경량화 모델 및 기타 file commit

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+model/gaepago-20-lite filter=lfs diff=lfs merge=lfs -text
+model/gaepago-20-lite/model_quant_int8.pt filter=lfs diff=lfs merge=lfs -text

.ipynb_checkpoints/app-checkpoint.py

@@ -4,7 +4,7 @@
 from transformers import AutoModelForAudioClassification
 from transformers import AutoFeatureExtractor
 from transformers import pipeline
-from datasets import Dataset
+from datasets import Dataset, Audio
 import gradio as gr
 import torch
 
@@ -13,7 +13,10 @@ MODEL_NAME = "Gae8J/gaepago-20"
 DATASET_NAME = "Gae8J/modeling_v1"
 
 # Import Model & feature extractor
-model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)
+# model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)
+from transformers import AutoConfig
+config = AutoConfig.from_pretrained(MODEL_NAME)
+model = torch.jit.load(f"./model/gaepago-20-lite/model_quant_int8.pt")
 feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)
 
 # 모델 cpu로 변경하여 진행
@@ -27,9 +30,12 @@ def gaepago_fn(tmp_audio_dir):
                                ,sampling_rate=audio_dataset[0]["audio"]["sampling_rate"]
                                ,return_tensors="pt")
     with torch.no_grad():
-        logits = model(**inputs).logits
+        # logits = model(**inputs).logits
+        logits = model(**inputs)["logits"]        
+    # predicted_class_ids = torch.argmax(logits).item()
+    # predicted_label = model.config.id2label[predicted_class_ids]
     predicted_class_ids = torch.argmax(logits).item()
-    predicted_label = model.config.id2label[predicted_class_ids]
+    predicted_label = config.id2label[predicted_class_ids]
     
     return predicted_label
 
@@ -47,4 +53,4 @@ with main_api:
     b1.click(gaepago_fn, inputs=audio, outputs=transcription)
 #    examples = gr.Examples(examples=example_list,
 #                           inputs=[audio])
-main_api.launch()
+main_api.launch(share=True)

.ipynb_checkpoints/eval_and_inference-checkpoint.ipynb

@@ -0,0 +1,279 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "544a588c-68ff-440f-be5c-389f1f02a0b7",
+   "metadata": {},
+   "source": [
+    "# example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "7ef8c97c-cefd-4905-8d63-af303c412d1a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "MODEL_NAME = \"gaepago-20\"\n",
+    "DATASET_NAME = \"Gae8J/modeling_v1\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "044499ce-7821-4b59-9f4b-5971b6a24cce",
+   "metadata": {},
+   "source": [
+    "## load dataset (test data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "e827e3bb-820d-46b3-b2e8-fdb97787bde1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f078fd108d2044b48a961bee6ed49747",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['test']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d0c16b3d-32dd-4e61-86bd-e21232840e98",
+   "metadata": {},
+   "source": [
+    "## run"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "d504778d-4ba3-43d3-b22b-76ce838a5edf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoModelForAudioClassification\n",
+    "from transformers import AutoFeatureExtractor\n",
+    "import torch\n",
+    "\n",
+    "model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)\n",
+    "\n",
+    "preds = []\n",
+    "gts = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = model(**inputs).logits\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = model.config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(model.config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f200bec5-c2d9-4549-8bb8-1400c484f499",
+   "metadata": {},
+   "source": [
+    "## performance"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "be97683d-da60-4d23-abc9-0be9b86cd636",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark       0.56      0.62      0.59         8\n",
+      "    growling       1.00      0.83      0.91         6\n",
+      "        howl       0.75      0.86      0.80         7\n",
+      "     panting       1.00      0.80      0.89        10\n",
+      "     whimper       0.38      0.43      0.40         7\n",
+      "\n",
+      "    accuracy                           0.71        38\n",
+      "   macro avg       0.74      0.71      0.72        38\n",
+      "weighted avg       0.75      0.71      0.72        38\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "test_performance = classification_report(gts, preds)\n",
+    "print(test_performance)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ea3ee48d-19c7-4f9d-9c2c-4b03d4748acb",
+   "metadata": {},
+   "source": [
+    "## load dataset (validation data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "33e5051e-75a2-4523-905c-fe1dbc81eda2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "WARNING:datasets.builder:Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cf5cfe439c174b8284b4668419af6dca",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['validation']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "36bee3b3-e66f-46dc-8030-cef3cb62ff97",
+   "metadata": {},
+   "source": [
+    "## run"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "914a471c-5d76-482b-a4f3-3c5eeebdd697",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoModelForAudioClassification\n",
+    "import torch\n",
+    "\n",
+    "model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)\n",
+    "\n",
+    "preds = []\n",
+    "gts = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = model(**inputs).logits\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = model.config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(model.config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4f1d5bab-4f88-4628-918e-d14b29c2143b",
+   "metadata": {},
+   "source": [
+    "## performance"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "26e0c704-b5b6-4bf0-8b58-1e3615b76cb7",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark       0.75      0.67      0.71         9\n",
+      "    growling       1.00      0.71      0.83         7\n",
+      "        howl       0.86      0.86      0.86         7\n",
+      "     panting       1.00      0.70      0.82        10\n",
+      "     whimper       0.54      1.00      0.70         7\n",
+      "\n",
+      "    accuracy                           0.78        40\n",
+      "   macro avg       0.83      0.79      0.78        40\n",
+      "weighted avg       0.84      0.78      0.78        40\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "valid_performance = classification_report(gts, preds)\n",
+    "print(valid_performance)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "g3p8",
+   "language": "python",
+   "name": "g3p8"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

