Add emotion recognition models (CNN, TFLite, TFJS)

README.md

@@ -1,3 +1,115 @@
 ---
 license: mit
+tags:
+  - image-classification
+  - tensorflow
+  - keras
+  - emotion-recognition
+  - facial-expression
+  - cnn
+  - tflite
+  - tfjs
+datasets:
+  - samithsachidanandan/human-face-emotions
+metrics:
+  - accuracy
+pipeline_tag: image-classification
+library_name: keras
 ---
+
+# Human Emotion Recognition
+
+Deep learning models for classifying human facial emotions.
+
+## Emotion Classes
+- 😠 Angry
+- 😨 Fear  
+- 😊 Happy
+- 😢 Sad
+- 😲 Surprise
+
+## Model Performance
+
+| Model | Test Accuracy | Test Loss | Epochs |
+|-------|---------------|-----------|--------|
+| **Base CNN** | **92.41%** | 0.268 | 33 |
+| MobileNetV3Small | 81.56% | 0.551 | 50 |
+
+> 🏆 Best model: **Base CNN** with 92.41% test accuracy
+
+## Models
+
+| File | Format | Input Size | Description |
+|------|--------|------------|-------------|
+| `model_base.h5` | Keras H5 | 128x128x1 | Custom CNN (Grayscale) |
+| `model_transfer_learning.keras` | Keras | 224x224x3 | MobileNetV3Small (RGB) |
+| `tflite/best_model.tflite` | TFLite | 128x128x1 | Mobile/Edge |
+| `tfjs_model/` | TF.js | 128x128x1 | Web deployment |
+
+## Usage
+
+### Python
+
+```python
+from huggingface_hub import hf_hub_download
+import tensorflow as tf
+import numpy as np
+
+# Download model
+model_path = hf_hub_download(
+    repo_id="dafisnadhif/human-emotion-recognition",
+    filename="model_base.h5"
+)
+
+# Load model
+model = tf.keras.models.load_model(model_path)
+
+# Predict
+CLASS_NAMES = ['Angry', 'Fear', 'Happy', 'Sad', 'Surprise']
+predictions = model.predict(img_batch)
+print(CLASS_NAMES[np.argmax(predictions[0])])
+```
+
+### TensorFlow Lite
+
+```python
+from huggingface_hub import hf_hub_download
+import tensorflow as tf
+
+tflite_path = hf_hub_download(
+    repo_id="dafisnadhif/human-emotion-recognition",
+    filename="tflite/best_model.tflite"
+)
+
+interpreter = tf.lite.Interpreter(model_path=tflite_path)
+interpreter.allocate_tensors()
+```
+
+### TensorFlow.js
+
+```javascript
+const model = await tf.loadLayersModel(
+  'https://huggingface.co/dafisnadhif/human-emotion-recognition/resolve/main/tfjs_model/model.json'
+);
+```
+
+## Training Details
+
+| Parameter | Value |
+|-----------|-------|
+| **Dataset** | [Human Face Emotions](https://www.kaggle.com/datasets/samithsachidanandan/human-face-emotions) |
+| **Images** | ~47,000 facial images |
+| **Source Code** | [GitHub](https://github.com/DafisNadhifSaputra/human-emotion-recognition) |
+| **Framework** | TensorFlow 2.x / Keras |
+| **Optimizer** | AdamW (lr=1e-3, weight_decay=1e-4) |
+| **Loss** | Sparse Categorical Crossentropy |
+| **Batch Size** | 256 |
+| **Callbacks** | EarlyStopping (patience=8), ReduceLROnPlateau |
+
+## License
+
+MIT License
+
+## Author
+
+Dafis Nadhif Saputra

model_base.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:964e54bed5f53efddb1dbcfaebdc16f1909b4b9961aaee1c87003cd4f78a70ac
+size 13376000

tfjs_model/group1-shard1of2.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3c58be73182d8a009b235d8ff6c8befc0a3748fc6bc988374fd446faafa4780
+size 4194304

tfjs_model/group1-shard2of2.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7ea3f59763dc3c26d4eeda0f6bb9398bf1124b571b0b06da1ffa52b53132f79
+size 235796

