{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"gpuType": "T4"
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "code",
"source": [
"!pip install librosa numpy tensorflow scikit-learn sounddevice\n",
"!pip install gradio"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "QdDz0RCbIBwe",
"outputId": "8cb2c152-52ae-4a62-a0bf-c3067481af1f"
},
"execution_count": 1,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Requirement already satisfied: librosa in /usr/local/lib/python3.11/dist-packages (0.10.2.post1)\n",
"Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (1.26.4)\n",
"Requirement already satisfied: tensorflow in /usr/local/lib/python3.11/dist-packages (2.18.0)\n",
"Requirement already satisfied: scikit-learn in /usr/local/lib/python3.11/dist-packages (1.6.1)\n",
"Requirement already satisfied: sounddevice in /usr/local/lib/python3.11/dist-packages (0.5.1)\n",
"Requirement already satisfied: audioread>=2.1.9 in /usr/local/lib/python3.11/dist-packages (from librosa) (3.0.1)\n",
"Requirement already satisfied: scipy>=1.2.0 in /usr/local/lib/python3.11/dist-packages (from librosa) (1.14.1)\n",
"Requirement already satisfied: joblib>=0.14 in /usr/local/lib/python3.11/dist-packages (from librosa) (1.4.2)\n",
"Requirement already satisfied: decorator>=4.3.0 in /usr/local/lib/python3.11/dist-packages (from librosa) (4.4.2)\n",
"Requirement already satisfied: numba>=0.51.0 in /usr/local/lib/python3.11/dist-packages (from librosa) (0.60.0)\n",
"Requirement already satisfied: soundfile>=0.12.1 in /usr/local/lib/python3.11/dist-packages (from librosa) (0.13.1)\n",
"Requirement already satisfied: pooch>=1.1 in /usr/local/lib/python3.11/dist-packages (from librosa) (1.8.2)\n",
"Requirement already satisfied: soxr>=0.3.2 in /usr/local/lib/python3.11/dist-packages (from librosa) (0.5.0.post1)\n",
"Requirement already satisfied: typing-extensions>=4.1.1 in /usr/local/lib/python3.11/dist-packages (from librosa) (4.12.2)\n",
"Requirement already satisfied: lazy-loader>=0.1 in /usr/local/lib/python3.11/dist-packages (from librosa) (0.4)\n",
"Requirement already satisfied: msgpack>=1.0 in /usr/local/lib/python3.11/dist-packages (from librosa) (1.1.0)\n",
"Requirement already satisfied: absl-py>=1.0.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.4.0)\n",
"Requirement already satisfied: astunparse>=1.6.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.6.3)\n",
"Requirement already satisfied: flatbuffers>=24.3.25 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (25.2.10)\n",
"Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.6.0)\n",
"Requirement already satisfied: google-pasta>=0.1.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.2.0)\n",
"Requirement already satisfied: libclang>=13.0.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (18.1.1)\n",
"Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.4.0)\n",
"Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from tensorflow) (24.2)\n",
"Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<6.0.0dev,>=3.20.3 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (4.25.6)\n",
"Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.32.3)\n",
"Requirement already satisfied: setuptools in /usr/local/lib/python3.11/dist-packages (from tensorflow) (75.1.0)\n",
"Requirement already satisfied: six>=1.12.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.17.0)\n",
"Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.5.0)\n",
"Requirement already satisfied: wrapt>=1.11.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.17.2)\n",
"Requirement already satisfied: grpcio<2.0,>=1.24.3 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (1.71.0)\n",
"Requirement already satisfied: tensorboard<2.19,>=2.18 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (2.18.0)\n",
"Requirement already satisfied: keras>=3.5.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.8.0)\n",
"Requirement already satisfied: h5py>=3.11.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (3.12.1)\n",
"Requirement already satisfied: ml-dtypes<0.5.0,>=0.4.0 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.4.1)\n",
"Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /usr/local/lib/python3.11/dist-packages (from tensorflow) (0.37.1)\n",
"Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.11/dist-packages (from scikit-learn) (3.5.0)\n",
"Requirement already satisfied: CFFI>=1.0 in /usr/local/lib/python3.11/dist-packages (from sounddevice) (1.17.1)\n",
"Requirement already satisfied: wheel<1.0,>=0.23.0 in /usr/local/lib/python3.11/dist-packages (from astunparse>=1.6.0->tensorflow) (0.45.1)\n",
"Requirement already satisfied: pycparser in /usr/local/lib/python3.11/dist-packages (from CFFI>=1.0->sounddevice) (2.22)\n",
"Requirement already satisfied: rich in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (13.9.4)\n",
"Requirement already satisfied: namex in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (0.0.8)\n",
"Requirement already satisfied: optree in /usr/local/lib/python3.11/dist-packages (from keras>=3.5.0->tensorflow) (0.14.1)\n",
"Requirement already satisfied: llvmlite<0.44,>=0.43.0dev0 in /usr/local/lib/python3.11/dist-packages (from numba>=0.51.0->librosa) (0.43.0)\n",
"Requirement already satisfied: platformdirs>=2.5.0 in /usr/local/lib/python3.11/dist-packages (from pooch>=1.1->librosa) (4.3.6)\n",
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (3.4.1)\n",
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (3.10)\n",
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (2.3.0)\n",
