Delete huggingface-space

huggingface-space/.dvc/.gitignore

@@ -1,2 +0,0 @@
-/config.local
-/tmp

huggingface-space/.gitattributes

@@ -1 +0,0 @@
-sota_model/model.safetensors filter=lfs diff=lfs merge=lfs -text

huggingface-space/.gitignore

@@ -1,112 +0,0 @@
-# MLOps & Data Science Artifacts
-# -------------------------------------------------------------------
-# Ignore all data, models, and artifacts. These should be tracked by DVC.
-/artifacts/
-/data/
-/sota_model/
-# Ignore the DVC local cache. This is where the actual data files are stored.
-.dvc/cache
-
-# Ignore MLflow experiment tracking output
-/mlruns/
-
-# Ignore logs
-/logs/
-*.log
-
-# Ignore common model file extensions, just in case
-*.h5
-*.pkl
-*.model
-*.onnx
-
-
-# Python Virtual Environments
-# -------------------------------------------------------------------
-/venv/
-/myenv/
-/.venv/
-/env/
-/ENV/
-*/.venv/
-*/venv/
-*/myenv/
-
-
-# Python Byte-code and Caches
-# -------------------------------------------------------------------
-__pycache__/
-*.py[cod]
-*$py.class
-
-
-# Python Packaging & Distribution
-# -------------------------------------------------------------------
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-*.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
-
-
-# IDE and Editor Configuration
-# -------------------------------------------------------------------
-# PyCharm
-.idea/
-
-# Visual Studio Code (allow sharing of recommended extensions)
-.vscode/*
-!.vscode/extensions.json
-
-# Sublime Text
-*.sublime-project
-*.sublime-workspace
-
-
-# Secrets and Environment Variables
-# -------------------------------------------------------------------
-# NEVER commit secrets or environment variables
-.env
-*.env
-secrets.yaml
-secrets.json
-
-
-# Operating System Files
-# -------------------------------------------------------------------
-# macOS
-.DS_Store
-
-# Windows
-Thumbs.db
-desktop.ini
-
-
-# Jupyter Notebook Checkpoints
-# -------------------------------------------------------------------
-.ipynb_checkpoints/
-
-
-# Other
-# -------------------------------------------------------------------
-# Temporary files
-*.tmp
-*.bak
-*.swp
-
-.env
-*.env
-secrets.yaml
-secrets.json
-processed_video.mp4

huggingface-space/LICENSE

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) 2025 ALYYAN
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

huggingface-space/README.md

@@ -1,53 +0,0 @@
----
-title: Facial Emotion Detector
-emoji: 🎭
-colorFrom: purple
-colorTo: indigo
-sdk: gradio
-sdk_version: "3.50.2"
-app_file: app.py
-pinned: false
----
-
-# 🎭 End-to-End Facial Emotion Recognition
-
- <!-- Replace with a link to your final app screenshot -->
-
-This repository contains a complete, end-to-end MLOps pipeline and a production-ready web application for real-time facial emotion recognition. The project leverages a state-of-the-art Vision Transformer model and is deployed as a user-friendly Gradio application on Hugging Face Spaces.
-
-**Live Demo:** [🚀 Click here to try the application on Hugging Face Spaces!](https://huggingface.co/spaces/ALYYAN/Emotion-Recognition) <!-- Replace with your HF Space URL -->
-
----
-
-## ✨ Features
-
--   **Real-time Emotion Detection:** Analyzes your webcam feed to predict emotions in real-time.
--   **High Accuracy:** Powered by a pre-trained Swin Transformer model fine-tuned on the massive AffectNet dataset for superior performance on "in the wild" faces.
--   **Static Image & Video Analysis:** Upload your own images or videos for emotion prediction.
--   **Polished UI:** A professional and responsive user interface with an animated background, built with Gradio.
--   **Reproducible MLOps Pipeline:** The entire model training and data processing workflow is managed by DVC, ensuring 100% reproducibility.
--   **Containerized for Deployment:** The application is packaged with Docker for easy and consistent deployment anywhere.
-
-## 🛠️ Tech Stack
-
--   **Model:** Swin Transformer (`PangPang/affectnet-swin-tiny-patch4-window7-224`)
--   **ML/Ops:** Python, TensorFlow/Keras, DVC, MLflow, Hugging Face `transformers`
--   **Backend & UI:** Gradio
--   **Face Detection:** MTCNN
--   **Deployment:** Hugging Face Spaces, Docker
-
-## 🚀 Getting Started
-
-Follow these steps to run the project locally.
-
-### Prerequisites
-
--   Python 3.10+
--   Git and Git LFS ([installation guide](https://git-lfs.github.com))
--   An NVIDIA GPU with CUDA drivers is recommended for the training pipeline, but the deployed app runs on CPU.
-
-### 1. Clone the Repository
-
-```bash
-git clone https://github.com/YOUR-USERNAME/Emotion-Recognition-MLOps.git
-cd Emotion-Recognition-MLOps