.ipynb_checkpoints/eval_and_inference_lite_v1-checkpoint.ipynb

@@ -0,0 +1,189 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "544a588c-68ff-440f-be5c-389f1f02a0b7",
+   "metadata": {},
+   "source": [
+    "# example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "7ef8c97c-cefd-4905-8d63-af303c412d1a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "MODEL_NAME = \"gaepago-20-lite\"\n",
+    "DATASET_NAME = \"Gae8J/modeling_v1\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "044499ce-7821-4b59-9f4b-5971b6a24cce",
+   "metadata": {},
+   "source": [
+    "## load dataset (test data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "e827e3bb-820d-46b3-b2e8-fdb97787bde1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "WARNING:datasets.builder:Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4438f0b33464423b92fecc698c1935e5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "from transformers import AutoFeatureExtractor\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['test']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "779c547a-7e27-4481-8a66-fd9900e41964",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoConfig\n",
+    "config = AutoConfig.from_pretrained(MODEL_NAME)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "03659af7-3d90-4431-a4ea-a8d99e93602f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "0f58cfcf-ba2d-45e4-b4e9-87df88e9dbad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "loaded_quantized_model = torch.jit.load(\"gaepago-20-lite/model_quant_int8.pt\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "52212656-a3e9-4bd2-ac2d-427acb5795c6",
+   "metadata": {},
+   "source": [
+    "## 모델결과"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "3d4f5365-d6f1-4163-9c47-ce8c89e13884",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "preds = []\n",
+    "gts = []\n",
+    "# quant_logits_list = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = loaded_quantized_model(**inputs)['logits']\n",
+    "#     quant_logits_list.append(logits)\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "93b3c424-bab6-4774-915e-9e9f534f762d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark     0.5556    0.6250    0.5882         8\n",
+      "    growling     1.0000    0.8333    0.9091         6\n",
+      "        howl     0.7500    0.8571    0.8000         7\n",
+      "     panting     1.0000    0.8000    0.8889        10\n",
+      "     whimper     0.3750    0.4286    0.4000         7\n",
+      "\n",
+      "    accuracy                         0.7105        38\n",
+      "   macro avg     0.7361    0.7088    0.7172        38\n",
+      "weighted avg     0.7452    0.7105    0.7224        38\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "test_performance = classification_report(gts, preds,digits=4)\n",
+    "print(test_performance)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "99a3ea38-54c8-4aed-9bbf-12f98bf09dc5",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "g3p8",
+   "language": "python",
+   "name": "g3p8"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