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/dist-packages (from requests<3,>=2.21.0->tensorflow) (2025.1.31)\n",
"Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (3.7)\n",
"Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (0.7.2)\n",
"Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.11/dist-packages (from tensorboard<2.19,>=2.18->tensorflow) (3.1.3)\n",
"Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.11/dist-packages (from werkzeug>=1.0.1->tensorboard<2.19,>=2.18->tensorflow) (2.1.5)\n",
"Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.11/dist-packages (from rich->keras>=3.5.0->tensorflow) (3.0.0)\n",
"Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.11/dist-packages (from rich->keras>=3.5.0->tensorflow) (2.18.0)\n",
"Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.11/dist-packages (from markdown-it-py>=2.2.0->rich->keras>=3.5.0->tensorflow) (0.1.2)\n",
"Requirement already satisfied: gradio in /usr/local/lib/python3.11/dist-packages (5.21.0)\n",
"Requirement already satisfied: aiofiles<24.0,>=22.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (23.2.1)\n",
"Requirement already satisfied: anyio<5.0,>=3.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (3.7.1)\n",
"Requirement already satisfied: fastapi<1.0,>=0.115.2 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.115.11)\n",
"Requirement already satisfied: ffmpy in /usr/local/lib/python3.11/dist-packages (from gradio) (0.5.0)\n",
"Requirement already satisfied: gradio-client==1.7.2 in /usr/local/lib/python3.11/dist-packages (from gradio) (1.7.2)\n",
"Requirement already satisfied: groovy~=0.1 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.1.2)\n",
"Requirement already satisfied: httpx>=0.24.1 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.28.1)\n",
"Requirement already satisfied: huggingface-hub>=0.28.1 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.28.1)\n",
"Requirement already satisfied: jinja2<4.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (3.1.6)\n",
"Requirement already satisfied: markupsafe~=2.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (2.1.5)\n",
"Requirement already satisfied: numpy<3.0,>=1.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (1.26.4)\n",
"Requirement already satisfied: orjson~=3.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (3.10.15)\n",
"Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from gradio) (24.2)\n",
"Requirement already satisfied: pandas<3.0,>=1.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (2.2.2)\n",
"Requirement already satisfied: pillow<12.0,>=8.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (11.1.0)\n",
"Requirement already satisfied: pydantic>=2.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (2.10.6)\n",
"Requirement already satisfied: pydub in /usr/local/lib/python3.11/dist-packages (from gradio) (0.25.1)\n",
"Requirement already satisfied: python-multipart>=0.0.18 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.0.20)\n",
"Requirement already satisfied: pyyaml<7.0,>=5.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (6.0.2)\n",
"Requirement already satisfied: ruff>=0.9.3 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.11.0)\n",
"Requirement already satisfied: safehttpx<0.2.0,>=0.1.6 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.1.6)\n",
"Requirement already satisfied: semantic-version~=2.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (2.10.0)\n",
"Requirement already satisfied: starlette<1.0,>=0.40.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.46.1)\n",
"Requirement already satisfied: tomlkit<0.14.0,>=0.12.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.13.2)\n",
"Requirement already satisfied: typer<1.0,>=0.12 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.15.2)\n",
"Requirement already satisfied: typing-extensions~=4.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (4.12.2)\n",
"Requirement already satisfied: uvicorn>=0.14.0 in /usr/local/lib/python3.11/dist-packages (from gradio) (0.34.0)\n",
"Requirement already satisfied: fsspec in /usr/local/lib/python3.11/dist-packages (from gradio-client==1.7.2->gradio) (2024.10.0)\n",
"Requirement already satisfied: websockets<16.0,>=10.0 in /usr/local/lib/python3.11/dist-packages (from gradio-client==1.7.2->gradio) (14.2)\n",
"Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.11/dist-packages (from anyio<5.0,>=3.0->gradio) (3.10)\n",
"Requirement already satisfied: sniffio>=1.1 in /usr/local/lib/python3.11/dist-packages (from anyio<5.0,>=3.0->gradio) (1.3.1)\n",
"Requirement already satisfied: certifi in /usr/local/lib/python3.11/dist-packages (from httpx>=0.24.1->gradio) (2025.1.31)\n",
"Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.11/dist-packages (from httpx>=0.24.1->gradio) (1.0.7)\n",
"Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.11/dist-packages (from httpcore==1.*->httpx>=0.24.1->gradio) (0.14.0)\n",
"Requirement already satisfied: filelock in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.28.1->gradio) (3.17.0)\n",
"Requirement already satisfied: requests in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.28.1->gradio) (2.32.3)\n",
"Requirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.28.1->gradio) (4.67.1)\n",
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/dist-packages (from pandas<3.0,>=1.0->gradio) (2.8.2)\n",
"Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.11/dist-packages (from pandas<3.0,>=1.0->gradio) (2025.1)\n",
"Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.11/dist-packages (from pandas<3.0,>=1.0->gradio) (2025.1)\n",