huggingface-space/app.py

@@ -1,181 +0,0 @@
-import gradio as gr
-import os
-import cv2
-import time
-
-# Ensure the correct predictor class is imported
-from src.EmotionRecognition.pipeline.hf_predictor import HFPredictor
-
-# --- INITIALIZE THE MODEL ---
-print("[INFO] Initializing predictor...")
-try:
-    predictor = HFPredictor()
-    print("[INFO] Predictor initialized successfully.")
-except Exception as e:
-    predictor = None
-    print(f"[FATAL ERROR] Failed to initialize predictor: {e}")
-
-# --- UI CONTENT & STYLING ---
-CSS = """
-/* Animated Gradient Background */
-body {
-    background: linear-gradient(-45deg, #0b0f19, #131a2d, #2a2a72, #522a72);
-    background-size: 400% 400%;
-    animation: gradient 15s ease infinite;
-}
-@keyframes gradient { 0% { background-position: 0% 50%; } 50% { background-position: 100% 50%; } 100% { background-position: 0% 50%; } }
-
-/* General Layout & Typography */
-.gradio-container { max-width: 1320px !important; margin: auto !important; }
-#title { text-align: center; font-size: 3rem !important; font-weight: 700; color: #FFF; margin-bottom: 0.5rem; }
-#subtitle { text-align: center; color: #bebebe; margin-top: 0; margin-bottom: 40px; font-size: 1.2rem; font-weight: 300; }
-.gr-button { font-weight: bold !important; }
-
-/* Main Content Card */
-#main-card {
-    background: rgba(22, 22, 34, 0.65);
-    border-radius: 16px;
-    box-shadow: 0 8px 32px 0 rgba(0, 0, 0, 0.37);
-    backdrop-filter: blur(12px); -webkit-backdrop-filter: blur(12px);
-    border: 1px solid rgba(255, 255, 255, 0.18);
-    padding: 1rem;
-}
-
-/* Prediction Bar Styling */
-#predictions-column { background-color: transparent !important; padding: 1.5rem; }
-#predictions-column > .gr-label { display: none; }
-.prediction-list { list-style-type: none; padding: 0; margin-top: 1.5rem; }
-.prediction-list li { display: flex; align-items: center; margin-bottom: 12px; font-size: 1.1rem; }
-.prediction-list .label { width: 100px; text-transform: capitalize; color: #e0e0e0; }
-.prediction-list .bar-container { flex-grow: 1; height: 24px; background-color: rgba(255,255,255,0.1); border-radius: 12px; margin: 0 15px; overflow: hidden; }
-.prediction-list .bar { height: 100%; background: linear-gradient(90deg, #8A2BE2, #C71585); border-radius: 12px; transition: width: 0.1s linear; }
-.prediction-list .percent { width: 60px; text-align: right; font-weight: bold; color: #FFF; }
-footer { display: none !important; }
-"""
-
-ABOUT_MARKDOWN = """
-## 🚀 About This Project
-
-This application is the culmination of a complete, end-to-end MLOps project, demonstrating the full lifecycle from research and experimentation to a final, deployed, state-of-the-art solution.
-
-**💻 [View Project on GitHub](https://github.com/YOUR-USERNAME/Emotion-Recognition-MLOps)** <!--- REPLACE WITH YOUR GITHUB REPO LINK --->
-
----
-
-### Key Technical Features:
-
-*   **State-of-the-Art AI Model:** Utilizes a **Swin Transformer**, a powerful Vision Transformer (ViT) architecture, pre-trained on the massive **AffectNet** dataset. This ensures high accuracy and robust generalization to real-world, "in the wild" facial expressions.
-*   **Reproducible MLOps Pipeline:** The original model training and data processing workflows were built using **DVC (Data Version Control)**, ensuring that every experiment is versioned and reproducible.
-*   **Full-Stack & Deployment:** The application architecture evolved from a Python-only script to a decoupled **FastAPI backend** and a **React frontend**, and was ultimately deployed as this streamlined and robust **Gradio** application.
-*   **Containerized & Automated:** The entire application is packaged with **Docker** and is set up for **CI/CD with GitHub Actions**, enabling automated testing and deployment to cloud platforms like Hugging Face Spaces.
-
----
-
-### 🛠️ Architecture & Tech Stack
-
-*   **Machine Learning & CV:** Python, PyTorch, Hugging Face `transformers`, MTCNN, OpenCV
-*   **MLOps & DevOps:** DVC, GitHub Actions, Docker, Git LFS
-*   **Application & UI:** Gradio
-
-"""
-
-# --- BACKEND LOGIC ---
-
-def create_prediction_html(probabilities):
-    """Generates clean HTML for the prediction bars."""
-    if not probabilities:
-        return "<div style='padding: 2rem; text-align: center; color: #999;'>Waiting for prediction...</div>"
-    html = "<ul class='prediction-list'>"
-    sorted_preds = sorted(probabilities.items(), key=lambda item: item[1], reverse=True)
-    for emotion, prob in sorted_preds:
-        html += f"""
-        <li>
-            <strong class='label'>{emotion}</strong>
-            <div class='bar-container'><div class='bar' style='width: {prob*100:.1f}%;'></div></div>
-            <span class='percent'>{(prob*100):.1f}%</span>
-        </li>
-        """
-    html += "</ul>"
-    return html
-
-def unified_prediction_function(frame):
-    """A single, robust function that takes any frame (from webcam or upload) and returns the annotated frame and the prediction HTML."""
-    if frame is None:
-        return None, create_prediction_html({})
-    
-    # The predictor class handles all annotation and prediction logic
-    annotated_frame, probabilities = predictor.process_frame(frame)
-    
-    return annotated_frame, create_prediction_html(probabilities)
-
-def process_video(video_path, progress=gr.Progress(track_tqdm=True)):
-    """Processes an uploaded video file frame-by-frame."""
-    if video_path is None: 
-        return None
-    try:
-        cap = cv2.VideoCapture(video_path)
-        frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
-        output_path = "processed_video.mp4"
-        fourcc = cv2.VideoWriter_fourcc(*'mp4v')
-        fps = cap.get(cv2.CAP_PROP_FPS)
-        width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-        height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-        out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))
-        for _ in progress.tqdm(range(frame_count), desc="Processing Video"):
-            ret, frame = cap.read()
-            if not ret: break
-            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            annotated_frame, _ = predictor.process_frame(frame_rgb)
-            if annotated_frame is not None:
-                out.write(cv2.cvtColor(annotated_frame, cv2.COLOR_RGB2BGR))
-        cap.release()
-        out.release()
-        return output_path
-    except Exception as e:
-        print(f"[ERROR] Video processing failed: {e}")
-        return None
-
-# --- GRADIO UI ---
-with gr.Blocks(css=CSS, theme=gr.themes.Base()) as demo:
-    gr.Markdown("# Facial Emotion Detector", elem_id="title")
-    gr.Markdown("A real-time AI application powered by Vision Transformers", elem_id="subtitle")
-
-    with gr.Box(elem_id="main-card"):
-        with gr.Tabs():
-            with gr.TabItem("Live Detection"):
-                with gr.Row(equal_height=False):
-                    with gr.Column(scale=3):
-                        # The single, correct component for a live webcam feed.
-                        live_feed = gr.Image(source="webcam", streaming=True, type="numpy", label="Live Feed", height=550, mirror_webcam=True)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        gr.Markdown("### Emotion Probabilities")
-                        live_predictions = gr.HTML()
-            
-            with gr.TabItem("Upload Image"):
-                with gr.Row(equal_height=False):
-                    with gr.Column(scale=3):
-                        image_input = gr.Image(type="numpy", label="Upload an Image", height=550)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        image_predictions = gr.HTML()
-                image_button = gr.Button("Analyze Image", variant="primary")
-
-            with gr.TabItem("Upload Video"):
-                with gr.Row(equal_height=False):
-                    video_input = gr.Video(label="Upload a Video File")
-                    video_output = gr.Video(label="Processed Video")
-                video_button = gr.Button("Analyze Video", variant="primary")
-            
-            with gr.TabItem("About"):
-                gr.Markdown(ABOUT_MARKDOWN)
-
-    # --- EVENT LISTENERS ---
-    live_feed.stream(fn=unified_prediction_function, inputs=live_feed, outputs=[live_feed, live_predictions])
-    image_button.click(fn=unified_prediction_function, inputs=[image_input], outputs=[image_input, image_predictions])
-    video_button.click(fn=process_video, inputs=[video_input], outputs=[video_output])
-
-# --- LAUNCH THE APP ---
-if predictor:
-    # Enabling the queue is essential for the video processing progress bar.
-    demo.queue().launch(debug=True)
-else:
-    print("\n[FATAL ERROR] Could not start the application.")

huggingface-space/config/config.yaml

@@ -1,25 +0,0 @@
-artifacts_root: artifacts
-
-data_preparation: # This is our Stage 1
-  root_dir: artifacts/data_preparation
-  # Inputs from raw data
-  ferplus_pixels_csv: data/raw/fer2013.csv
-  ferplus_labels_csv: data/raw/fer2013new.csv
-  ckplus_dir: data/raw/CK+48
-  # Outputs
-  combined_train_dir: artifacts/data_preparation/train
-  ferplus_test_dir: artifacts/data_preparation/test
-
-model_trainer:
-  root_dir: artifacts/training
-  # The trainer now takes its input directly from the preparation stage
-  train_data_dir: artifacts/data_preparation/train
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-
-model_evaluation:
-  root_dir: artifacts/evaluation
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-  metrics_file_name: artifacts/evaluation/metrics.json
-  mlflow_uri: https://dagshub.com/AlyyanAhmed21/Emotion-Recognition-MLOps.mlflow # Example for DagsHub