.ipynb_checkpoints/text_label-checkpoint.json

@@ -0,0 +1,528 @@
+{
+	"bark": [
+		[
+			"너무 신나서 어쩌지?",
+			"긍정"
+		],
+		[
+			"집사, 놀아줘!",
+			"긍정"
+		],
+		[
+			"지금 너무 신나!",
+			"긍정"
+		],
+		[
+			"누가 왔나 봐!",
+			"긍정"
+		],
+		[
+			"놀아줘!! 놀아달란말이야!!",
+			"긍정"
+		],
+		[
+			"안녕 🐶",
+			"긍정"
+		],
+		[
+			"난 너를 좋아하는 걸, 그런데 너는 나를 좋아해?",
+			"긍정"
+		],
+		[
+			"주목해줘! 놀자!",
+			"긍정"
+		],
+		[
+			"놀이 시간이야, 같이 놀자!",
+			"긍정"
+		],
+		[
+			"다가오지마!",
+			"부정"
+		],
+		[
+			"뭔가 이상한 소리 들려!",
+			"부정"
+		],
+		[
+			"경계해, 경계해!",
+			"부정"
+		],
+		[
+			"아니야, 아니야!",
+			"부정"
+		],
+		[
+			"건들지마!!!!",
+			"부정"
+		],
+		[
+			"뭔가 불안해, 도와줘!",
+			"부정"
+		],
+		[
+			"주인~ 뭐해~?",
+			"중립"
+		],
+		[
+			"밖에 뭐가 있는 거 같아!",
+			"중립"
+		],
+		[
+			"이리 와봐!",
+			"중립"
+		],
+		[
+			"날 보고있어?",
+			"중립"
+		],
+		[
+			"밖에 뭐 있어?",
+			"중립"
+		],
+		[
+			"이거 내꺼야!",
+			"중립"
+		],
+		[
+			"물 마실래, 마실 것 좀 줘.",
+			"중립"
+		],
+		[
+			"목이 말라, 물 좀 줄래?",
+			"중립"
+		]
+	],
+	"growling": [
+		[
+			"나 좀 내버려 둬!",
+			"부정"
+		],
+		[
+			"더 이상 다가오지마!",
+			"부정"
+		],
+		[
+			"너무 까다로워!",
+			"부정"
+		],
+		[
+			"내가 경계하고 있어!",
+			"부정"
+		],
+		[
+			"빨리 이리 와!",
+			"부정"
+		],
+		[
+			"나 너무 화나!",
+			"부정"
+		],
+		[
+			"나 싸울 준비됐어!",
+			"부정"
+		],
+		[
+			"그만 좀 해!",
+			"부정"
+		],
+		[
+			"내게 장난치지마!",
+			"부정"
+		],
+		[
+			"나 지금 너무 짜증나!",
+			"부정"
+		],
+		[
+			"나 지금 안 좋아!",
+			"부정"
+		],
+		[
+			"다가오지마!",
+			"부정"
+		],
+		[
+			"너에게 화난 거야!",
+			"부정"
+		],
+		[
+			"좀 멀리 가!",
+			"부정"
+		],
+		[
+			"나 싸우려고 준비됐어!",
+			"부정"
+		],
+		[
+			"한번 더 건드리면 물어버릴거야!!!",
+			"부정"
+		],
+		[
+			"나한테 이렇게 위협적으로 다가오지마!",
+			"부정"
+		],
+		[
+			"나의 영역을 침범하면 안돼! 이해해줘!",
+			"부정"
+		],
+		[
+			"그만 좀 귀찮게 해! 내가 분명히 경고했잖아!",
+			"부정"
+		],
+		[
+			"불편해, 물러서줘.",
+			"부정"
+		],
+		[
+			"경고하는 거야, 가까이 오지 마.",
+			"부정"
+		],
+		[
+			"좀 너무 가까워, 거리 좀 둬.",
+			"부정"
+		],
+		[
+			"나를 방해하지 마, 신경 써줘.",
+			"부정"
+		],
+		[
+			"내가 불편해, 거리 좀 두고 있어.",
+			"부정"
+		],
+		[
+			"가까이 오지 마.",
+			"부정"
+		],
+		[
+			"나를 방해하지 마, 존중해줘. Respect Me!!",
+			"부정"
+		]
+	],
+	"howl": [
+		[
+			"나 여기있어, 봐줘!",
+			"중립"
+		],
+		[
+			"너 어디 갔어?!",
+			"중립"
+		],
+		[
+			"나 너무 외로워!",
+			"중립"
+		],
+		[
+			"이리 와봐, 나 있는 곳으로!",
+			"중립"
+		],
+		[
+			"너 없으면 너무 심심해!",
+			"중립"
+		],
+		[
+			"나도 같이 가고 싶어!",
+			"중립"
+		],
+		[
+			"나 심심해",
+			"중립"
+		],
+		[
+			"어디야? 나 찾아봐!",
+			"중립"
+		],
+		[
+			"언제 오려고 그래?",
+			"중립"
+		],
+		[
+			"나는 여기 있는데!",
+			"중립"
+		],
+		[
+			"빨리 돌아와줘!",
+			"중립"
+		],
+		[
+			"나 혼자 남겨두지 마!",
+			"중립"
+		],
+		[
+			"나 여기있어!! 나좀 봐줘!!!",
+			"중립"
+		],
+		[
+			"나 잘 보고 있어? 나 괜찮아?",
+			"중립"
+		],
+		[
+			"주인, 나 좀 안아줄 수 있을까?",
+			"중립"
+		],
+		[
+			"외로워, 보고 싶어.",
+			"중립"
+		],
+		[
+			"다른 강아지와 '합창'하고 싶어.",
+			"중립"
+		],
+		[
+			"너를 보고싶어, 언제 와?",
+			"중립"
+		],
+		[
+			"무언가 알려고 하는 중이야.",
+			"중립"
+		],
+		[
+			"다른 강아지들이랑 노래하고 싶어.",
+			"긍정"
+		]
+	],
+	"panting": [
+		[
+			"더워~ 에어컨 켜줘.",
+			"부정"
+		],
+		[
+			"운동 후 휴식 중이야.",
+			"중립"
+		],
+		[
+			"숨이 차, 좀 도와줘.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 좀 쉬자.",
+			"부정"
+		],
+		[
+			"너무 더워, 물 좀 줄래?",
+			"부정"
+		],
+		[
+			"너무 더워, 바람 좀 쐬자.",
+			"부정"
+		],
+		[
+			"힘들게 운동했어, 휴식 좀!",
+			"부정"
+		],
+		[
+			"숨이 차, 쉬는 시간이 필요해.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 조용히 좀...",
+			"부정"
+		],
+		[
+			"물 좀 마시고 싶어, 줄래?",
+			"중립"
+		],
+		[
+			"많이 뛰어서 힘들어, 휴식이 필요해.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 좀 더 쉬자.",
+			"중립"
+		],
+		[
+			"너무 더워서 물 좀 마시고 싶어.",
+			"중립"
+		],
+		[
+			"좀 더운 �� 같아, 바람 좀 쐬고 싶어.",
+			"중립"
+		],
+		[
+			"지금 좀 쉴 시간이 필요해, 잠시만 기다려.",
+			"중립"
+		],
+		[
+			"지금 진정할 시간이 필요해!!!",
+			"중립"
+		],
+		[
+			"나 지금 너무 신나",
+			"긍정"
+		],
+		[
+			"너랑 놀면 더 재밌을 것 같아",
+			"긍정"
+		],
+		[
+			"나랑 놀지 않을래?",
+			"긍정"
+		],
+		[
+			"밖에 나가면 재미난 일이 있을 것 같아!",
+			"긍정"
+		],
+		[
+			"오늘은 무슨 일이 있을까? 좋은 일이 생길 것 같아!",
+			"긍정"
+		],
+		[
+			"세상 모든 것들이 반가워~",
+			"긍정"
+		],
+		[
+			"너랑 친해지고 싶어~",
+			"긍정"
+		],
+		[
+			"오늘 기분 아주 나이스~",
+			"긍정"
+		],
+		[
+			"세상에서 제일 좋아!!",
+			"긍정"
+		],
+		[
+			"나 지금 기분이가 좋아~",
+			"긍정"
+		],
+		[
+			"너랑 놀고싶어~",
+			"긍정"
+		],
+		[
+			"오늘 되게 행복한 하루다~",
+			"긍정"
+		],
+		[
+			"오늘 내 생일인가? 너무 행복해><",
+			"긍정"
+		],
+		[
+			"만나서 반가워",
+			"긍정"
+		],
+		[
+			"너는 이름이 뭐니?",
+			"긍정"
+		],
+		[
+			"난 너가 좋아!!",
+			"긍정"
+		],
+		[
+			"나 매우 재밌어",
+			"긍정"
+		],
+		[
+			"나랑 같이 놀러 나가자",
+			"긍정"
+		]
+	],
+	"whimper": [
+		[
+			"나 너무 두려워",
+			"부정"
+		],
+		[
+			"나 지금 너무 외로워",
+			"부정"
+		],
+		[
+			"나 너무 슬퍼",
+			"부정"
+		],
+		[
+			"나 좀 안아줘",
+			"부정"
+		],
+		[
+			"나 지금 너무 불편해",
+			"부정"
+		],
+		[
+			"나 너무 피곤해",
+			"부정"
+		],
+		[
+			"조금만 더 안아줘",
+			"부정"
+		],
+		[
+			"나 좀 위로해줘",
+			"부정"
+		],
+		[
+			"나 기다리는 중",
+			"부정"
+		],
+		[
+			"외로워서 눈물이 나",
+			"부정"
+		],
+		[
+			"나 상처받았어, 너무 두려워...ㅠㅡㅠ",
+			"부정"
+		],
+		[
+			"나 놀래쪄ㅠㅡㅠ 힝구힝구..",
+			"부정"
+		],
+		[
+			"무셔워... 안아죠~~~",
+			"부정"
+		],
+		[
+			"너무 슬퍼서 맘이 아파... 안아줘...",
+			"부정"
+		],
+		[
+			"나 기분이 너무 안 좋아... 어떻게 해줄래?",
+			"부정"
+		],
+		[
+			"힝...미안해...",
+			"부정"
+		],
+		[
+			"불안해, 곁에 있어줘.",
+			"부정"
+		],
+		[
+			"밖으로 나가고 싶어.",
+			"중립"
+		],
+		[
+			"미안해, 실수했어.",
+			"부정"
+		],
+		[
+			"너무 슬퍼, 위로 좀 해줘.",
+			"부정"
+		],
+		[
+			"스트레스 받았어, 도와줘.",
+			"부정"
+		],
+		[
+			"내가 불안해, 붙어있어줘.",
+			"부정"
+		],
+		[
+			"너무 외로워, 애정을 보여줘.",
+			"부정"
+		],
+		[
+			"산책 좀 가고 싶어.",
+			"중립"
+		],
+		[
+			"정말 슬퍼, 안아줘.",
+			"부정"
+		],
+		[
+			"스트레스가 너무 많아, 안아줘.",
+			"부정"
+		]
+	]
+}

