.k_proj.bias\n",
      "encoder.layers.7.attention.k_proj.weight\n",
      "encoder.layers.7.attention.out_proj.bias\n",
      "encoder.layers.7.attention.out_proj.weight\n",
      "encoder.layers.7.attention.q_proj.bias\n",
      "encoder.layers.7.attention.q_proj.weight\n",
      "encoder.layers.7.attention.v_proj.bias\n",
      "encoder.layers.7.attention.v_proj.weight\n",
      "encoder.layers.7.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.7.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.7.feed_forward.output_dense.bias\n",
      "encoder.layers.7.feed_forward.output_dense.weight\n",
      "encoder.layers.7.final_layer_norm.bias\n",
      "encoder.layers.7.final_layer_norm.weight\n",
      "encoder.layers.7.layer_norm.bias\n",
      "encoder.layers.7.layer_norm.weight\n",
      "encoder.layers.8.attention.k_proj.bias\n",
      "encoder.layers.8.attention.k_proj.weight\n",
      "encoder.layers.8.attention.out_proj.bias\n",
      "encoder.layers.8.attention.out_proj.weight\n",
      "encoder.layers.8.attention.q_proj.bias\n",
      "encoder.layers.8.attention.q_proj.weight\n",
      "encoder.layers.8.attention.v_proj.bias\n",
      "encoder.layers.8.attention.v_proj.weight\n",
      "encoder.layers.8.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.8.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.8.feed_forward.output_dense.bias\n",
      "encoder.layers.8.feed_forward.output_dense.weight\n",
      "encoder.layers.8.final_layer_norm.bias\n",
      "encoder.layers.8.final_layer_norm.weight\n",
      "encoder.layers.8.layer_norm.bias\n",
      "encoder.layers.8.layer_norm.weight\n",
      "encoder.layers.9.attention.k_proj.bias\n",
      "encoder.layers.9.attention.k_proj.weight\n",
      "encoder.layers.9.attention.out_proj.bias\n",
      "encoder.layers.9.attention.out_proj.weight\n",
      "encoder.layers.9.attention.q_proj.bias\n",
      "encoder.layers.9.attention.q_proj.weight\n",
      "encoder.layers.9.attention.v_proj.bias\n",
      "encoder.layers.9.attention.v_proj.weight\n",
      "encoder.layers.9.feed_forward.intermediate_dense.bias\n",
      "encoder.layers.9.feed_forward.intermediate_dense.weight\n",
      "encoder.layers.9.feed_forward.output_dense.bias\n",
      "encoder.layers.9.feed_forward.output_dense.weight\n",
      "encoder.layers.9.final_layer_norm.bias\n",
      "encoder.layers.9.final_layer_norm.weight\n",
      "encoder.layers.9.layer_norm.bias\n",
      "encoder.layers.9.layer_norm.weight\n",
      "encoder.pos_conv_embed.conv.bias\n",
      "encoder.pos_conv_embed.conv.weight_g\n",
      "encoder.pos_conv_embed.conv.weight_v\n",
      "feature_extractor.conv_layers.0.conv.weight\n",
      "feature_extractor.conv_layers.0.layer_norm.bias\n",
      "feature_extractor.conv_layers.0.layer_norm.weight\n",
      "feature_extractor.conv_layers.1.conv.weight\n",
      "feature_extractor.conv_layers.2.conv.weight\n",
      "feature_extractor.conv_layers.3.conv.weight\n",
      "feature_extractor.conv_layers.4.conv.weight\n",
      "feature_extractor.conv_layers.5.conv.weight\n",
      "feature_extractor.conv_layers.6.conv.weight\n",
      "feature_projection.layer_norm.bias\n",
      "feature_projection.layer_norm.weight\n",
      "feature_projection.projection.bias\n",
      "feature_projection.projection.weight\n",
      "masked_spec_embed\n"
     ]
    }
   ],
   "source": [
    "for key in sorted(model_original.state_dict().keys()):\n",
    "    print(key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5f76688d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:42:39.938533Z",
     "start_time": "2023-08-30T20:42:37.546064Z"
    }
   },
   "outputs": [],
   "source": [
    "sd = torch.load(\"/app/suno/checkpoints/mert_25hz/checkpoint_last.pt\", map_location=\"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "566ed5dd",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:44:49.866014Z",
     "start_time": "2023-08-30T20:44:49.861745Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "211"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(model_original.state_dict().keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d106807e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:44:40.236236Z",
     "start_time": "2023-08-30T20:44:40.233333Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "220"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(sd[\"model\"].keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "96974b67",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:25:39.645006Z",
     "start_time": "2023-08-30T20:25:39.641202Z"
    }
   },
   "outputs": [],
   "source": [
    "def validate_model(test_model, test_processor):\n",
    "    audio_arr = Audio.from_file(\"audios/kitchen_sink_temperature_semantic_0.85_0.wav\").convert(24_000, 2, 1).get_segment(to_s=10).array_float\n",