huggingface-space/dvc.lock

@@ -1,203 +0,0 @@
-schema: '2.0'
-stages:
-  data_validation:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_validation.py
-    deps:
-    - path: artifacts/data_ingestion
-      hash: md5
-      md5: 9208f64defb6697b78bab62e943d955d.dir
-      size: 302675528
-      nfiles: 2
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: dacf4230e18681185b786aa280cdec5e
-      size: 4275
-    - path: src/EmotionRecognition/pipeline/stage_02_data_validation.py
-      hash: md5
-      md5: 18a3d78c83dc5b278e14523077035e41
-      size: 1141
-    outs:
-    - path: artifacts/data_validation/status.txt
-      hash: md5
-      md5: 86e6a2f694c57a675b3e2da6b95ff9ba
-      size: 23
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-    - path: data/raw/CK+48
-      hash: md5
-      md5: a1559eddfd0d86b541e5df18b4b8205e.dir
-      size: 1715162
-      nfiles: 981
-    - path: data/raw/fer2013.csv
-      hash: md5
-      md5: f8428a1edbd21e88f42c73edd2a14f95
-      size: 301072766
-    - path: data/raw/fer2013new.csv
-      hash: md5
-      md5: 413eba86d6e454536b99705b8c7fc5c5
-      size: 1602762
-    - path: src/EmotionRecognition/components/data_preparation.py
-      hash: md5
-      md5: 228140227aaedb9f07b4c00462f267c6
-      size: 5776
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: 8786c8d41e2e50a49b4ca6d5bf59ad44
-      size: 2910
-    - path: src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      hash: md5
-      md5: 1a324b8f1cf01e4e60e0a8529b23b577
-      size: 1110
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/model_trainer.py
-      hash: md5
-      md5: 5192acef195c9a9b03a88490476ead1c
-      size: 3916
-    - path: src/EmotionRecognition/pipeline/stage_02_model_training.py
-      hash: md5
-      md5: 2ee36d6e30a3a262e8327a26e71a37e9
-      size: 1076
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-        TRAINING_PARAMS:
-          EPOCHS: 50
-          LEARNING_RATE: 0.0001
-          OPTIMIZER: Adam
-          LOSS_FUNCTION: CategoricalCrossentropy
-          METRICS:
-          - accuracy
-          DROPOUT_RATE: 0.5
-    outs:
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-    - path: src/EmotionRecognition/components/model_evaluation.py
-      hash: md5
-      md5: 8b327667db406dd7c6489937747b8537
-      size: 2429
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-    outs:
-    - path: artifacts/evaluation/metrics.json
-      hash: md5
-      md5: 3e8f938b34095f56c597110c5d86064e
-      size: 72
-  data_preprocessing:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/data_preprocessing.py
-      hash: md5
-      md5: bc85964fdf86afb289051c2498037eb8
-      size: 3903
-    - path: src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-      hash: md5
-      md5: 5631296a6b7bace5c2f6979eda5ca081
-      size: 971
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preprocessing/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preprocessing/train
-      hash: md5
-      md5: 3dc8382a4774d1a1f1d1e5dfe3ca4c1b.dir
-      size: 18389122
-      nfiles: 10500

huggingface-space/dvc.yaml

@@ -1,39 +0,0 @@
-stages:
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      - src/EmotionRecognition/components/data_preparation.py
-      - data/raw/fer2013.csv
-      - data/raw/fer2013new.csv
-      - data/raw/CK+48
-    params:
-      - DATA_PARAMS.CLASSES
-    outs:
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_02_model_training.py
-      - src/EmotionRecognition/components/model_trainer.py
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-    params:
-      - DATA_PARAMS
-      - TRAINING_PARAMS
-    outs:
-      - artifacts/training/model.keras
-
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-      - src/EmotionRecognition/components/model_evaluation.py
-      - artifacts/data_preparation/test
-      - artifacts/training/model.keras
-    params:
-      - DATA_PARAMS
-    metrics:
-      - artifacts/evaluation/metrics.json:
-          cache: false

huggingface-space/gpuCheck.py

@@ -1,42 +0,0 @@
-import os
-import tensorflow as tf
-
-# --- THE WORKAROUND ---
-# Define the full path to the CUDA bin directory
-cuda_bin_path = r"E:\Nvidia\CUDA\v11.2\bin"
-
-# Add this path to the OS environment's DLL search path
-# This MUST be done BEFORE importing tensorflow
-try:
-    os.add_dll_directory(cuda_bin_path)
-    print(f"Successfully added {cuda_bin_path} to DLL search path.")
-except AttributeError:
-    # This function was added in Python 3.8. For older versions, you might need
-    # to add the path to the system PATH environment variable manually.
-    print("os.add_dll_directory not available. Ensure CUDA bin is in the system PATH.")
-# --- END WORKAROUND ---
-
-
-print(f"TensorFlow Version: {tf.__version__}")
-print("-" * 30)
-
-# Check for GPU devices
-gpu_devices = tf.config.list_physical_devices('GPU')
-print(f"Num GPUs Available: {len(gpu_devices)}")
-print("-" * 30)
-
-if gpu_devices:
-    print("GPU Device Details:")
-    for gpu in gpu_devices:
-        tf.config.experimental.set_memory_growth(gpu, True)
-        print(f"- {gpu.name}, Type: {gpu.device_type}")
-    print("\nSUCCESS: TensorFlow is configured to use the GPU!")
-else:
-    print("\nFAILURE: TensorFlow did not detect a GPU.")
-
-
-import tensorflow as tf
-from tensorflow.python.client import device_lib
-
-print("Verbose device list:")
-print(device_lib.list_local_devices())

huggingface-space/huggingface-space/.dvc/.gitignore

@@ -1,2 +0,0 @@
-/config.local
-/tmp

huggingface-space/huggingface-space/.gitattributes

@@ -1 +0,0 @@
-sota_model/model.safetensors filter=lfs diff=lfs merge=lfs -text

huggingface-space/huggingface-space/.gitignore

@@ -1,112 +0,0 @@
-# MLOps & Data Science Artifacts
-# -------------------------------------------------------------------
-# Ignore all data, models, and artifacts. These should be tracked by DVC.
-/artifacts/
-/data/
-/sota_model/
-# Ignore the DVC local cache. This is where the actual data files are stored.
-.dvc/cache
-
-# Ignore MLflow experiment tracking output
-/mlruns/
-
-# Ignore logs
-/logs/
-*.log
-
-# Ignore common model file extensions, just in case
-*.h5
-*.pkl
-*.model
-*.onnx
-
-
-# Python Virtual Environments
-# -------------------------------------------------------------------
-/venv/
-/myenv/
-/.venv/
-/env/
-/ENV/
-*/.venv/
-*/venv/
-*/myenv/
-
-
-# Python Byte-code and Caches
-# -------------------------------------------------------------------
-__pycache__/
-*.py[cod]
-*$py.class
-
-
-# Python Packaging & Distribution
-# -------------------------------------------------------------------
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-*.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
-
-
-# IDE and Editor Configuration
-# -------------------------------------------------------------------
-# PyCharm
-.idea/
-
-# Visual Studio Code (allow sharing of recommended extensions)
-.vscode/*
-!.vscode/extensions.json
-
-# Sublime Text
-*.sublime-project
-*.sublime-workspace
-
-
-# Secrets and Environment Variables
-# -------------------------------------------------------------------
-# NEVER commit secrets or environment variables
-.env
-*.env
-secrets.yaml
-secrets.json
-
-
-# Operating System Files
-# -------------------------------------------------------------------
-# macOS
-.DS_Store
-
-# Windows
-Thumbs.db
-desktop.ini
-
-
-# Jupyter Notebook Checkpoints
-# -------------------------------------------------------------------
-.ipynb_checkpoints/
-
-
-# Other
-# -------------------------------------------------------------------
-# Temporary files
-*.tmp
-*.bak
-*.swp
-
-.env
-*.env
-secrets.yaml
-secrets.json
-processed_video.mp4