.ipynb_checkpoints/text_mapping_example-checkpoint.ipynb

@@ -0,0 +1,90 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "8f925fb7-86ba-487f-ab85-88754d777860",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import json\n",
+    "with open(\"text/text_label.json\",\"r\",encoding='utf-8')  as f:\n",
+    "    text_label = json.load(f)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "d2c0a048-1db7-4236-9f26-539ed31d3d27",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import random\n",
+    "random.seed(0)\n",
+    "def post_process(model_output,text_label):\n",
+    "    text_list = text_label[model_output]\n",
+    "    text,sent = random.sample(text_list,1)[0]\n",
+    "    return {'label' : model_output,\n",
+    "            'text' : text,\n",
+    "            'sentiment' : sent}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "f8ca0ad8-bc0c-4766-8e13-fe093c5290df",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'label': 'bark', 'text': '아니야, 아니야!', 'sentiment': '부정'}"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model_output = 'bark'\n",
+    "post_process(model_output,text_label)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "da690a64-4dea-4b2a-89c1-23ea8bad955c",
+   "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": "3.10.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

app.py

@@ -13,7 +13,10 @@ MODEL_NAME = "Gae8J/gaepago-20"
 DATASET_NAME = "Gae8J/modeling_v1"
 
 # Import Model & feature extractor
-model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)
+# model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)
+from transformers import AutoConfig
+config = AutoConfig.from_pretrained(MODEL_NAME)
+model = torch.jit.load(f"./model/gaepago-20-lite/model_quant_int8.pt")
 feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)
 
 # 모델 cpu로 변경하여 진행
@@ -27,9 +30,12 @@ def gaepago_fn(tmp_audio_dir):
                                ,sampling_rate=audio_dataset[0]["audio"]["sampling_rate"]
                                ,return_tensors="pt")
     with torch.no_grad():
-        logits = model(**inputs).logits
+        # logits = model(**inputs).logits
+        logits = model(**inputs)["logits"]        
+    # predicted_class_ids = torch.argmax(logits).item()
+    # predicted_label = model.config.id2label[predicted_class_ids]
     predicted_class_ids = torch.argmax(logits).item()
-    predicted_label = model.config.id2label[predicted_class_ids]
+    predicted_label = config.id2label[predicted_class_ids]
     
     return predicted_label
 
@@ -47,4 +53,4 @@ with main_api:
     b1.click(gaepago_fn, inputs=audio, outputs=transcription)
 #    examples = gr.Examples(examples=example_list,
 #                           inputs=[audio])
-main_api.launch()
+main_api.launch(share=True)