"Requirement already satisfied: annotated-types>=0.6.0 in /usr/local/lib/python3.11/dist-packages (from pydantic>=2.0->gradio) (0.7.0)\n",
"Requirement already satisfied: pydantic-core==2.27.2 in /usr/local/lib/python3.11/dist-packages (from pydantic>=2.0->gradio) (2.27.2)\n",
"Requirement already satisfied: click>=8.0.0 in /usr/local/lib/python3.11/dist-packages (from typer<1.0,>=0.12->gradio) (8.1.8)\n",
"Requirement already satisfied: shellingham>=1.3.0 in /usr/local/lib/python3.11/dist-packages (from typer<1.0,>=0.12->gradio) (1.5.4)\n",
"Requirement already satisfied: rich>=10.11.0 in /usr/local/lib/python3.11/dist-packages (from typer<1.0,>=0.12->gradio) (13.9.4)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.11/dist-packages (from python-dateutil>=2.8.2->pandas<3.0,>=1.0->gradio) (1.17.0)\n",
"Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.11/dist-packages (from rich>=10.11.0->typer<1.0,>=0.12->gradio) (3.0.0)\n",
"Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.11/dist-packages (from rich>=10.11.0->typer<1.0,>=0.12->gradio) (2.18.0)\n",
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.28.1->gradio) (3.4.1)\n",
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.28.1->gradio) (2.3.0)\n",
"Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.11/dist-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->typer<1.0,>=0.12->gradio) (0.1.2)\n"
]
}
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"id": "D5BPAg-OGnnD"
},
"outputs": [],
"source": [
"import os\n",
"import numpy as np\n",
"import librosa\n",
"import tensorflow as tf\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.preprocessing import LabelEncoder\n",
"from tensorflow.keras.utils import to_categorical\n",
"from transformers import TFWav2Vec2Model, Wav2Vec2FeatureExtractor\n",
"from imblearn.over_sampling import SMOTE\n",
"from pydub import AudioSegment\n",
"import gradio as gr\n",
"import joblib\n",
"from sklearn.utils.class_weight import compute_class_weight"
]
},
{
"cell_type": "code",
"source": [
"# === Configuration ===\n",
"SAMPLE_RATE = 16000 # Matches Wav2Vec2 requirements\n",
"DURATION = 3 # Standardize audio clips to 3 seconds\n",
"MAX_AUDIO_LENGTH = SAMPLE_RATE * DURATION\n",
"PRETRAINED_MODEL_NAME = \"facebook/wav2vec2-base-960h\"\n",
"DATASET_PATH = \"/content/drive/MyDrive/dataset/YAF DATASET\" # Adjust this path as needed\n",
"\n",
"# Initialize Wav2Vec2 model and feature extractor\n",
"wav2vec2 = TFWav2Vec2Model.from_pretrained(PRETRAINED_MODEL_NAME)\n",
"feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(PRETRAINED_MODEL_NAME)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "bWV4e2kKGs39",
"outputId": "0d721d1c-307a-4117-8e7a-9ded57986367"
},
"execution_count": 3,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"/usr/local/lib/python3.11/dist-packages/huggingface_hub/utils/_auth.py:94: UserWarning: \n",
"The secret `HF_TOKEN` does not exist in your Colab secrets.\n",
"To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n",
"You will be able to reuse this secret in all of your notebooks.\n",
"Please note that authentication is recommended but still optional to access public models or datasets.\n",
" warnings.warn(\n",
"\n",
"TFWav2Vec2Model has backpropagation operations that are NOT supported on CPU. If you wish to train/fine-tune this model, you need a GPU or a TPU\n",
"Some weights of the PyTorch model were not used when initializing the TF 2.0 model TFWav2Vec2Model: ['lm_head.weight', 'lm_head.bias']\n",
"- This IS expected if you are initializing TFWav2Vec2Model from a PyTorch model trained on another task or with another architecture (e.g. initializing a TFBertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing TFWav2Vec2Model from a PyTorch model that you expect to be exactly identical (e.g. initializing a TFBertForSequenceClassification model from a BertForSequenceClassification model).\n",
"Some weights or buffers of the TF 2.0 model TFWav2Vec2Model were not initialized from the PyTorch model and are newly initialized: ['wav2vec2.masked_spec_embed']\n",
"You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# === Audio Preprocessing ===\n",
"def load_and_preprocess_audio(file_path):\n",
" \"\"\"Load and preprocess audio to a fixed length.\"\"\"\n",
" try:\n",
" audio, _ = librosa.load(file_path, sr=SAMPLE_RATE, duration=DURATION)\n",
" except Exception as e:\n",
" print(f\"Error loading {file_path}: {e}\")\n",
" return None\n",
" if len(audio) < MAX_AUDIO_LENGTH:\n",
" audio = np.pad(audio, (0, MAX_AUDIO_LENGTH - len(audio)))\n",
" else:\n",
" audio = audio[:MAX_AUDIO_LENGTH]\n",
" return audio"
],
"metadata": {
"id": "meaDR05VG8ke"
},
"execution_count": 4,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# === Standard Data Augmentation ===\n",
"def augment(audio):\n",
" \"\"\"Apply random pitch shifting and noise addition (no time stretching).\"\"\"\n",
" if np.random.rand() < 0.5:\n",
" audio = librosa.effects.pitch_shift(audio, sr=SAMPLE_RATE, n_steps=np.random.uniform(-2, 2))\n",
" if np.random.rand() < 0.5:\n",
" noise = np.random.normal(0, 0.005, audio.shape)\n",
" audio = audio + noise\n",
" return audio\n"
],
"metadata": {
"id": "PiEncyiTG9VE"
},
"execution_count": 5,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# === Feature Extraction ===\n",
"def extract_features(audio):\n",
" \"\"\"Extract full Wav2Vec2 feature sequence.\"\"\"\n",
" inputs = feature_extractor(audio, sampling_rate=SAMPLE_RATE, return_tensors=\"tf\")\n",
" wav_features = wav2vec2(inputs.input_values).last_hidden_state[0] # Shape: (time_steps, 768)\n",
" return wav_features.numpy()"
],
"metadata": {
"id": "Pnk5-vbEHA-V"
},
"execution_count": 6,