huggingface-space/huggingface-space/LICENSE

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) 2025 ALYYAN
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

huggingface-space/huggingface-space/README.md

@@ -1,42 +0,0 @@
-# 🎭 End-to-End Facial Emotion Recognition
-
- <!-- Replace with a link to your final app screenshot -->
-
-This repository contains a complete, end-to-end MLOps pipeline and a production-ready web application for real-time facial emotion recognition. The project leverages a state-of-the-art Vision Transformer model and is deployed as a user-friendly Gradio application on Hugging Face Spaces.
-
-**Live Demo:** [🚀 Click here to try the application on Hugging Face Spaces!](https://huggingface.co/spaces/ALYYAN/Emotion-Recognition) <!-- Replace with your HF Space URL -->
-
----
-
-## ✨ Features
-
--   **Real-time Emotion Detection:** Analyzes your webcam feed to predict emotions in real-time.
--   **High Accuracy:** Powered by a pre-trained Swin Transformer model fine-tuned on the massive AffectNet dataset for superior performance on "in the wild" faces.
--   **Static Image & Video Analysis:** Upload your own images or videos for emotion prediction.
--   **Polished UI:** A professional and responsive user interface with an animated background, built with Gradio.
--   **Reproducible MLOps Pipeline:** The entire model training and data processing workflow is managed by DVC, ensuring 100% reproducibility.
--   **Containerized for Deployment:** The application is packaged with Docker for easy and consistent deployment anywhere.
-
-## 🛠️ Tech Stack
-
--   **Model:** Swin Transformer (`PangPang/affectnet-swin-tiny-patch4-window7-224`)
--   **ML/Ops:** Python, TensorFlow/Keras, DVC, MLflow, Hugging Face `transformers`
--   **Backend & UI:** Gradio
--   **Face Detection:** MTCNN
--   **Deployment:** Hugging Face Spaces, Docker
-
-## 🚀 Getting Started
-
-Follow these steps to run the project locally.
-
-### Prerequisites
-
--   Python 3.10+
--   Git and Git LFS ([installation guide](https://git-lfs.github.com))
--   An NVIDIA GPU with CUDA drivers is recommended for the training pipeline, but the deployed app runs on CPU.
-
-### 1. Clone the Repository
-
-```bash
-git clone https://github.com/YOUR-USERNAME/Emotion-Recognition-MLOps.git
-cd Emotion-Recognition-MLOps

huggingface-space/huggingface-space/app.py

@@ -1,195 +0,0 @@
-import gradio as gr
-import os
-import cv2
-import time
-
-from src.EmotionRecognition.pipeline.hf_predictor import HFPredictor
-
-# --- INITIALIZE THE MODEL ---
-print("[INFO] Initializing predictor...")
-try:
-    predictor = HFPredictor()
-    print("[INFO] Predictor initialized successfully.")
-except Exception as e:
-    predictor = None
-    print(f"[FATAL ERROR] Failed to initialize predictor: {e}")
-
-# --- UI CONTENT & STYLING ---
-# In app.py
-
-CSS = """
-/* Animated Gradient Background */
-body {
-    background: linear-gradient(-45deg, #0b0f19, #131a2d, #2a2a72, #522a72);
-    background-size: 400% 400%;
-    animation: gradient 15s ease infinite;
-    color: #e0e0e0;
-}
-@keyframes gradient {
-    0% { background-position: 0% 50%; }
-    50% { background-position: 100% 50%; }
-    100% { background-position: 0% 50%; }
-}
-
-/* General Layout & Typography */
-.gradio-container { max-width: 1320px !important; margin: auto !important; }
-#title { text-align: center; font-size: 3rem !important; font-weight: 700; color: #FFF; margin-bottom: 0.5rem; }
-#subtitle { text-align: center; color: #bebebe; margin-top: 0; margin-bottom: 40px; font-size: 1.2rem; font-weight: 300; }
-.gr-button { font-weight: bold !important; }
-
-/* --- NEW: The "Glass Card" effect --- */
-#main-card {
-    background: rgba(22, 22, 34, 0.65); /* Semi-transparent dark background */
-    border-radius: 16px;
-    box-shadow: 0 8px 32px 0 rgba(0, 0, 0, 0.37);
-    backdrop-filter: blur(12px); /* The "frosted glass" effect */
-    -webkit-backdrop-filter: blur(12px); /* For Safari */
-    border: 1px solid rgba(255, 255, 255, 0.18);
-    padding: 1rem;
-}
-/* --- END NEW --- */
-
-/* Prediction Bar Styling - now inside the card */
-#predictions-column { background-color: transparent !important; border-radius: 12px; padding: 1.5rem; }
-#predictions-column > .gr-label { display: none; }
-.prediction-list { list-style-type: none; padding: 0; margin-top: 0; }
-.prediction-list li { display: flex; align-items: center; margin-bottom: 12px; font-size: 1.1rem; }
-.prediction-list .label { width: 100px; text-transform: capitalize; color: #e0e0e0; }
-.prediction-list .bar-container { flex-grow: 1; height: 24px; background-color: rgba(255,255,255,0.1); border-radius: 12px; margin: 0 15px; overflow: hidden; }
-.prediction-list .bar { height: 100%; background: linear-gradient(90deg, #8A2BE2, #C71585); border-radius: 12px; transition: width 0.2s ease-in-out; }
-.prediction-list .percent { width: 60px; text-align: right; font-weight: bold; color: #FFF; }
-footer { display: none !important; }
-"""
-
-ABOUT_MARKDOWN = """
-### Model: Vision Transformer (ViT)
-This application uses a Vision Transformer model, fine-tuned for facial emotion recognition.
-### Dataset
-The model was fine-tuned on the **Emotion Recognition Dataset** from Kaggle, a large, curated collection of labeled facial images. This diverse dataset allows the model to generalize to a wide variety of real-world faces and expressions.
-*Dataset Link:* [https://www.kaggle.com/datasets/sujaykapadnis/emotion-recognition-dataset](https://www.kaggle.com/datasets/sujaykapadnis/emotion-recognition-dataset)
-### MLOps Pipeline
-This entire application, from data processing to training and deployment, was built using a reproducible MLOps pipeline, ensuring consistency and quality at every step.
-"""
-
-# --- BACKEND LOGIC ---
-def create_prediction_html(probabilities):
-    if not probabilities:
-        return "<div style='padding: 2rem; text-align: center; color: #999;'>Waiting for prediction...</div>"
-    html = "<ul class='prediction-list'>"
-    sorted_preds = sorted(probabilities.items(), key=lambda item: item[1], reverse=True)
-    for emotion, prob in sorted_preds:
-        html += f"""
-        <li>
-            <strong class='label'>{emotion}</strong>
-            <div class='bar-container'><div class='bar' style='width: {prob*100:.1f}%;'></div></div>
-            <span class='percent'>{(prob*100):.1f}%</span>
-        </li>
-        """
-    html += "</ul>"
-    return html
-
-def live_detection_stream():
-    """A generator function that runs the live feed loop. This is the definitive fix."""
-    cap = cv2.VideoCapture(0)
-    if not cap.isOpened():
-        print("[ERROR] Cannot open webcam")
-        return
-    try:
-        while True:
-            ret, frame = cap.read()
-            if not ret:
-                time.sleep(0.01)
-                continue
-            
-            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            annotated_frame, probabilities = predictor.process_frame(frame_rgb)
-            yield annotated_frame, create_prediction_html(probabilities)
-            time.sleep(0.05) # Controls FPS. 0.05 = ~20 FPS target. The model inference will be the main bottleneck.
-    finally:
-        print("[INFO] Live feed stopped. Releasing webcam.")
-        cap.release()
-
-def process_image(image):
-    if image is None: return None, create_prediction_html({})
-    annotated_frame, probabilities = predictor.process_frame(image)
-    return annotated_frame, create_prediction_html(probabilities)
-
-def process_video(video_path, progress=gr.Progress(track_tqdm=True)):
-    if video_path is None: return None
-    try:
-        cap = cv2.VideoCapture(video_path)
-        frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
-        output_path = "processed_video.mp4"
-        fourcc = cv2.VideoWriter_fourcc(*'mp4v')
-        fps = cap.get(cv2.CAP_PROP_FPS)
-        width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-        height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-        out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))
-        for _ in progress.tqdm(range(frame_count), desc="Processing Video"):
-            ret, frame = cap.read()
-            if not ret: break
-            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            annotated_frame, _ = predictor.process_frame(frame_rgb)
-            if annotated_frame is not None:
-                out.write(cv2.cvtColor(annotated_frame, cv2.COLOR_RGB2BGR))
-        cap.release()
-        out.release()
-        return output_path
-    except Exception as e:
-        print(f"[ERROR] Video processing failed: {e}")
-        return None
-
-# --- GRADIO UI ---
-with gr.Blocks(css=CSS, theme=gr.themes.Base()) as demo:
-    gr.Markdown("# Facial Emotion Detector", elem_id="title")
-    gr.Markdown("A real-time AI application powered by Vision Transformers", elem_id="subtitle")
-
-    # --- NEW: Wrapper for the glass card effect ---
-    with gr.Box(elem_id="main-card"):
-        with gr.Tabs():
-            with gr.TabItem("Live Detection"):
-                with gr.Row(equal_height=True):
-                    with gr.Column(scale=3):
-                        live_output = gr.Image(label="Live Feed", interactive=False, height=550)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        gr.Markdown("### Emotion Probabilities") # Title for the panel
-                        live_predictions = gr.HTML()
-                with gr.Row():
-                    start_button = gr.Button("Start Webcam", variant="primary", scale=1)
-                    stop_button = gr.Button("Stop Webcam", variant="secondary", scale=1)
-                
-                stream_state = gr.State("Stop")
-
-            with gr.TabItem("Upload Image"):
-                with gr.Row(equal_height=True):
-                    with gr.Column(scale=3):
-                        image_input = gr.Image(type="numpy", label="Upload an Image", height=550)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        gr.Markdown("### Emotion Probabilities")
-                        image_predictions = gr.HTML()
-                image_button = gr.Button("Analyze Image", variant="primary")
-
-            with gr.TabItem("Upload Video"):
-                with gr.Row(equal_height=True):
-                    video_input = gr.Video(label="Upload a Video File")
-                    video_output = gr.Video(label="Processed Video")
-                video_button = gr.Button("Analyze Video", variant="primary")
-            
-            with gr.TabItem("About"):
-                gr.Markdown(ABOUT_MARKDOWN)
-    # --- END WRAPPER ---
-
-    # --- EVENT LISTENERS (No changes needed here) ---
-    start_event = start_button.click(lambda: "Start", None, stream_state, queue=False)
-    live_stream = start_event.then(live_detection_stream, stream_state, [live_output, live_predictions])
-    
-    stop_button.click(fn=None, inputs=None, outputs=None, cancels=[live_stream])
-
-    image_button.click(process_image, [image_input], [image_input, image_predictions])
-    video_button.click(process_video, [video_input], [video_output])
-
-# --- LAUNCH THE APP ---
-if predictor:
-    demo.queue().launch(debug=True, share=True)
-else:
-    print("\n[FATAL ERROR] Could not start the application.")