eval_and_inference.ipynb

@@ -0,0 +1,279 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "544a588c-68ff-440f-be5c-389f1f02a0b7",
+   "metadata": {},
+   "source": [
+    "# example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "7ef8c97c-cefd-4905-8d63-af303c412d1a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "MODEL_NAME = \"gaepago-20\"\n",
+    "DATASET_NAME = \"Gae8J/modeling_v1\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "044499ce-7821-4b59-9f4b-5971b6a24cce",
+   "metadata": {},
+   "source": [
+    "## load dataset (test data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "e827e3bb-820d-46b3-b2e8-fdb97787bde1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f078fd108d2044b48a961bee6ed49747",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['test']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d0c16b3d-32dd-4e61-86bd-e21232840e98",
+   "metadata": {},
+   "source": [
+    "## run"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "d504778d-4ba3-43d3-b22b-76ce838a5edf",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoModelForAudioClassification\n",
+    "from transformers import AutoFeatureExtractor\n",
+    "import torch\n",
+    "\n",
+    "model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)\n",
+    "\n",
+    "preds = []\n",
+    "gts = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = model(**inputs).logits\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = model.config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(model.config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f200bec5-c2d9-4549-8bb8-1400c484f499",
+   "metadata": {},
+   "source": [
+    "## performance"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "be97683d-da60-4d23-abc9-0be9b86cd636",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark       0.56      0.62      0.59         8\n",
+      "    growling       1.00      0.83      0.91         6\n",
+      "        howl       0.75      0.86      0.80         7\n",
+      "     panting       1.00      0.80      0.89        10\n",
+      "     whimper       0.38      0.43      0.40         7\n",
+      "\n",
+      "    accuracy                           0.71        38\n",
+      "   macro avg       0.74      0.71      0.72        38\n",
+      "weighted avg       0.75      0.71      0.72        38\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "test_performance = classification_report(gts, preds)\n",
+    "print(test_performance)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ea3ee48d-19c7-4f9d-9c2c-4b03d4748acb",
+   "metadata": {},
+   "source": [
+    "## load dataset (validation data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "33e5051e-75a2-4523-905c-fe1dbc81eda2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "WARNING:datasets.builder:Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cf5cfe439c174b8284b4668419af6dca",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['validation']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "36bee3b3-e66f-46dc-8030-cef3cb62ff97",
+   "metadata": {},
+   "source": [
+    "## run"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "914a471c-5d76-482b-a4f3-3c5eeebdd697",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoModelForAudioClassification\n",
+    "import torch\n",
+    "\n",
+    "model = AutoModelForAudioClassification.from_pretrained(MODEL_NAME)\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)\n",
+    "\n",
+    "preds = []\n",
+    "gts = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = model(**inputs).logits\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = model.config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(model.config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4f1d5bab-4f88-4628-918e-d14b29c2143b",
+   "metadata": {},
+   "source": [
+    "## performance"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "26e0c704-b5b6-4bf0-8b58-1e3615b76cb7",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark       0.75      0.67      0.71         9\n",
+      "    growling       1.00      0.71      0.83         7\n",
+      "        howl       0.86      0.86      0.86         7\n",
+      "     panting       1.00      0.70      0.82        10\n",
+      "     whimper       0.54      1.00      0.70         7\n",
+      "\n",
+      "    accuracy                           0.78        40\n",
+      "   macro avg       0.83      0.79      0.78        40\n",
+      "weighted avg       0.84      0.78      0.78        40\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "valid_performance = classification_report(gts, preds)\n",
+    "print(valid_performance)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "g3p8",
+   "language": "python",
+   "name": "g3p8"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

eval_and_inference_lite_v1.ipynb

@@ -0,0 +1,189 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "544a588c-68ff-440f-be5c-389f1f02a0b7",
+   "metadata": {},
+   "source": [
+    "# example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "7ef8c97c-cefd-4905-8d63-af303c412d1a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "MODEL_NAME = \"gaepago-20-lite\"\n",
+    "DATASET_NAME = \"Gae8J/modeling_v1\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "044499ce-7821-4b59-9f4b-5971b6a24cce",
+   "metadata": {},
+   "source": [
+    "## load dataset (test data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "e827e3bb-820d-46b3-b2e8-fdb97787bde1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "WARNING:datasets.builder:Found cached dataset parquet (/home/jovyan/.cache/huggingface/datasets/Gae8J___parquet/Gae8J--modeling_v1-b480c78c61a26816/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4438f0b33464423b92fecc698c1935e5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset, Audio\n",
+    "from transformers import AutoFeatureExtractor\n",
+    "dataset = load_dataset(DATASET_NAME)\n",
+    "dataset = dataset.cast_column(\"audio\", Audio(sampling_rate=16000))\n",
+    "test_data = dataset['test']\n",
+    "sampling_rate = test_data.features[\"audio\"].sampling_rate\n",
+    "feature_extractor = AutoFeatureExtractor.from_pretrained(MODEL_NAME)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "779c547a-7e27-4481-8a66-fd9900e41964",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import AutoConfig\n",
+    "config = AutoConfig.from_pretrained(MODEL_NAME)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "03659af7-3d90-4431-a4ea-a8d99e93602f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "0f58cfcf-ba2d-45e4-b4e9-87df88e9dbad",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "loaded_quantized_model = torch.jit.load(\"gaepago-20-lite/model_quant_int8.pt\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "52212656-a3e9-4bd2-ac2d-427acb5795c6",
+   "metadata": {},
+   "source": [
+    "## 모델결과"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "3d4f5365-d6f1-4163-9c47-ce8c89e13884",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "preds = []\n",
+    "gts = []\n",
+    "# quant_logits_list = []\n",
+    "for i in range(len(test_data)):\n",
+    "    inputs = feature_extractor(test_data[i][\"audio\"][\"array\"], sampling_rate=sampling_rate, return_tensors=\"pt\")\n",
+    "    with torch.no_grad():\n",
+    "        logits = loaded_quantized_model(**inputs)['logits']\n",
+    "#     quant_logits_list.append(logits)\n",
+    "    predicted_class_ids = torch.argmax(logits).item()\n",
+    "    predicted_label = config.id2label[predicted_class_ids]\n",
+    "    preds.append(predicted_label)\n",
+    "    gts.append(config.id2label[test_data[i]['label']])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "93b3c424-bab6-4774-915e-9e9f534f762d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        bark     0.5556    0.6250    0.5882         8\n",
+      "    growling     1.0000    0.8333    0.9091         6\n",
+      "        howl     0.7500    0.8571    0.8000         7\n",
+      "     panting     1.0000    0.8000    0.8889        10\n",
+      "     whimper     0.3750    0.4286    0.4000         7\n",
+      "\n",
+      "    accuracy                         0.7105        38\n",
+      "   macro avg     0.7361    0.7088    0.7172        38\n",
+      "weighted avg     0.7452    0.7105    0.7224        38\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import classification_report\n",
+    "test_performance = classification_report(gts, preds,digits=4)\n",
+    "print(test_performance)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "99a3ea38-54c8-4aed-9bbf-12f98bf09dc5",
+   "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": "3.8.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