"outputs": []
},
{
"cell_type": "code",
"source": [
"\n",
"# === Enhanced Data Loading ===\n",
"def load_enhanced_dataset(dataset_path):\n",
" \"\"\"Load dataset with speaker IDs and apply augmentation to minority classes.\"\"\"\n",
" features, labels, speakers = [], [], []\n",
" for emotion in os.listdir(dataset_path):\n",
" emotion_dir = os.path.join(dataset_path, emotion)\n",
" if os.path.isdir(emotion_dir):\n",
" for file in os.listdir(emotion_dir):\n",
" if file.endswith(\".wav\"):\n",
" file_path = os.path.join(emotion_dir, file)\n",
" audio = load_and_preprocess_audio(file_path)\n",
" if audio is None:\n",
" continue\n",
" feat = extract_features(audio)\n",
" features.append(feat)\n",
" # Extract speaker ID from emotion label (e.g., 'YAF' from 'YAF_happy')\n",
" speaker_id = emotion.split('_')[0]\n",
" speakers.append(speaker_id)\n",
" labels.append(emotion)\n",
" # Augment for minority classes\n",
" if emotion in ['YAF_happy', 'YAF_sad']:\n",
" aug_audio = augment(audio)\n",
" aug_feat = extract_features(aug_audio)\n",
" features.append(aug_feat)\n",
" speakers.append(speaker_id)\n",
" labels.append(emotion)\n",
" # Encode speaker IDs numerically\n",
" speaker_encoder = LabelEncoder()\n",
" speakers_encoded = speaker_encoder.fit_transform(speakers)\n",
" return np.array(features), np.array(labels), speakers_encoded, speaker_encoder\n"
],
"metadata": {
"id": "ZMGSVxAzHFW_"
},
"execution_count": 7,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# === Advanced Model Architecture ===\n",
"def build_enhanced_model(time_steps, feature_dim, num_speakers, num_classes):\n",
" \"\"\"Build a model with sequence input, speaker embedding, LSTM, and attention.\"\"\"\n",
" audio_input = tf.keras.Input(shape=(time_steps, feature_dim), name='audio_input')\n",
" speaker_input = tf.keras.Input(shape=(1,), name='speaker_input')\n",
" # Embed speaker ID and repeat across time steps\n",
" speaker_embed = tf.keras.layers.Embedding(num_speakers, 8)(speaker_input)\n",
" speaker_embed = tf.keras.layers.Flatten()(speaker_embed)\n",
" speaker_embed = tf.keras.layers.RepeatVector(time_steps)(speaker_embed)\n",
" # Combine audio and speaker features\n",
" combined = tf.keras.layers.concatenate([audio_input, speaker_embed], axis=-1)\n",
" # Bidirectional LSTM with attention\n",
" x = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(128, return_sequences=True))(combined)\n",
" x = tf.keras.layers.Attention()([x, x])\n",
" x = tf.keras.layers.GlobalAveragePooling1D()(x)\n",
" x = tf.keras.layers.Dense(128, activation='relu')(x)\n",
" x = tf.keras.layers.Dropout(0.5)(x)\n",
" # Final Dense layer outputs raw logits (no activation)\n",
" outputs = tf.keras.layers.Dense(num_classes, activation=None)(x)\n",
"\n",
" model = tf.keras.Model(inputs=[audio_input, speaker_input], outputs=outputs)\n",
" # Fine-tuning: unfreeze wav2vec2 for adaptation\n",
" wav2vec2.trainable = True\n",
" optimizer = tf.keras.optimizers.Adam(1e-5) # Low learning rate\n",
" model.compile(optimizer=optimizer,\n",
" loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),\n",
" metrics=['accuracy'])\n",
" return model"
],
"metadata": {
"id": "nh4EeWwiHHpJ"
},
"execution_count": 8,
"outputs": []
},
{
"cell_type": "code",
"source": [
"\n",
"# === Main Execution ===\n",
"if __name__ == \"__main__\":\n",
" # Load dataset with speaker information\n",
" X, y, speakers, speaker_encoder = load_enhanced_dataset(DATASET_PATH)\n",
" label_encoder = LabelEncoder()\n",
" y_encoded = label_encoder.fit_transform(y)\n",
" y_onehot = to_categorical(y_encoded)\n",
"\n",
" # Handle class imbalance with SMOTE\n",
" X_flat = X.reshape(-1, X.shape[1] * X.shape[2])\n",
" smote = SMOTE()\n",
" X_res_flat, y_res = smote.fit_resample(X_flat, y_encoded)\n",
" X_res = X_res_flat.reshape(-1, X.shape[1], X.shape[2])\n",
" y_res = to_categorical(y_res)\n",
""
],
"metadata": {
"id": "rphJ4dtmHJjm"
},
"execution_count": 9,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Split dataset\n",
"X_train, X_test, y_train, y_test = train_test_split(\n",
" X_res, y_res, test_size=0.2, random_state=42, stratify=y_res # Stratify by y_res\n",
")\n",
"\n",
"# Resample the speaker data to match the new data shape\n",
"speaker_res = np.repeat(speakers, np.ceil(len(X_res) / len(speakers)))[:len(X_res)]\n",
"\n",
"# Now, split the resampled speaker data along with the data\n",
"speaker_train, speaker_test = train_test_split(\n",
" speaker_res, test_size=0.2, random_state=42, stratify=y_res # Stratify by y_res\n",
")\n",
"\n",
"time_steps, feature_dim = X_train.shape[1], X_train.shape[2]\n",
"num_speakers = len(speaker_encoder.classes_) # Removed extra indentation\n",
"num_classes = y_onehot.shape[1] # Removed extra indentation"
],
"metadata": {
"id": "pqR9blxIHNqJ"
},
"execution_count": 13,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Build and train model\n",
"model = build_enhanced_model(time_steps, feature_dim, num_speakers, num_classes)\n",
"model.summary()\n",
"class_weights = compute_class_weight('balanced', classes=np.unique(y_encoded), y=y_encoded)\n",
"\n",
"history = model.fit(\n",
" [X_train, speaker_train], y_train,\n",
" validation_data=([X_test, speaker_test], y_test),\n",
" epochs=50,\n",
" batch_size=32,\n",
" class_weight=dict(enumerate(class_weights)),\n",
" callbacks=[\n",
" tf.keras.callbacks.EarlyStopping(patience=5, restore_best_weights=True),\n",
" tf.keras.callbacks.ReduceLROnPlateau(factor=0.5, patience=3)\n",
" ]\n",
" )"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"id": "hyJSdOskHXh2",
"outputId": "00abc60f-bd38-435c-efa7-193b2ced64f5"
},