tfjs_model/model.json

@@ -0,0 +1 @@
+{"format": "layers-model", "generatedBy": "keras v3.8.0", "convertedBy": "TensorFlow.js Converter v4.22.0", "modelTopology": {"keras_version": "3.8.0", "backend": "tensorflow", "model_config": {"class_name": "Sequential", "config": {"name": "sequential_33", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "layers": [{"class_name": "InputLayer", "config": {"batch_shape": [null, 128, 128, 1], "dtype": "float32", "sparse": false, "name": "input_layer_56"}}, {"class_name": "Conv2D", "config": {"name": "conv2d_28", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "filters": 32, "kernel_size": [3, 3], "strides": [1, 1], "padding": "same", "data_format": "channels_last", "dilation_rate": [1, 1], "groups": 1, "activation": "linear", "use_bias": true, "kernel_initializer": {"module": "keras.initializers", "class_name": "HeNormal", "config": {"seed": null}, "registered_name": null}, "bias_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "BatchNormalization", "config": {"name": "batch_normalization_37", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "axis": -1, "momentum": 0.99, "epsilon": 0.001, "center": true, "scale": true, "beta_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "gamma_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "moving_mean_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "moving_variance_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "beta_regularizer": null, "gamma_regularizer": null, "beta_constraint": null, "gamma_constraint": null, "synchronized": false}}, {"class_name": "Activation", "config": {"name": "activation_313", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "activation": "relu"}}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d_28", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "pool_size": [2, 2], "padding": "valid", "strides": [2, 2], "data_format": "channels_last"}}, {"class_name": "Dropout", "config": {"name": "dropout_47", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "rate": 0.1, "seed": null, "noise_shape": null}}, {"class_name": "Conv2D", "config": {"name": "conv2d_29", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "filters": 64, "kernel_size": [3, 3], "strides": [1, 1], "padding": "same", "data_format": "channels_last", "dilation_rate": [1, 1], "groups": 1, "activation": "linear", "use_bias": true, "kernel_initializer": {"module": "keras.initializers", "class_name": "HeNormal", "config": {"seed": null}, "registered_name": null}, "bias_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "BatchNormalization", "config": {"name": "batch_normalization_38", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "axis": -1, "momentum": 0.99, "epsilon": 0.001, "center": true, "scale": true, "beta_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "gamma_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "moving_mean_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "moving_variance_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "beta_regularizer": null, "gamma_regularizer": null, "beta_constraint": null, "gamma_constraint": null, "synchronized": false}}, {"class_name": "Activation", "config": {"name": "activation_314", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "activation": "relu"}}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d_29", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "pool_size": [2, 2], "padding": "valid", "strides": [2, 2], "data_format": "channels_last"}}, {"class_name": "Conv2D", "config": {"name": "conv2d_30", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "filters": 64, "kernel_size": [3, 3], "strides": [1, 1], "padding": "same", "data_format": "channels_last", "dilation_rate": [1, 1], "groups": 1, "activation": "linear", "use_bias": true, "kernel_initializer": {"module": "keras.initializers", "class_name": "HeNormal", "config": {"seed": null}, "registered_name": null}, "bias_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "BatchNormalization", "config": {"name": "batch_normalization_39", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "axis": -1, "momentum": 0.99, "epsilon": 0.001, "center": true, "scale": true, "beta_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "gamma_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "moving_mean_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "moving_variance_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "beta_regularizer": null, "gamma_regularizer": null, "beta_constraint": null, "gamma_constraint": null, "synchronized": false}}, {"class_name": "Activation", "config": {"name": "activation_315", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "activation": "relu"}}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d_30", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "pool_size": [4, 4], "padding": "valid", "strides": [4, 4], "data_format": "channels_last"}}, {"class_name": "Dropout", "config": {"name": "dropout_48", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "rate": 