model/gaepago-20-lite/.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,131 @@
+{
+  "_name_or_path": "gaepago-20",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "sum",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "howl",
+    "1": "growling",
+    "2": "bark",
+    "3": "panting",
+    "4": "whimper"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "bark": "2",
+    "growling": "1",
+    "howl": "0",
+    "panting": "3",
+    "whimper": "4"
+  },
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 256,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.29.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

model/gaepago-20-lite/config.json

@@ -0,0 +1,131 @@
+{
+  "_name_or_path": "gaepago-20",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "sum",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "howl",
+    "1": "growling",
+    "2": "bark",
+    "3": "panting",
+    "4": "whimper"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "bark": "2",
+    "growling": "1",
+    "howl": "0",
+    "panting": "3",
+    "whimper": "4"
+  },
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 256,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.29.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

model/gaepago-20-lite/model_quant_int8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f76b504bb04245a11ec92b145dcfb53391b2105fa204b082fd5c58a862447769
+size 122374341

model/gaepago-20-lite/preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

requirements.txt

@@ -3,15 +3,19 @@ aiohttp==3.8.4
 aiosignal==1.3.1
 altair==5.0.1
 anyio==3.7.0
+appdirs==1.4.4
 async-timeout==4.0.2
 attrs==23.1.0
+audioread==3.0.0
 certifi==2023.5.7
+cffi==1.15.1
 charset-normalizer==3.1.0
 click==8.1.3
 cmake==3.26.4
 contourpy==1.1.0
 cycler==0.11.0
 datasets==2.13.0
+decorator==5.1.1
 dill==0.3.6
 exceptiongroup==1.1.1
 fastapi==0.97.0
@@ -27,21 +31,28 @@ httpcore==0.17.2
 httpx==0.24.1
 huggingface-hub==0.15.1
 idna==3.4
+importlib-metadata==6.7.0
 importlib-resources==5.12.0
 Jinja2==3.1.2
+joblib==1.2.0
 jsonschema==4.17.3
 kiwisolver==1.4.4
+lazy_loader==0.2
+librosa==0.10.0.post2
 linkify-it-py==2.0.2
 lit==16.0.6
+llvmlite==0.40.1rc1
 markdown-it-py==2.2.0
 MarkupSafe==2.1.3
 matplotlib==3.7.1
 mdit-py-plugins==0.3.3
 mdurl==0.1.2
 mpmath==1.3.0
+msgpack==1.0.5
 multidict==6.0.4
 multiprocess==0.70.14
 networkx==3.1
+numba==0.57.0
 numpy==1.24.3
 nvidia-cublas-cu11==11.10.3.66
 nvidia-cuda-cupti-cu11==11.7.101
@@ -59,7 +70,9 @@ packaging==23.1
 pandas==2.0.2
 Pillow==9.5.0
 pkgutil_resolve_name==1.3.10
+pooch==1.6.0
 pyarrow==12.0.1
+pycparser==2.21
 pydantic==1.10.9
 pydub==0.25.1
 Pygments==2.15.1
@@ -72,11 +85,16 @@ PyYAML==6.0
 regex==2023.6.3
 requests==2.31.0
 safetensors==0.3.1
+scikit-learn==1.2.2
+scipy==1.10.1
 semantic-version==2.10.0
 six==1.16.0
 sniffio==1.3.0
+soundfile==0.12.1
+soxr==0.3.5
 starlette==0.27.0
 sympy==1.12
+threadpoolctl==3.1.0
 tokenizers==0.13.3
 toolz==0.12.0
 torch==2.0.1