huggingface-space/huggingface-space/config/config.yaml

@@ -1,25 +0,0 @@
-artifacts_root: artifacts
-
-data_preparation: # This is our Stage 1
-  root_dir: artifacts/data_preparation
-  # Inputs from raw data
-  ferplus_pixels_csv: data/raw/fer2013.csv
-  ferplus_labels_csv: data/raw/fer2013new.csv
-  ckplus_dir: data/raw/CK+48
-  # Outputs
-  combined_train_dir: artifacts/data_preparation/train
-  ferplus_test_dir: artifacts/data_preparation/test
-
-model_trainer:
-  root_dir: artifacts/training
-  # The trainer now takes its input directly from the preparation stage
-  train_data_dir: artifacts/data_preparation/train
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-
-model_evaluation:
-  root_dir: artifacts/evaluation
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-  metrics_file_name: artifacts/evaluation/metrics.json
-  mlflow_uri: https://dagshub.com/AlyyanAhmed21/Emotion-Recognition-MLOps.mlflow # Example for DagsHub

huggingface-space/huggingface-space/dvc.lock

@@ -1,203 +0,0 @@
-schema: '2.0'
-stages:
-  data_validation:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_validation.py
-    deps:
-    - path: artifacts/data_ingestion
-      hash: md5
-      md5: 9208f64defb6697b78bab62e943d955d.dir
-      size: 302675528
-      nfiles: 2
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: dacf4230e18681185b786aa280cdec5e
-      size: 4275
-    - path: src/EmotionRecognition/pipeline/stage_02_data_validation.py
-      hash: md5
-      md5: 18a3d78c83dc5b278e14523077035e41
-      size: 1141
-    outs:
-    - path: artifacts/data_validation/status.txt
-      hash: md5
-      md5: 86e6a2f694c57a675b3e2da6b95ff9ba
-      size: 23
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-    - path: data/raw/CK+48
-      hash: md5
-      md5: a1559eddfd0d86b541e5df18b4b8205e.dir
-      size: 1715162
-      nfiles: 981
-    - path: data/raw/fer2013.csv
-      hash: md5
-      md5: f8428a1edbd21e88f42c73edd2a14f95
-      size: 301072766
-    - path: data/raw/fer2013new.csv
-      hash: md5
-      md5: 413eba86d6e454536b99705b8c7fc5c5
-      size: 1602762
-    - path: src/EmotionRecognition/components/data_preparation.py
-      hash: md5
-      md5: 228140227aaedb9f07b4c00462f267c6
-      size: 5776
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: 8786c8d41e2e50a49b4ca6d5bf59ad44
-      size: 2910
-    - path: src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      hash: md5
-      md5: 1a324b8f1cf01e4e60e0a8529b23b577
-      size: 1110
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/model_trainer.py
-      hash: md5
-      md5: 5192acef195c9a9b03a88490476ead1c
-      size: 3916
-    - path: src/EmotionRecognition/pipeline/stage_02_model_training.py
-      hash: md5
-      md5: 2ee36d6e30a3a262e8327a26e71a37e9
-      size: 1076
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-        TRAINING_PARAMS:
-          EPOCHS: 50
-          LEARNING_RATE: 0.0001
-          OPTIMIZER: Adam
-          LOSS_FUNCTION: CategoricalCrossentropy
-          METRICS:
-          - accuracy
-          DROPOUT_RATE: 0.5
-    outs:
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-    - path: src/EmotionRecognition/components/model_evaluation.py
-      hash: md5
-      md5: 8b327667db406dd7c6489937747b8537
-      size: 2429
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-    outs:
-    - path: artifacts/evaluation/metrics.json
-      hash: md5
-      md5: 3e8f938b34095f56c597110c5d86064e
-      size: 72
-  data_preprocessing:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/data_preprocessing.py
-      hash: md5
-      md5: bc85964fdf86afb289051c2498037eb8
-      size: 3903
-    - path: src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-      hash: md5
-      md5: 5631296a6b7bace5c2f6979eda5ca081
-      size: 971
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preprocessing/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preprocessing/train
-      hash: md5
-      md5: 3dc8382a4774d1a1f1d1e5dfe3ca4c1b.dir
-      size: 18389122
-      nfiles: 10500