"execution_count": 15,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1mModel: \"functional\"\u001b[0m\n"
],
"text/html": [
"Model: \"functional\"\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ speaker_input │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
"│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ embedding (\u001b[38;5;33mEmbedding\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m8\u001b[0m) │ \u001b[38;5;34m8\u001b[0m │ speaker_input[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ embedding[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ audio_input (\u001b[38;5;33mInputLayer\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m768\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ repeat_vector │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m8\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ flatten[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mRepeatVector\u001b[0m) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ concatenate (\u001b[38;5;33mConcatenate\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m776\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ audio_input[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ repeat_vector[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ bidirectional │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m926,720\u001b[0m │ concatenate[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mBidirectional\u001b[0m) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ attention (\u001b[38;5;33mAttention\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m149\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bidirectional[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ bidirectional[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ global_average_pooling1d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ attention[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mGlobalAveragePooling1D\u001b[0m) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m32,896\u001b[0m │ global_average_poolin… │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m) │ \u001b[38;5;34m645\u001b[0m │ dropout[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"└───────────────────────────┴────────────────────────┴────────────────┴────────────────────────┘\n"
],
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃ Connected to ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ speaker_input │ (None, 1) │ 0 │ - │\n",
"│ (InputLayer) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ embedding (Embedding) │ (None, 1, 8) │ 8 │ speaker_input[0][0] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ flatten (Flatten) │ (None, 8) │ 0 │ embedding[0][0] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ audio_input (InputLayer) │ (None, 149, 768) │ 0 │ - │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ repeat_vector │ (None, 149, 8) │ 0 │ flatten[0][0] │\n",
"│ (RepeatVector) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ concatenate (Concatenate) │ (None, 149, 776) │ 0 │ audio_input[0][0], │\n",
"│ │ │ │ repeat_vector[0][0] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ bidirectional │ (None, 149, 256) │ 926,720 │ concatenate[0][0] │\n",
"│ (Bidirectional) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ attention (Attention) │ (None, 149, 256) │ 0 │ bidirectional[0][0], │\n",
"│ │ │ │ bidirectional[0][0] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ global_average_pooling1d │ (None, 256) │ 0 │ attention[0][0] │\n",
"│ (GlobalAveragePooling1D) │ │ │ │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dense (Dense) │ (None, 128) │ 32,896 │ global_average_poolin… │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dropout (Dropout) │ (None, 128) │ 0 │ dense[0][0] │\n",
"├───────────────────────────┼────────────────────────┼────────────────┼────────────────────────┤\n",
"│ dense_1 (Dense) │ (None, 5) │ 645 │ dropout[0][0] │\n",
"└───────────────────────────┴────────────────────────┴────────────────┴────────────────────────┘\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m960,269\u001b[0m (3.66 MB)\n"
],
"text/html": [
" Total params: 960,269 (3.66 MB)\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m960,269\u001b[0m (3.66 MB)\n"
],
"text/html": [
" Trainable params: 960,269 (3.66 MB)\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
],
"text/html": [
" Non-trainable params: 0 (0.00 B)\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"Epoch 1/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 52ms/step - accuracy: 0.2126 - loss: 1.8167 - val_accuracy: 0.3550 - val_loss: 1.5980 - learning_rate: 1.0000e-05\n",
"Epoch 2/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.2259 - loss: 1.7496 - val_accuracy: 0.3725 - val_loss: 1.5877 - learning_rate: 1.0000e-05\n",
"Epoch 3/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 38ms/step - accuracy: 0.2732 - loss: 1.7249 - val_accuracy: 0.3900 - val_loss: 1.5707 - learning_rate: 1.0000e-05\n",
"Epoch 4/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.2897 - loss: 1.6866 - val_accuracy: 0.4175 - val_loss: 1.5469 - learning_rate: 1.0000e-05\n",
"Epoch 5/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.3233 - loss: 1.6432 - val_accuracy: 0.4200 - val_loss: 1.5183 - learning_rate: 1.0000e-05\n",
"Epoch 6/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.3692 - loss: 1.5907 - val_accuracy: 0.4400 - val_loss: 1.4816 - learning_rate: 1.0000e-05\n",
"Epoch 7/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.3881 - loss: 1.5541 - val_accuracy: 0.4650 - val_loss: 1.4324 - learning_rate: 1.0000e-05\n",