    "    inputs = test_processor(audio_arr, sampling_rate=24_000, return_tensors=\"pt\")\n",
    "    with torch.no_grad():\n",
    "        outputs = test_model(**inputs, output_hidden_states=True)\n",
    "\n",
    "    audio_arr_2 = Audio.from_file(\"audios/kitchen_sink_temperature_semantic_0.85_0.wav\").convert(24_000, 2, 1).get_segment(to_s=5).array_float\n",
    "    inputs_2 = test_processor(audio_arr_2, sampling_rate=24_000, return_tensors=\"pt\")\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        outputs_2 = test_model(**inputs_2, output_hidden_states=True)\n",
    "    n_layer = 6\n",
    "    print(outputs.hidden_states[n_layer].numpy().shape, outputs_2.hidden_states[n_layer].numpy().shape)\n",
    "    plt.plot(outputs.hidden_states[n_layer][0].numpy().mean(-1))\n",
    "    plt.plot(outputs_2.hidden_states[n_layer][0].numpy().mean(-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f82918f8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:25:46.703198Z",
     "start_time": "2023-08-30T20:25:39.891590Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 749, 768) (1, 374, 768)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "validate_model(model_original, processor_original)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "795e7709",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:30:24.260978Z",
     "start_time": "2023-08-30T20:30:24.258496Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Wav2Vec2FeatureExtractor {\n",
       "  \"do_normalize\": true,\n",
       "  \"feature_extractor_type\": \"Wav2Vec2FeatureExtractor\",\n",
       "  \"feature_size\": 1,\n",
       "  \"padding_side\": \"right\",\n",
       "  \"padding_value\": 0,\n",
       "  \"return_attention_mask\": true,\n",
       "  \"sampling_rate\": 24000\n",
       "}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processor_original"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4d290924",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:30:17.640426Z",
     "start_time": "2023-08-30T20:30:17.637820Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Wav2Vec2FeatureExtractor {\n",
       "  \"do_normalize\": true,\n",
       "  \"feature_extractor_type\": \"Wav2Vec2FeatureExtractor\",\n",
       "  \"feature_size\": 1,\n",
       "  \"padding_side\": \"right\",\n",
       "  \"padding_value\": 0,\n",
       "  \"return_attention_mask\": true,\n",
       "  \"sampling_rate\": 24000\n",
       "}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6587cd28",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:25:48.157223Z",
     "start_time": "2023-08-30T20:25:46.704896Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import torch\n",
    "from transformers import Wav2Vec2FeatureExtractor\n",
    "from suno_utils.models.mert.modeling_MERT import MERTModel, MERTConfig\n",
    "from suno_utils.audio import Audio\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/preprocessor_config.json\") as f:\n",
    "    processor = Wav2Vec2FeatureExtractor(**json.load(f))\n",
    "\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/config_small_25hz.json\") as f:\n",
    "    cfg = MERTConfig(**json.load(f))\n",
    "model = MERTModel(cfg)\n",
    "sd = torch.load(\"/home/tony/Data/MERT/mert_test_25hz.pt\")\n",
    "model.load_state_dict(sd);\n",
    "model.eval();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2021bbe7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-30T20:25:53.279731Z",
     "start_time": "2023-08-30T20:25:48.158657Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 249, 768) (1, 124, 768)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "validate_model(model, processor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "683ada34",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "f520b381",
   "metadata": {},
   "source": [
    "# Seriously"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4ad9a716",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:51.036344Z",
     "start_time": "2023-08-31T04:35:46.026144Z"
    }
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.insert(0, \"/home/tony/Work/MERT/mert_fairseq/models/mert\")\n",
    "\n",
    "from s3prl.upstream.utils import load_fairseq_ckpt, merge_with_parent\n",
    "from mert_model import MERTModel, MERTConfig, HubertPretrainingConfig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7d3d1758",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:54.195525Z",
     "start_time": "2023-08-31T04:35:51.037674Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import torch\n",
    "from transformers import Wav2Vec2FeatureExtractor\n",
    "from suno_utils.models.mert.modeling_MERT import MERTModel as SunoMERT, MERTConfig as SunoMERTConfig\n",