huggingface-space/huggingface-space/dvc.yaml

@@ -1,39 +0,0 @@
-stages:
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      - src/EmotionRecognition/components/data_preparation.py
-      - data/raw/fer2013.csv
-      - data/raw/fer2013new.csv
-      - data/raw/CK+48
-    params:
-      - DATA_PARAMS.CLASSES
-    outs:
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_02_model_training.py
-      - src/EmotionRecognition/components/model_trainer.py
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-    params:
-      - DATA_PARAMS
-      - TRAINING_PARAMS
-    outs:
-      - artifacts/training/model.keras
-
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-      - src/EmotionRecognition/components/model_evaluation.py
-      - artifacts/data_preparation/test
-      - artifacts/training/model.keras
-    params:
-      - DATA_PARAMS
-    metrics:
-      - artifacts/evaluation/metrics.json:
-          cache: false

huggingface-space/huggingface-space/gpuCheck.py

@@ -1,42 +0,0 @@
-import os
-import tensorflow as tf
-
-# --- THE WORKAROUND ---
-# Define the full path to the CUDA bin directory
-cuda_bin_path = r"E:\Nvidia\CUDA\v11.2\bin"
-
-# Add this path to the OS environment's DLL search path
-# This MUST be done BEFORE importing tensorflow
-try:
-    os.add_dll_directory(cuda_bin_path)
-    print(f"Successfully added {cuda_bin_path} to DLL search path.")
-except AttributeError:
-    # This function was added in Python 3.8. For older versions, you might need
-    # to add the path to the system PATH environment variable manually.
-    print("os.add_dll_directory not available. Ensure CUDA bin is in the system PATH.")
-# --- END WORKAROUND ---
-
-
-print(f"TensorFlow Version: {tf.__version__}")
-print("-" * 30)
-
-# Check for GPU devices
-gpu_devices = tf.config.list_physical_devices('GPU')
-print(f"Num GPUs Available: {len(gpu_devices)}")
-print("-" * 30)
-
-if gpu_devices:
-    print("GPU Device Details:")
-    for gpu in gpu_devices:
-        tf.config.experimental.set_memory_growth(gpu, True)
-        print(f"- {gpu.name}, Type: {gpu.device_type}")
-    print("\nSUCCESS: TensorFlow is configured to use the GPU!")
-else:
-    print("\nFAILURE: TensorFlow did not detect a GPU.")
-
-
-import tensorflow as tf
-from tensorflow.python.client import device_lib
-
-print("Verbose device list:")
-print(device_lib.list_local_devices())

huggingface-space/huggingface-space/huggingface-space/.dvc/.gitignore

@@ -1,2 +0,0 @@
-/config.local
-/tmp

huggingface-space/huggingface-space/huggingface-space/.gitattributes

@@ -1 +0,0 @@
-sota_model/model.safetensors filter=lfs diff=lfs merge=lfs -text

huggingface-space/huggingface-space/huggingface-space/.gitignore

@@ -1,112 +0,0 @@
-# MLOps & Data Science Artifacts
-# -------------------------------------------------------------------
-# Ignore all data, models, and artifacts. These should be tracked by DVC.
-/artifacts/
-/data/
-/sota_model/
-# Ignore the DVC local cache. This is where the actual data files are stored.
-.dvc/cache
-
-# Ignore MLflow experiment tracking output
-/mlruns/
-
-# Ignore logs
-/logs/
-*.log
-
-# Ignore common model file extensions, just in case
-*.h5
-*.pkl
-*.model
-*.onnx
-
-
-# Python Virtual Environments
-# -------------------------------------------------------------------
-/venv/
-/myenv/
-/.venv/
-/env/
-/ENV/
-*/.venv/
-*/venv/
-*/myenv/
-
-
-# Python Byte-code and Caches
-# -------------------------------------------------------------------
-__pycache__/
-*.py[cod]
-*$py.class
-
-
-# Python Packaging & Distribution
-# -------------------------------------------------------------------
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-*.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
-
-
-# IDE and Editor Configuration
-# -------------------------------------------------------------------
-# PyCharm
-.idea/
-
-# Visual Studio Code (allow sharing of recommended extensions)
-.vscode/*
-!.vscode/extensions.json
-
-# Sublime Text
-*.sublime-project
-*.sublime-workspace
-
-
-# Secrets and Environment Variables
-# -------------------------------------------------------------------
-# NEVER commit secrets or environment variables
-.env
-*.env
-secrets.yaml
-secrets.json
-
-
-# Operating System Files
-# -------------------------------------------------------------------
-# macOS
-.DS_Store
-
-# Windows
-Thumbs.db
-desktop.ini
-
-
-# Jupyter Notebook Checkpoints
-# -------------------------------------------------------------------
-.ipynb_checkpoints/
-
-
-# Other
-# -------------------------------------------------------------------
-# Temporary files
-*.tmp
-*.bak
-*.swp
-
-.env
-*.env
-secrets.yaml
-secrets.json
-processed_video.mp4

huggingface-space/huggingface-space/huggingface-space/LICENSE

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) 2025 ALYYAN
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

huggingface-space/huggingface-space/huggingface-space/README.md

@@ -1,42 +0,0 @@
-# 🎭 End-to-End Facial Emotion Recognition
-
- <!-- Replace with a link to your final app screenshot -->
-
-This repository contains a complete, end-to-end MLOps pipeline and a production-ready web application for real-time facial emotion recognition. The project leverages a state-of-the-art Vision Transformer model and is deployed as a user-friendly Gradio application on Hugging Face Spaces.
-
-**Live Demo:** [🚀 Click here to try the application on Hugging Face Spaces!](https://huggingface.co/spaces/ALYYAN/Emotion-Recognition) <!-- Replace with your HF Space URL -->
-
----
-
-## ✨ Features
-
--   **Real-time Emotion Detection:** Analyzes your webcam feed to predict emotions in real-time.
--   **High Accuracy:** Powered by a pre-trained Swin Transformer model fine-tuned on the massive AffectNet dataset for superior performance on "in the wild" faces.
--   **Static Image & Video Analysis:** Upload your own images or videos for emotion prediction.
--   **Polished UI:** A professional and responsive user interface with an animated background, built with Gradio.
--   **Reproducible MLOps Pipeline:** The entire model training and data processing workflow is managed by DVC, ensuring 100% reproducibility.
--   **Containerized for Deployment:** The application is packaged with Docker for easy and consistent deployment anywhere.
-
-## 🛠️ Tech Stack
-
--   **Model:** Swin Transformer (`PangPang/affectnet-swin-tiny-patch4-window7-224`)
--   **ML/Ops:** Python, TensorFlow/Keras, DVC, MLflow, Hugging Face `transformers`
--   **Backend & UI:** Gradio
--   **Face Detection:** MTCNN
--   **Deployment:** Hugging Face Spaces, Docker
-
-## 🚀 Getting Started
-
-Follow these steps to run the project locally.
-
-### Prerequisites
-
--   Python 3.10+
--   Git and Git LFS ([installation guide](https://git-lfs.github.com))
--   An NVIDIA GPU with CUDA drivers is recommended for the training pipeline, but the deployed app runs on CPU.
-
-### 1. Clone the Repository
-
-```bash
-git clone https://github.com/YOUR-USERNAME/Emotion-Recognition-MLOps.git
-cd Emotion-Recognition-MLOps