"Epoch 8/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 33ms/step - accuracy: 0.4236 - loss: 1.4783 - val_accuracy: 0.4575 - val_loss: 1.3681 - learning_rate: 1.0000e-05\n",
"Epoch 9/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.4236 - loss: 1.4111 - val_accuracy: 0.4675 - val_loss: 1.3021 - learning_rate: 1.0000e-05\n",
"Epoch 10/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.4405 - loss: 1.3170 - val_accuracy: 0.4775 - val_loss: 1.2408 - learning_rate: 1.0000e-05\n",
"Epoch 11/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.4661 - loss: 1.2461 - val_accuracy: 0.5200 - val_loss: 1.1904 - learning_rate: 1.0000e-05\n",
"Epoch 12/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.4960 - loss: 1.1909 - val_accuracy: 0.5400 - val_loss: 1.1487 - learning_rate: 1.0000e-05\n",
"Epoch 13/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 31ms/step - accuracy: 0.4760 - loss: 1.1721 - val_accuracy: 0.5600 - val_loss: 1.1179 - learning_rate: 1.0000e-05\n",
"Epoch 14/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.5220 - loss: 1.1201 - val_accuracy: 0.5775 - val_loss: 1.0832 - learning_rate: 1.0000e-05\n",
"Epoch 15/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.5509 - loss: 1.0576 - val_accuracy: 0.5675 - val_loss: 1.0315 - learning_rate: 1.0000e-05\n",
"Epoch 16/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.5761 - loss: 1.0416 - val_accuracy: 0.6150 - val_loss: 0.9876 - learning_rate: 1.0000e-05\n",
"Epoch 17/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.5882 - loss: 1.0076 - val_accuracy: 0.6075 - val_loss: 0.9579 - learning_rate: 1.0000e-05\n",
"Epoch 18/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 31ms/step - accuracy: 0.5964 - loss: 0.9858 - val_accuracy: 0.6375 - val_loss: 0.9172 - learning_rate: 1.0000e-05\n",
"Epoch 19/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.6365 - loss: 0.9032 - val_accuracy: 0.6425 - val_loss: 0.9103 - learning_rate: 1.0000e-05\n",
"Epoch 20/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.6461 - loss: 0.8624 - val_accuracy: 0.6525 - val_loss: 0.8516 - learning_rate: 1.0000e-05\n",
"Epoch 21/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.6431 - loss: 0.8475 - val_accuracy: 0.6750 - val_loss: 0.8211 - learning_rate: 1.0000e-05\n",
"Epoch 22/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 32ms/step - accuracy: 0.6869 - loss: 0.8131 - val_accuracy: 0.6875 - val_loss: 0.7836 - learning_rate: 1.0000e-05\n",
"Epoch 23/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 33ms/step - accuracy: 0.6748 - loss: 0.8001 - val_accuracy: 0.7050 - val_loss: 0.7506 - learning_rate: 1.0000e-05\n",
"Epoch 24/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.6901 - loss: 0.7635 - val_accuracy: 0.7075 - val_loss: 0.7369 - learning_rate: 1.0000e-05\n",
"Epoch 25/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 35ms/step - accuracy: 0.7058 - loss: 0.7296 - val_accuracy: 0.7275 - val_loss: 0.6809 - learning_rate: 1.0000e-05\n",
"Epoch 26/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 44ms/step - accuracy: 0.7418 - loss: 0.6508 - val_accuracy: 0.7550 - val_loss: 0.6529 - learning_rate: 1.0000e-05\n",
"Epoch 27/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 50ms/step - accuracy: 0.7381 - loss: 0.6755 - val_accuracy: 0.7275 - val_loss: 0.6935 - learning_rate: 1.0000e-05\n",
"Epoch 28/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.7410 - loss: 0.6279 - val_accuracy: 0.7625 - val_loss: 0.6355 - learning_rate: 1.0000e-05\n",
"Epoch 29/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 52ms/step - accuracy: 0.7481 - loss: 0.6329 - val_accuracy: 0.7600 - val_loss: 0.6302 - learning_rate: 1.0000e-05\n",
"Epoch 30/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 45ms/step - accuracy: 0.7426 - loss: 0.6093 - val_accuracy: 0.7850 - val_loss: 0.5961 - learning_rate: 1.0000e-05\n",
"Epoch 31/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 75ms/step - accuracy: 0.7727 - loss: 0.5551 - val_accuracy: 0.8000 - val_loss: 0.5790 - learning_rate: 1.0000e-05\n",
"Epoch 32/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 61ms/step - accuracy: 0.7872 - loss: 0.5304 - val_accuracy: 0.7975 - val_loss: 0.5452 - learning_rate: 1.0000e-05\n",
"Epoch 33/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 35ms/step - accuracy: 0.8003 - loss: 0.5277 - val_accuracy: 0.8150 - val_loss: 0.5350 - learning_rate: 1.0000e-05\n",
"Epoch 34/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 36ms/step - accuracy: 0.7944 - loss: 0.5542 - val_accuracy: 0.7700 - val_loss: 0.6045 - learning_rate: 1.0000e-05\n",
"Epoch 35/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.8164 - loss: 0.4469 - val_accuracy: 0.8050 - val_loss: 0.5263 - learning_rate: 1.0000e-05\n",
"Epoch 36/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 38ms/step - accuracy: 0.7986 - loss: 0.4953 - val_accuracy: 0.8300 - val_loss: 0.4949 - learning_rate: 1.0000e-05\n",
"Epoch 37/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 39ms/step - accuracy: 0.8164 - loss: 0.4580 - val_accuracy: 0.8350 - val_loss: 0.4910 - learning_rate: 1.0000e-05\n",
"Epoch 38/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 33ms/step - accuracy: 0.8153 - loss: 0.4393 - val_accuracy: 0.8500 - val_loss: 0.4688 - learning_rate: 1.0000e-05\n",
"Epoch 39/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.8406 - loss: 0.4233 - val_accuracy: 0.8475 - val_loss: 0.4530 - learning_rate: 1.0000e-05\n",
"Epoch 40/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8276 - loss: 0.4257 - val_accuracy: 0.8450 - val_loss: 0.4463 - learning_rate: 1.0000e-05\n",