    "from suno_utils.audio import Audio\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "with open(\"/home/tony/Work/glockenspiel/suno_utils/suno_utils/models/mert/preprocessor_config.json\") as f:\n",
    "    processor = Wav2Vec2FeatureExtractor(**json.load(f))\n",
    "\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/config_small_25hz.json\") as f:\n",
    "    cfg = SunoMERTConfig(**json.load(f))\n",
    "model = SunoMERT(cfg)\n",
    "sd = torch.load(\"/home/tony/Data/MERT/mert_test_25hz.pt\")\n",
    "model.load_state_dict(sd);\n",
    "model.eval();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5a299b4d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:54.915016Z",
     "start_time": "2023-08-31T04:35:54.197788Z"
    }
   },
   "outputs": [],
   "source": [
    "from_fp = \"/app/suno/checkpoints/mert_25hz/checkpoint_last.pt\"\n",
    "fair_state, fair_cfg = load_fairseq_ckpt(from_fp)\n",
    "output_state = {\n",
    "        \"task_cfg\": fair_cfg[\"task\"],\n",
    "        \"model_cfg\": fair_cfg[\"model\"],\n",
    "        \"model_weight\": fair_state[\"model\"],\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "921273e5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:54.918022Z",
     "start_time": "2023-08-31T04:35:54.916373Z"
    }
   },
   "outputs": [],
   "source": [
    "# task_cfg = merge_with_parent(HubertPretrainingConfig, ckpt_state[\"task_cfg\"])\n",
    "# model_cfg = merge_with_parent(HubertConfig, ckpt_state[\"model_cfg\"])\n",
    "# model = HubertModel(model_cfg, task_cfg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "13b0f9c9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:54.968081Z",
     "start_time": "2023-08-31T04:35:54.919132Z"
    }
   },
   "outputs": [],
   "source": [
    "task_cfg = merge_with_parent(HubertPretrainingConfig, output_state[\"task_cfg\"])\n",
    "model_cfg = merge_with_parent(MERTConfig, output_state[\"model_cfg\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "507ac0f3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:55.017753Z",
     "start_time": "2023-08-31T04:35:54.969105Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "HubertPretrainingConfig(_name='mert_pretraining', data='/app/suno/data/mert_25hz/audio_tsv', fine_tuning=False, labels=['codec_0', 'codec_1', 'codec_2', 'codec_3', 'codec_4', 'codec_5', 'codec_6', 'codec_7'], label_dir='/app/suno/data/mert_25hz/label', label_rate=25.0, sample_rate=24000, normalize=False, enable_padding=False, max_keep_size=None, max_sample_size=120000, min_sample_size=72000, single_target=False, random_crop=True, pad_audio=False)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "task_cfg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a036e483",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:56.758447Z",
     "start_time": "2023-08-31T04:35:55.018809Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-31 04:35:55 | INFO | mert_model | MERTModel Config: MERTConfig(_name='mert', label_rate=25.0, extractor_mode='default', encoder_layers=12, encoder_embed_dim=768, encoder_ffn_embed_dim=3072, encoder_attention_heads=12, activation_fn='gelu', layer_type='transformer', dropout=0.1, attention_dropout=0.1, activation_dropout=0.0, encoder_layerdrop=0.05, dropout_input=0.1, dropout_features=0.1, final_dim=64, untie_final_proj=True, layer_norm_first=False, audio_extract_type='w2v_conv', music_conv_nmel=80, music_conv_hoplen=40, conv_feature_layers='[(512,10,5)] + [(512,5,3)] + [(512,3,2)] * 4 + [(512,2,2)] * 2', conv_bias=False, logit_temp=0.1, target_glu=False, feature_grad_mult=0.1, conv_pos=128, conv_pos_groups=16, pos_conv_depth=1, mask_length=5, mask_prob=0.8, mask_dynamic_prob_step='[]', mask_dynamic_prob='[]', mask_dynamic_len_step='[]', mask_dynamic_len='[]', mask_selection='static', mask_other=0.0, no_mask_overlap=False, mask_min_space=1, mask_replace=0.0, mask_replace_type='in_sample', mask_origin=0.0, mask_channel_length=10, mask_channel_prob=0.0, mask_channel_selection='static', mask_channel_other=0.0, no_mask_channel_overlap=False, mask_channel_min_space=1, latent_temp=[2.0, 0.5, 0.999995], skip_masked=False, skip_nomask=True, checkpoint_activations=False, required_seq_len_multiple=2, depthwise_conv_kernel_size=31, attn_type='', pos_enc_type='abs', fp16=False, audio_cqt_loss_m=True, audio_cqt_bins=336, audio_mel_loss_m=False, audio_mel_bins=84, feature_extractor_cqt=False, feature_extractor_cqt_bins=84, mixture_prob=0.5, inbatch_noise_augment_len_range='[12000, 24000]', inbatch_noise_augment_number_range='[1, 3]', inbatch_noise_augment_volume=1.0, learnable_temp=False, learnable_temp_init=0.1, learnable_temp_max=100.0, chunk_nce_cal=-1, pretrained_weights='', random_codebook=-1, deepnorm=False, subln=False, emb_grad_mult=1.0, attention_relax=-1.0, do_cnn_feat_stable_layernorm=False, wav_normalize=False)\n",