huggingface-space/huggingface-space/huggingface-space/app.py

@@ -1,195 +0,0 @@
-import gradio as gr
-import os
-import cv2
-import time
-
-from src.EmotionRecognition.pipeline.hf_predictor import HFPredictor
-
-# --- INITIALIZE THE MODEL ---
-print("[INFO] Initializing predictor...")
-try:
-    predictor = HFPredictor()
-    print("[INFO] Predictor initialized successfully.")
-except Exception as e:
-    predictor = None
-    print(f"[FATAL ERROR] Failed to initialize predictor: {e}")
-
-# --- UI CONTENT & STYLING ---
-# In app.py
-
-CSS = """
-/* Animated Gradient Background */
-body {
-    background: linear-gradient(-45deg, #0b0f19, #131a2d, #2a2a72, #522a72);
-    background-size: 400% 400%;
-    animation: gradient 15s ease infinite;
-    color: #e0e0e0;
-}
-@keyframes gradient {
-    0% { background-position: 0% 50%; }
-    50% { background-position: 100% 50%; }
-    100% { background-position: 0% 50%; }
-}
-
-/* General Layout & Typography */
-.gradio-container { max-width: 1320px !important; margin: auto !important; }
-#title { text-align: center; font-size: 3rem !important; font-weight: 700; color: #FFF; margin-bottom: 0.5rem; }
-#subtitle { text-align: center; color: #bebebe; margin-top: 0; margin-bottom: 40px; font-size: 1.2rem; font-weight: 300; }
-.gr-button { font-weight: bold !important; }
-
-/* --- NEW: The "Glass Card" effect --- */
-#main-card {
-    background: rgba(22, 22, 34, 0.65); /* Semi-transparent dark background */
-    border-radius: 16px;
-    box-shadow: 0 8px 32px 0 rgba(0, 0, 0, 0.37);
-    backdrop-filter: blur(12px); /* The "frosted glass" effect */
-    -webkit-backdrop-filter: blur(12px); /* For Safari */
-    border: 1px solid rgba(255, 255, 255, 0.18);
-    padding: 1rem;
-}
-/* --- END NEW --- */
-
-/* Prediction Bar Styling - now inside the card */
-#predictions-column { background-color: transparent !important; border-radius: 12px; padding: 1.5rem; }
-#predictions-column > .gr-label { display: none; }
-.prediction-list { list-style-type: none; padding: 0; margin-top: 0; }
-.prediction-list li { display: flex; align-items: center; margin-bottom: 12px; font-size: 1.1rem; }
-.prediction-list .label { width: 100px; text-transform: capitalize; color: #e0e0e0; }
-.prediction-list .bar-container { flex-grow: 1; height: 24px; background-color: rgba(255,255,255,0.1); border-radius: 12px; margin: 0 15px; overflow: hidden; }
-.prediction-list .bar { height: 100%; background: linear-gradient(90deg, #8A2BE2, #C71585); border-radius: 12px; transition: width 0.2s ease-in-out; }
-.prediction-list .percent { width: 60px; text-align: right; font-weight: bold; color: #FFF; }
-footer { display: none !important; }
-"""
-
-ABOUT_MARKDOWN = """
-### Model: Vision Transformer (ViT)
-This application uses a Vision Transformer model, fine-tuned for facial emotion recognition.
-### Dataset
-The model was fine-tuned on the **Emotion Recognition Dataset** from Kaggle, a large, curated collection of labeled facial images. This diverse dataset allows the model to generalize to a wide variety of real-world faces and expressions.
-*Dataset Link:* [https://www.kaggle.com/datasets/sujaykapadnis/emotion-recognition-dataset](https://www.kaggle.com/datasets/sujaykapadnis/emotion-recognition-dataset)
-### MLOps Pipeline
-This entire application, from data processing to training and deployment, was built using a reproducible MLOps pipeline, ensuring consistency and quality at every step.
-"""
-
-# --- BACKEND LOGIC ---
-def create_prediction_html(probabilities):
-    if not probabilities:
-        return "<div style='padding: 2rem; text-align: center; color: #999;'>Waiting for prediction...</div>"
-    html = "<ul class='prediction-list'>"
-    sorted_preds = sorted(probabilities.items(), key=lambda item: item[1], reverse=True)
-    for emotion, prob in sorted_preds:
-        html += f"""
-        <li>
-            <strong class='label'>{emotion}</strong>
-            <div class='bar-container'><div class='bar' style='width: {prob*100:.1f}%;'></div></div>
-            <span class='percent'>{(prob*100):.1f}%</span>
-        </li>
-        """
-    html += "</ul>"
-    return html
-
-def live_detection_stream():
-    """A generator function that runs the live feed loop. This is the definitive fix."""
-    cap = cv2.VideoCapture(0)
-    if not cap.isOpened():
-        print("[ERROR] Cannot open webcam")
-        return
-    try:
-        while True:
-            ret, frame = cap.read()
-            if not ret:
-                time.sleep(0.01)
-                continue
-            
-            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            annotated_frame, probabilities = predictor.process_frame(frame_rgb)
-            yield annotated_frame, create_prediction_html(probabilities)
-            time.sleep(0.05) # Controls FPS. 0.05 = ~20 FPS target. The model inference will be the main bottleneck.
-    finally:
-        print("[INFO] Live feed stopped. Releasing webcam.")
-        cap.release()
-
-def process_image(image):
-    if image is None: return None, create_prediction_html({})
-    annotated_frame, probabilities = predictor.process_frame(image)
-    return annotated_frame, create_prediction_html(probabilities)
-
-def process_video(video_path, progress=gr.Progress(track_tqdm=True)):
-    if video_path is None: return None
-    try:
-        cap = cv2.VideoCapture(video_path)
-        frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
-        output_path = "processed_video.mp4"
-        fourcc = cv2.VideoWriter_fourcc(*'mp4v')
-        fps = cap.get(cv2.CAP_PROP_FPS)
-        width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-        height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-        out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))
-        for _ in progress.tqdm(range(frame_count), desc="Processing Video"):
-            ret, frame = cap.read()
-            if not ret: break
-            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-            annotated_frame, _ = predictor.process_frame(frame_rgb)
-            if annotated_frame is not None:
-                out.write(cv2.cvtColor(annotated_frame, cv2.COLOR_RGB2BGR))
-        cap.release()
-        out.release()
-        return output_path
-    except Exception as e:
-        print(f"[ERROR] Video processing failed: {e}")
-        return None
-
-# --- GRADIO UI ---
-with gr.Blocks(css=CSS, theme=gr.themes.Base()) as demo:
-    gr.Markdown("# Facial Emotion Detector", elem_id="title")
-    gr.Markdown("A real-time AI application powered by Vision Transformers", elem_id="subtitle")
-
-    # --- NEW: Wrapper for the glass card effect ---
-    with gr.Box(elem_id="main-card"):
-        with gr.Tabs():
-            with gr.TabItem("Live Detection"):
-                with gr.Row(equal_height=True):
-                    with gr.Column(scale=3):
-                        live_output = gr.Image(label="Live Feed", interactive=False, height=550)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        gr.Markdown("### Emotion Probabilities") # Title for the panel
-                        live_predictions = gr.HTML()
-                with gr.Row():
-                    start_button = gr.Button("Start Webcam", variant="primary", scale=1)
-                    stop_button = gr.Button("Stop Webcam", variant="secondary", scale=1)
-                
-                stream_state = gr.State("Stop")
-
-            with gr.TabItem("Upload Image"):
-                with gr.Row(equal_height=True):
-                    with gr.Column(scale=3):
-                        image_input = gr.Image(type="numpy", label="Upload an Image", height=550)
-                    with gr.Column(scale=2, elem_id="predictions-column"):
-                        gr.Markdown("### Emotion Probabilities")
-                        image_predictions = gr.HTML()
-                image_button = gr.Button("Analyze Image", variant="primary")
-
-            with gr.TabItem("Upload Video"):
-                with gr.Row(equal_height=True):
-                    video_input = gr.Video(label="Upload a Video File")
-                    video_output = gr.Video(label="Processed Video")
-                video_button = gr.Button("Analyze Video", variant="primary")
-            
-            with gr.TabItem("About"):
-                gr.Markdown(ABOUT_MARKDOWN)
-    # --- END WRAPPER ---
-
-    # --- EVENT LISTENERS (No changes needed here) ---
-    start_event = start_button.click(lambda: "Start", None, stream_state, queue=False)
-    live_stream = start_event.then(live_detection_stream, stream_state, [live_output, live_predictions])
-    
-    stop_button.click(fn=None, inputs=None, outputs=None, cancels=[live_stream])
-
-    image_button.click(process_image, [image_input], [image_input, image_predictions])
-    video_button.click(process_video, [video_input], [video_output])
-
-# --- LAUNCH THE APP ---
-if predictor:
-    demo.queue().launch(debug=True, share=True)
-else:
-    print("\n[FATAL ERROR] Could not start the application.")