"Epoch 41/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8446 - loss: 0.4167 - val_accuracy: 0.8500 - val_loss: 0.4663 - learning_rate: 1.0000e-05\n",
"Epoch 42/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8445 - loss: 0.4211 - val_accuracy: 0.8350 - val_loss: 0.4317 - learning_rate: 1.0000e-05\n",
"Epoch 43/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 39ms/step - accuracy: 0.8697 - loss: 0.3649 - val_accuracy: 0.8350 - val_loss: 0.4538 - learning_rate: 1.0000e-05\n",
"Epoch 44/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 41ms/step - accuracy: 0.8651 - loss: 0.3714 - val_accuracy: 0.8525 - val_loss: 0.4089 - learning_rate: 1.0000e-05\n",
"Epoch 45/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8670 - loss: 0.3753 - val_accuracy: 0.8675 - val_loss: 0.4030 - learning_rate: 1.0000e-05\n",
"Epoch 46/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 34ms/step - accuracy: 0.8578 - loss: 0.3713 - val_accuracy: 0.8625 - val_loss: 0.4232 - learning_rate: 1.0000e-05\n",
"Epoch 47/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8739 - loss: 0.3369 - val_accuracy: 0.8775 - val_loss: 0.3967 - learning_rate: 1.0000e-05\n",
"Epoch 48/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 34ms/step - accuracy: 0.8895 - loss: 0.3170 - val_accuracy: 0.8700 - val_loss: 0.3946 - learning_rate: 1.0000e-05\n",
"Epoch 49/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 36ms/step - accuracy: 0.8870 - loss: 0.3139 - val_accuracy: 0.8625 - val_loss: 0.3739 - learning_rate: 1.0000e-05\n",
"Epoch 50/50\n",
"\u001b[1m50/50\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 40ms/step - accuracy: 0.9062 - loss: 0.2966 - val_accuracy: 0.8800 - val_loss: 0.3581 - learning_rate: 1.0000e-05\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Post-hoc calibration with temperature scaling\n",
"scaled_logits = model.output / 2.0 # Temperature scaling factor\n",
"calibrated_outputs = tf.keras.layers.Activation('softmax')(scaled_logits)\n",
"calibrated_model = tf.keras.Model(inputs=model.input, outputs=calibrated_outputs)\n",
"\n",
"# Save calibrated model and encoders\n",
"calibrated_model.save(\"improved_emotion_model.keras\")\n",
"joblib.dump(label_encoder, 'label_encoder.pkl')\n",
"joblib.dump(speaker_encoder, 'speaker_encoder.pkl')"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "EbxtL-GIHYCV",
"outputId": "8622160b-6575-4e7b-b161-6204c4efa38b"
},
"execution_count": 16,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"['speaker_encoder.pkl']"
]
},
"metadata": {},
"execution_count": 16
}
]
},
{
"cell_type": "code",
"source": [
"# === Gradio Interface ===\n",
"def convert_mp3_to_wav(mp3_filepath):\n",
" \"\"\"Convert MP3 to WAV for processing.\"\"\"\n",
" wav_filepath = mp3_filepath.replace('.mp3', '.wav')\n",
" sound = AudioSegment.from_mp3(mp3_filepath)\n",
" sound.export(wav_filepath, format=\"wav\")\n",
" return wav_filepath"
],
"metadata": {
"id": "p0swcOtTHdHE"
},
"execution_count": 17,
"outputs": []
},
{
"cell_type": "code",
"source": [
"def predict_emotion(audio_file):\n",
" \"\"\"Predict emotion from audio input using the trained model.\"\"\"\n",
" if audio_file.lower().endswith('.mp3'):\n",
" audio_file = convert_mp3_to_wav(audio_file)\n",
" audio = load_and_preprocess_audio(audio_file)\n",
" features = extract_features(audio)\n",
" speaker_id = np.array([0]) # Default speaker ID for inference\n",
" pred = calibrated_model.predict([np.expand_dims(features, 0), speaker_id.reshape(-1, 1)])\n",
" return label_encoder.inverse_transform([np.argmax(pred)])[0]"
],
"metadata": {
"id": "xvC9KHe7HjYZ"
},
"execution_count": 18,
"outputs": []
},
{
"cell_type": "code",
"source": [
"iface = gr.Interface(\n",
" fn=predict_emotion,\n",
" inputs=gr.Audio(type=\"filepath\"),\n",
" outputs=\"text\",\n",
" title=\"Enhanced Emotion Recognition\",\n",
" description=\"Record or upload audio (MP3/WAV) for emotion prediction with improved accuracy\"\n",
")\n"
],
"metadata": {
"id": "H0tGQXqepvYe"
},
"execution_count": 21,
"outputs": []
},
{
"cell_type": "code",
"source": [
"iface.launch()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 646
},
"id": "1WxOHPpypv_c",
"outputId": "78d7614d-00dc-4e42-c3f2-6376d4fa71b3"
},
"execution_count": 22,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Running Gradio in a Colab notebook requires sharing enabled. Automatically setting `share=True` (you can turn this off by setting `share=False` in `launch()` explicitly).\n",
"\n",
"Colab notebook detected. To show errors in colab notebook, set debug=True in launch()\n",
"* Running on public URL: https://962c8bd414079ad9b5.gradio.live\n",
"\n",
"This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from the terminal in the working directory to deploy to Hugging Face Spaces (https://huggingface.co/spaces)\n"
]
},
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"text/html": [
""
]
},
"metadata": {}
},
{
"output_type": "execute_result",
"data": {
"text/plain": []
},
"metadata": {},
"execution_count": 22
}
]
},
{
"cell_type": "code",
"source": [
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"from sklearn.metrics import classification_report, confusion_matrix"
],
"metadata": {
"id": "LC2BZUj-tsoz"
},
"execution_count": 23,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# Post-hoc calibration with temperature scaling\n",
"scaled_logits = model.output / 2.0 # Temperature scaling factor\n",
"calibrated_outputs = tf.keras.layers.Activation('softmax')(scaled_logits)\n",
"calibrated_model = tf.keras.Model(inputs=model.input, outputs=calibrated_outputs)\n",
"\n",
"# Re-compile the calibrated model with appropriate loss and metrics\n",