text_label.json

@@ -0,0 +1,528 @@
+{
+	"bark": [
+		[
+			"너무 신나서 어쩌지?",
+			"긍정"
+		],
+		[
+			"집사, 놀아줘!",
+			"긍정"
+		],
+		[
+			"지금 너무 신나!",
+			"긍정"
+		],
+		[
+			"누가 왔나 봐!",
+			"긍정"
+		],
+		[
+			"놀아줘!! 놀아달란말이야!!",
+			"긍정"
+		],
+		[
+			"안녕 🐶",
+			"긍정"
+		],
+		[
+			"난 너를 좋아하는 걸, 그런데 너는 나를 좋아해?",
+			"긍정"
+		],
+		[
+			"주목해줘! 놀자!",
+			"긍정"
+		],
+		[
+			"놀이 시간이야, 같이 놀자!",
+			"긍정"
+		],
+		[
+			"다가오지마!",
+			"부정"
+		],
+		[
+			"뭔가 이상한 소리 들려!",
+			"부정"
+		],
+		[
+			"경계해, 경계해!",
+			"부정"
+		],
+		[
+			"아니야, 아니야!",
+			"부정"
+		],
+		[
+			"건들지마!!!!",
+			"부정"
+		],
+		[
+			"뭔가 불안해, 도와줘!",
+			"부정"
+		],
+		[
+			"주인~ 뭐해~?",
+			"중립"
+		],
+		[
+			"밖에 뭐가 있는 거 같아!",
+			"중립"
+		],
+		[
+			"이리 와봐!",
+			"중립"
+		],
+		[
+			"날 보고있어?",
+			"중립"
+		],
+		[
+			"밖에 뭐 있어?",
+			"중립"
+		],
+		[
+			"이거 내꺼야!",
+			"중립"
+		],
+		[
+			"물 마실래, 마실 것 좀 줘.",
+			"중립"
+		],
+		[
+			"목이 말라, 물 좀 줄래?",
+			"중립"
+		]
+	],
+	"growling": [
+		[
+			"나 좀 내버려 둬!",
+			"부정"
+		],
+		[
+			"더 이상 다가오지마!",
+			"부정"
+		],
+		[
+			"너무 까다로워!",
+			"부정"
+		],
+		[
+			"내가 경계하고 있어!",
+			"부정"
+		],
+		[
+			"빨리 이리 와!",
+			"부정"
+		],
+		[
+			"나 너무 화나!",
+			"부정"
+		],
+		[
+			"나 싸울 준비됐어!",
+			"부정"
+		],
+		[
+			"그만 좀 해!",
+			"부정"
+		],
+		[
+			"내게 장난치지마!",
+			"부정"
+		],
+		[
+			"나 지금 너무 짜증나!",
+			"부정"
+		],
+		[
+			"나 지금 안 좋아!",
+			"부정"
+		],
+		[
+			"다가오지마!",
+			"부정"
+		],
+		[
+			"너에게 화난 거야!",
+			"부정"
+		],
+		[
+			"좀 멀리 가!",
+			"부정"
+		],
+		[
+			"나 싸우려고 준비됐어!",
+			"부정"
+		],
+		[
+			"한번 더 건드리면 물어버릴거야!!!",
+			"부정"
+		],
+		[
+			"나한테 이렇게 위협적으로 다가오지마!",
+			"부정"
+		],
+		[
+			"나의 영역을 침범하면 안돼! 이해해줘!",
+			"부정"
+		],
+		[
+			"그만 좀 귀찮게 해! 내가 분명히 경고했잖아!",
+			"부정"
+		],
+		[
+			"불편해, 물러서줘.",
+			"부정"
+		],
+		[
+			"경고하는 거야, 가까이 오지 마.",
+			"부정"
+		],
+		[
+			"좀 너무 가까워, 거리 좀 둬.",
+			"부정"
+		],
+		[
+			"나를 방해하지 마, 신경 써줘.",
+			"부정"
+		],
+		[
+			"내가 불편해, 거리 좀 두고 있어.",
+			"부정"
+		],
+		[
+			"가까이 오지 마.",
+			"부정"
+		],
+		[
+			"나를 방해하지 마, 존중해줘. Respect Me!!",
+			"부정"
+		]
+	],
+	"howl": [
+		[
+			"나 여기있어, 봐줘!",
+			"중립"
+		],
+		[
+			"너 어디 갔어?!",
+			"중립"
+		],
+		[
+			"나 너무 외로워!",
+			"중립"
+		],
+		[
+			"이리 와봐, 나 있는 곳으로!",
+			"중립"
+		],
+		[
+			"너 없으면 너무 심심해!",
+			"중립"
+		],
+		[
+			"나도 같이 가고 싶어!",
+			"중립"
+		],
+		[
+			"나 심심해",
+			"중립"
+		],
+		[
+			"어디야? 나 찾아봐!",
+			"중립"
+		],
+		[
+			"언제 오려고 그래?",
+			"중립"
+		],
+		[
+			"나는 여기 있는데!",
+			"중립"
+		],
+		[
+			"빨리 돌아와줘!",
+			"중립"
+		],
+		[
+			"나 혼자 남겨두지 마!",
+			"중립"
+		],
+		[
+			"나 여기있어!! 나좀 봐줘!!!",
+			"중립"
+		],
+		[
+			"나 잘 보고 있어? 나 괜찮아?",
+			"중립"
+		],
+		[
+			"주인, 나 좀 안아줄 수 있을까?",
+			"중립"
+		],
+		[
+			"외로워, 보고 싶어.",
+			"중립"
+		],
+		[
+			"다른 강아지와 '합창'하고 싶어.",
+			"중립"
+		],
+		[
+			"너를 보고싶어, 언제 와?",
+			"중립"
+		],
+		[
+			"무언가 알려고 하는 중이야.",
+			"중립"
+		],
+		[
+			"다른 강아지들이랑 노래하고 싶어.",
+			"긍정"
+		]
+	],
+	"panting": [
+		[
+			"더워~ 에어컨 켜줘.",
+			"부정"
+		],
+		[
+			"운동 후 휴식 중이야.",
+			"중립"
+		],
+		[
+			"숨이 차, 좀 도와줘.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 좀 쉬자.",
+			"부정"
+		],
+		[
+			"너무 더워, 물 좀 줄래?",
+			"부정"
+		],
+		[
+			"너무 더워, 바람 좀 쐬자.",
+			"부정"
+		],
+		[
+			"힘들게 운동했어, 휴식 좀!",
+			"부정"
+		],
+		[
+			"숨이 차, 쉬는 시간이 필요해.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 조용히 좀...",
+			"부정"
+		],
+		[
+			"물 좀 마시고 싶어, 줄래?",
+			"중립"
+		],
+		[
+			"많이 뛰어서 힘들어, 휴식이 필요해.",
+			"부정"
+		],