0.25, "seed": null, "noise_shape": null}}, {"class_name": "Flatten", "config": {"name": "flatten_12", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "data_format": "channels_last"}}, {"class_name": "Dense", "config": {"name": "dense_62", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "units": 256, "activation": "linear", "use_bias": true, "kernel_initializer": {"module": "keras.initializers", "class_name": "GlorotUniform", "config": {"seed": null}, "registered_name": null}, "bias_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "kernel_regularizer": null, "bias_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "BatchNormalization", "config": {"name": "batch_normalization_40", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "axis": -1, "momentum": 0.99, "epsilon": 0.001, "center": true, "scale": true, "beta_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "gamma_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "moving_mean_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "moving_variance_initializer": {"module": "keras.initializers", "class_name": "Ones", "config": {}, "registered_name": null}, "beta_regularizer": null, "gamma_regularizer": null, "beta_constraint": null, "gamma_constraint": null, "synchronized": false}}, {"class_name": "Activation", "config": {"name": "activation_316", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "activation": "relu"}}, {"class_name": "Dropout", "config": {"name": "dropout_49", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "rate": 0.5, "seed": null, "noise_shape": null}}, {"class_name": "Dense", "config": {"name": "dense_63", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "units": 5, "activation": "softmax", "use_bias": true, "kernel_initializer": {"module": "keras.initializers", "class_name": "GlorotUniform", "config": {"seed": null}, "registered_name": null}, "bias_initializer": {"module": "keras.initializers", "class_name": "Zeros", "config": {}, "registered_name": null}, "kernel_regularizer": null, "bias_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}], "build_input_shape": [null, 128, 128, 1]}}, "training_config": {"loss": "sparse_categorical_crossentropy", "loss_weights": null, "metrics": ["accuracy"], "weighted_metrics": null, "run_eagerly": false, "steps_per_execution": 1, "jit_compile": true, "optimizer_config": {"class_name": "AdamW", "config": {"name": "adamw", "learning_rate": 1.0000001111620804e-06, "weight_decay": 0.0001, "clipnorm": null, "global_clipnorm": null, "clipvalue": null, "use_ema": false, "ema_momentum": 0.99, "ema_overwrite_frequency": null, "loss_scale_factor": null, "gradient_accumulation_steps": null, "beta_1": 0.9, "beta_2": 0.999, "epsilon": 1e-07, "amsgrad": false}}}}, "weightsManifest": [{"paths": ["group1-shard1of2.bin", "group1-shard2of2.bin"], "weights": [{"name": "sequential_33/batch_normalization_37/gamma", "shape": [32], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_37/beta", "shape": [32], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_37/moving_mean", "shape": [32], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_37/moving_variance", "shape": [32], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_38/gamma", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_38/beta", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_38/moving_mean", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_38/moving_variance", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_39/gamma", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_39/beta", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_39/moving_mean", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_39/moving_variance", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_40/gamma", "shape": [256], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_40/beta", "shape": [256], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_40/moving_mean", "shape": [256], "dtype": "float32"}, {"name": "sequential_33/batch_normalization_40/moving_variance", "shape": [256], "dtype": "float32"}, {"name": "sequential_33/conv2d_28/kernel", "shape": [3, 3, 1, 32], "dtype": "float32"}, {"name": "sequential_33/conv2d_28/bias", "shape": [32], "dtype": "float32"}, {"name": "sequential_33/conv2d_29/kernel", "shape": [3, 3, 32, 64], "dtype": "float32"}, {"name": "sequential_33/conv2d_29/bias", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/conv2d_30/kernel", "shape": [3, 3, 64, 64], "dtype": "float32"}, {"name": "sequential_33/conv2d_30/bias", "shape": [64], "dtype": "float32"}, {"name": "sequential_33/dense_62/kernel", "shape": [4096, 256], "dtype": "float32"}, {"name": "sequential_33/dense_62/bias", "shape": [256], "dtype": "float32"}, {"name": "sequential_33/dense_63/kernel", "shape": [256, 5], "dtype": "float32"}, {"name": "sequential_33/dense_63/bias", "shape": [5], "dtype": "float32"}]}]}

tflite/best_model.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:988615a6c6efe100a6f204282c7b9a3d28bd2f3eb71f903c2a19a558970a3879
+size 1117480