"calibrated_model.compile(optimizer=model.optimizer, # Use the same optimizer\n",
" loss=tf.keras.losses.CategoricalCrossentropy(), # Without from_logits=True\n",
" metrics=['accuracy']) # Or any other desired metrics"
],
"metadata": {
"id": "0A4Wphz1t-8Q"
},
"execution_count": 26,
"outputs": []
},
{
"cell_type": "code",
"source": [
"# --- Evaluation ---\n",
"\n",
"# Evaluate the calibrated model on the test set\n",
"test_loss, test_accuracy = calibrated_model.evaluate([X_test, speaker_test], y_test)\n",
"print(\"Test Loss: {:.4f}\".format(test_loss))\n",
"print(\"Test Accuracy: {:.4f}\".format(test_accuracy))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "EtsSQ1sQuc_q",
"outputId": "c92b354e-6052-4b3a-df45-df64a3a41d1e"
},
"execution_count": 27,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"\u001b[1m13/13\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 23ms/step - accuracy: 0.8879 - loss: 0.5456\n",
"Test Loss: 0.5546\n",
"Test Accuracy: 0.8800\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Generate predictions on the test set\n",
"y_pred = calibrated_model.predict([X_test, speaker_test])\n",
"y_pred_classes = np.argmax(y_pred, axis=1)\n",
"y_true_classes = np.argmax(y_test, axis=1)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "X7JQHoVCufWR",
"outputId": "195bafd0-6179-413a-ef29-a0a1f17a1fd5"
},
"execution_count": 28,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"\u001b[1m13/13\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 39ms/step\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Print the classification report\n",
"print(\"Classification Report:\")\n",
"print(classification_report(y_true_classes, y_pred_classes, target_names=label_encoder.classes_))\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "uyaMW5WeukW-",
"outputId": "8c8a7c4c-05f9-4a56-c0c2-b794c4cb9e9d"
},
"execution_count": 29,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Classification Report:\n",
" precision recall f1-score support\n",
"\n",
" YAF_angry 0.94 0.91 0.92 80\n",
" YAF_fear 0.92 0.96 0.94 80\n",
" YAF_happy 0.83 0.85 0.84 80\n",
" YAF_neutral 0.85 0.96 0.90 80\n",
" YAF_sad 0.88 0.71 0.79 80\n",
"\n",
" accuracy 0.88 400\n",
" macro avg 0.88 0.88 0.88 400\n",
"weighted avg 0.88 0.88 0.88 400\n",
"\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# Plot training & validation accuracy and loss over epochs\n",
"plt.figure(figsize=(12, 5))\n",
"\n",
"# Accuracy Plot\n",
"plt.subplot(1, 2, 1)\n",
"plt.plot(history.history['accuracy'], label='Train Accuracy')\n",
"plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
"plt.title('Training and Validation Accuracy')\n",
"plt.xlabel('Epoch')\n",
"plt.ylabel('Accuracy')\n",
"plt.legend()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 504
},
"id": "g6NP3PuxumPf",
"outputId": "03722745-7562-491b-b21c-10d450ac38af"
},
"execution_count": 30,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
""
]
},
"metadata": {},
"execution_count": 30
},
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"image/png": "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\n"
},
"metadata": {}
}
]
},
{
"cell_type": "code",
"source": [
"# Loss Plot\n",
"plt.subplot(1, 2, 2)\n",
"plt.plot(history.history['loss'], label='Train Loss')\n",
"plt.plot(history.history['val_loss'], label='Validation Loss')\n",
"plt.title('Training and Validation Loss')\n",
"plt.xlabel('Epoch')\n",
"plt.ylabel('Loss')\n",
"plt.legend()\n",
"\n",
"plt.tight_layout()\n",
"plt.show()\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 487
},
"id": "nFTp6UPRuqTP",
"outputId": "071ef52d-ee44-4fd5-aba8-884ca2d22111"
},
"execution_count": 31,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"image/png": "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\n"
},
"metadata": {}
}
]
},
{
"cell_type": "code",
"source": [
"# Plot Confusion Matrix\n",
"cm = confusion_matrix(y_true_classes, y_pred_classes)\n",
"plt.figure(figsize=(10, 8))\n",
"sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\",\n",
" xticklabels=label_encoder.classes_,\n",
" yticklabels=label_encoder.classes_)\n",
"plt.title(\"Confusion Matrix\")\n",
"plt.xlabel(\"Predicted Label\")\n",
"plt.ylabel(\"True Label\")\n",
"plt.show()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 718
},
"id": "CA8crbNpuujn",
"outputId": "4f48f705-fb31-4c0c-c2bd-762e9e69d940"
},
"execution_count": 32,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"image/png": "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\n"
},
"metadata": {}
}
]
},
{
"cell_type": "code",
"source": [
"import matplotlib.pyplot as plt\n",
"from sklearn.metrics import precision_recall_curve, average_precision_score\n",
"\n",
"# Assuming y_test is one-hot encoded and y_pred are the predicted probabilities.\n",
"n_classes = y_test.shape[1]\n",
"\n",
"plt.figure(figsize=(10, 8))\n",
"for i in range(n_classes):\n",
" precision, recall, _ = precision_recall_curve(y_test[:, i], y_pred[:, i])\n",
" average_precision = average_precision_score(y_test[:, i], y_pred[:, i])\n",
" plt.plot(recall, precision, lw=2,\n",
" label=f'{label_encoder.classes_[i]} (AP = {average_precision:.2f})')\n",
"\n",
"plt.xlabel('Recall')\n",
"plt.ylabel('Precision')\n",
"plt.title('Precision-Recall Curve per Class')\n",
"plt.legend(loc='best')\n",
"plt.tight_layout()\n",
"plt.show()\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 807
},
"id": "QnUKDeLevPaf",
"outputId": "13826b63-ca66-4d8c-ba04-104c9c55a521"
},
"execution_count": 34,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"image/png": "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\n"
},
"metadata": {}
}
]
}
]
}