+		[
+			"휴식이 필요해, 좀 더 쉬자.",
+			"중립"
+		],
+		[
+			"너무 더워서 물 좀 마시고 싶어.",
+			"중립"
+		],
+		[
+			"좀 더운 �� 같아, 바람 좀 쐬고 싶어.",
+			"중립"
+		],
+		[
+			"지금 좀 쉴 시간이 필요해, 잠시만 기다려.",
+			"중립"
+		],
+		[
+			"지금 진정할 시간이 필요해!!!",
+			"중립"
+		],
+		[
+			"나 지금 너무 신나",
+			"긍정"
+		],
+		[
+			"너랑 놀면 더 재밌을 것 같아",
+			"긍정"
+		],
+		[
+			"나랑 놀지 않을래?",
+			"긍정"
+		],
+		[
+			"밖에 나가면 재미난 일이 있을 것 같아!",
+			"긍정"
+		],
+		[
+			"오늘은 무슨 일이 있을까? 좋은 일이 생길 것 같아!",
+			"긍정"
+		],
+		[
+			"세상 모든 것들이 반가워~",
+			"긍정"
+		],
+		[
+			"너랑 친해지고 싶어~",
+			"긍정"
+		],
+		[
+			"오늘 기분 아주 나이스~",
+			"긍정"
+		],
+		[
+			"세상에서 제일 좋아!!",
+			"긍정"
+		],
+		[
+			"나 지금 기분이가 좋아~",
+			"긍정"
+		],
+		[
+			"너랑 놀고싶어~",
+			"긍정"
+		],
+		[
+			"오늘 되게 행복한 하루다~",
+			"긍정"
+		],
+		[
+			"오늘 내 생일인가? 너무 행복해><",
+			"긍정"
+		],
+		[
+			"만나서 반가워",
+			"긍정"
+		],
+		[
+			"너는 이름이 뭐니?",
+			"긍정"
+		],
+		[
+			"난 너가 좋아!!",
+			"긍정"
+		],
+		[
+			"나 매우 재밌어",
+			"긍정"
+		],
+		[
+			"나랑 같이 놀러 나가자",
+			"긍정"
+		]
+	],
+	"whimper": [
+		[
+			"나 너무 두려워",
+			"부정"
+		],
+		[
+			"나 지금 너무 외로워",
+			"부정"
+		],
+		[
+			"나 너무 슬퍼",
+			"부정"
+		],
+		[
+			"나 좀 안아줘",
+			"부정"
+		],
+		[
+			"나 지금 너무 불편해",
+			"부정"
+		],
+		[
+			"나 너무 피곤해",
+			"부정"
+		],
+		[
+			"조금만 더 안아줘",
+			"부정"
+		],
+		[
+			"나 좀 위로해줘",
+			"부정"
+		],
+		[
+			"나 기다리는 중",
+			"부정"
+		],
+		[
+			"외로워서 눈물이 나",
+			"부정"
+		],
+		[
+			"나 상처받았어, 너무 두려워...ㅠㅡㅠ",
+			"부정"
+		],
+		[
+			"나 놀래쪄ㅠㅡㅠ 힝구힝구..",
+			"부정"
+		],
+		[
+			"무셔워... 안아죠~~~",
+			"부정"
+		],
+		[
+			"너무 슬퍼서 맘이 아파... 안아줘...",
+			"부정"
+		],
+		[
+			"나 기분이 너무 안 좋아... 어떻게 해줄래?",
+			"부정"
+		],
+		[
+			"힝...미안해...",
+			"부정"
+		],
+		[
+			"불안해, 곁에 있어줘.",
+			"부정"
+		],
+		[
+			"밖으로 나가고 싶어.",
+			"중립"
+		],
+		[
+			"미안해, 실수했어.",
+			"부정"
+		],
+		[
+			"너무 슬퍼, 위로 좀 해줘.",
+			"부정"
+		],
+		[
+			"스트레스 받았어, 도와줘.",
+			"부정"
+		],
+		[
+			"내가 불안해, 붙어있어줘.",
+			"부정"
+		],
+		[
+			"너무 외로워, 애정을 보여줘.",
+			"부정"
+		],
+		[
+			"산책 좀 가고 싶어.",
+			"중립"
+		],
+		[
+			"정말 슬퍼, 안아줘.",
+			"부정"
+		],
+		[
+			"스트레스가 너무 많아, 안아줘.",
+			"부정"
+		]
+	]
+}

text_mapping_example.ipynb

@@ -0,0 +1,90 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "8f925fb7-86ba-487f-ab85-88754d777860",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import json\n",
+    "with open(\"text/text_label.json\",\"r\",encoding='utf-8')  as f:\n",
+    "    text_label = json.load(f)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "d2c0a048-1db7-4236-9f26-539ed31d3d27",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import random\n",
+    "random.seed(0)\n",
+    "def post_process(model_output,text_label):\n",
+    "    text_list = text_label[model_output]\n",
+    "    text,sent = random.sample(text_list,1)[0]\n",
+    "    return {'label' : model_output,\n",
+    "            'text' : text,\n",
+    "            'sentiment' : sent}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "f8ca0ad8-bc0c-4766-8e13-fe093c5290df",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'label': 'bark', 'text': '아니야, 아니야!', 'sentiment': '부정'}"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model_output = 'bark'\n",
+    "post_process(model_output,text_label)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "da690a64-4dea-4b2a-89c1-23ea8bad955c",
+   "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": "3.8.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}