      "2023-08-31 04:35:56 | INFO | mert_model | cannot find dictionary. assume will be used for fine-tuning\n",
      "2023-08-31 04:35:56 | INFO | mert_model | train the model with extra task: reconstruct cqt from transformer output\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CQT kernels created, time used = 0.2489 seconds\n"
     ]
    }
   ],
   "source": [
    "local_model = MERTModel(model_cfg,  task_cfg, [None])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2754a792",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:56.762116Z",
     "start_time": "2023-08-31T04:35:56.759929Z"
    }
   },
   "outputs": [],
   "source": [
    "for k in [\n",
    "    \"label_embs_concat\", # \"final_proj.weight\", \"final_proj.bias\",\n",
    "]:\n",
    "    del output_state[\"model_weight\"][k]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c60d52c3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:56.817285Z",
     "start_time": "2023-08-31T04:35:56.763280Z"
    }
   },
   "outputs": [],
   "source": [
    "local_model.remove_pretraining_modules()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "599f559d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:57.412643Z",
     "start_time": "2023-08-31T04:35:56.819692Z"
    }
   },
   "outputs": [],
   "source": [
    "local_model.load_state_dict(output_state[\"model_weight\"])\n",
    "# fucking dropout\n",
    "_ = local_model.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1ab6c52d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:35:57.416493Z",
     "start_time": "2023-08-31T04:35:57.414088Z"
    }
   },
   "outputs": [],
   "source": [
    "processor.do_normalize = True\n",
    "with open(\"/home/georg/code/glockenspiel/suno_utils/suno_utils/models/mert/preprocessor_config.json\") as f:\n",
    "    processor_false = Wav2Vec2FeatureExtractor(**json.load(f))\n",
    "processor_false.do_normalize = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "3cd7c921",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:44:35.226483Z",
     "start_time": "2023-08-31T04:44:28.284075Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "for time_slice in [5]:\n",
    "    audio_arr = Audio.from_file(\"audios/kitchen_sink_temperature_semantic_0.85_0.wav\").convert(24_000, 2, 1).get_segment(to_s=time_slice).array_float\n",
    "    inputs = processor(audio_arr, sampling_rate=24_000, return_tensors=\"pt\") \n",
    "    inputs_false = processor_false(audio_arr, sampling_rate=24_000, return_tensors=\"pt\")\n",
    "    with torch.no_grad():\n",
    "        outputs = model(inputs[\"input_values\"], output_hidden_states=True)\n",
    "    with torch.no_grad():\n",
    "        local_outputs = local_model.forward(inputs[\"input_values\"], features_only=True, mask=False)\n",
    "    with torch.no_grad():\n",
    "        outputs_no_norm = model(inputs_false[\"input_values\"], output_hidden_states=True)\n",
    "    with torch.no_grad():\n",
    "        local_outputs_no_norm = local_model.forward(inputs_false[\"input_values\"], features_only=True, mask=False)\n",
    "    plt.plot(outputs.last_hidden_state[0].mean(-1), label=\"suno HF MERT\")\n",
    "    plt.plot(local_outputs[\"x\"][0].mean(-1),  label=\"local MERT\")\n",
    "    plt.plot(outputs_no_norm.last_hidden_state[0].mean(-1), label=\"suno HF MERT no processor norm\")\n",
    "    plt.plot(local_outputs_no_norm[\"x\"][0].mean(-1),  label=\"local MERT no processor norm\")\n",
    "plt.xlabel(\"time\")\n",
    "plt.ylabel(\"mean last hidden state weights\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "c16c1aff",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T04:37:43.557315Z",
     "start_time": "2023-08-31T04:37:43.555878Z"
    }
   },
   "outputs": [],
   "source": [
    "# test_input = torch.rand([1, 240000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "efac7fd3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T03:53:32.070899Z",
     "start_time": "2023-08-31T03:53:32.025412Z"
    }
   },
   "outputs": [],
   "source": [
    "# local_model.eval()\n",
    "# local_model.traning = False\n",
    "# local_model(test_input, features_only=True, mask=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a14c51b3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-31T03:53:32.121540Z",
     "start_time": "2023-08-31T03:53:32.071906Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# model.eval()\n",
    "# model.training = False\n",
    "# model(test_input).last_hidden_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2eab5df5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a714319b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}