huggingface-space/huggingface-space/huggingface-space/config/config.yaml

@@ -1,25 +0,0 @@
-artifacts_root: artifacts
-
-data_preparation: # This is our Stage 1
-  root_dir: artifacts/data_preparation
-  # Inputs from raw data
-  ferplus_pixels_csv: data/raw/fer2013.csv
-  ferplus_labels_csv: data/raw/fer2013new.csv
-  ckplus_dir: data/raw/CK+48
-  # Outputs
-  combined_train_dir: artifacts/data_preparation/train
-  ferplus_test_dir: artifacts/data_preparation/test
-
-model_trainer:
-  root_dir: artifacts/training
-  # The trainer now takes its input directly from the preparation stage
-  train_data_dir: artifacts/data_preparation/train
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-
-model_evaluation:
-  root_dir: artifacts/evaluation
-  test_data_dir: artifacts/data_preparation/test
-  trained_model_path: artifacts/training/model.keras
-  metrics_file_name: artifacts/evaluation/metrics.json
-  mlflow_uri: https://dagshub.com/AlyyanAhmed21/Emotion-Recognition-MLOps.mlflow # Example for DagsHub

huggingface-space/huggingface-space/huggingface-space/dvc.lock

@@ -1,203 +0,0 @@
-schema: '2.0'
-stages:
-  data_validation:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_validation.py
-    deps:
-    - path: artifacts/data_ingestion
-      hash: md5
-      md5: 9208f64defb6697b78bab62e943d955d.dir
-      size: 302675528
-      nfiles: 2
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: dacf4230e18681185b786aa280cdec5e
-      size: 4275
-    - path: src/EmotionRecognition/pipeline/stage_02_data_validation.py
-      hash: md5
-      md5: 18a3d78c83dc5b278e14523077035e41
-      size: 1141
-    outs:
-    - path: artifacts/data_validation/status.txt
-      hash: md5
-      md5: 86e6a2f694c57a675b3e2da6b95ff9ba
-      size: 23
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-    - path: data/raw/CK+48
-      hash: md5
-      md5: a1559eddfd0d86b541e5df18b4b8205e.dir
-      size: 1715162
-      nfiles: 981
-    - path: data/raw/fer2013.csv
-      hash: md5
-      md5: f8428a1edbd21e88f42c73edd2a14f95
-      size: 301072766
-    - path: data/raw/fer2013new.csv
-      hash: md5
-      md5: 413eba86d6e454536b99705b8c7fc5c5
-      size: 1602762
-    - path: src/EmotionRecognition/components/data_preparation.py
-      hash: md5
-      md5: 228140227aaedb9f07b4c00462f267c6
-      size: 5776
-    - path: src/EmotionRecognition/config/configuration.py
-      hash: md5
-      md5: 8786c8d41e2e50a49b4ca6d5bf59ad44
-      size: 2910
-    - path: src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      hash: md5
-      md5: 1a324b8f1cf01e4e60e0a8529b23b577
-      size: 1110
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/model_trainer.py
-      hash: md5
-      md5: 5192acef195c9a9b03a88490476ead1c
-      size: 3916
-    - path: src/EmotionRecognition/pipeline/stage_02_model_training.py
-      hash: md5
-      md5: 2ee36d6e30a3a262e8327a26e71a37e9
-      size: 1076
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-        TRAINING_PARAMS:
-          EPOCHS: 50
-          LEARNING_RATE: 0.0001
-          OPTIMIZER: Adam
-          LOSS_FUNCTION: CategoricalCrossentropy
-          METRICS:
-          - accuracy
-          DROPOUT_RATE: 0.5
-    outs:
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/training/model.keras
-      hash: md5
-      md5: 2c632cb4cbf3f2944145a8da1927f2cf
-      size: 11331400
-    - path: src/EmotionRecognition/components/model_evaluation.py
-      hash: md5
-      md5: 8b327667db406dd7c6489937747b8537
-      size: 2429
-    params:
-      params.yaml:
-        DATA_PARAMS:
-          IMAGE_SIZE:
-          - 224
-          - 224
-          CHANNELS: 3
-          BATCH_SIZE: 32
-          CLASSES:
-          - angry
-          - disgust
-          - fear
-          - happy
-          - neutral
-          - sad
-          - surprise
-          NUM_CLASSES: 7
-    outs:
-    - path: artifacts/evaluation/metrics.json
-      hash: md5
-      md5: 3e8f938b34095f56c597110c5d86064e
-      size: 72
-  data_preprocessing:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-    deps:
-    - path: artifacts/data_preparation/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preparation/train
-      hash: md5
-      md5: 750c0a305d28467341396ab591ed2731.dir
-      size: 51232879
-      nfiles: 29471
-    - path: src/EmotionRecognition/components/data_preprocessing.py
-      hash: md5
-      md5: bc85964fdf86afb289051c2498037eb8
-      size: 3903
-    - path: src/EmotionRecognition/pipeline/stage_02_data_preprocessing.py
-      hash: md5
-      md5: 5631296a6b7bace5c2f6979eda5ca081
-      size: 971
-    params:
-      params.yaml:
-        DATA_PARAMS.CLASSES:
-        - angry
-        - disgust
-        - fear
-        - happy
-        - neutral
-        - sad
-        - surprise
-    outs:
-    - path: artifacts/data_preprocessing/test
-      hash: md5
-      md5: 79c105a50ccbe2557fea9fab2c743fa5.dir
-      size: 6249935
-      nfiles: 3589
-    - path: artifacts/data_preprocessing/train
-      hash: md5
-      md5: 3dc8382a4774d1a1f1d1e5dfe3ca4c1b.dir
-      size: 18389122
-      nfiles: 10500

huggingface-space/huggingface-space/huggingface-space/dvc.yaml

@@ -1,39 +0,0 @@
-stages:
-  data_preparation:
-    cmd: python src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_01_data_preparation.py
-      - src/EmotionRecognition/components/data_preparation.py
-      - data/raw/fer2013.csv
-      - data/raw/fer2013new.csv
-      - data/raw/CK+48
-    params:
-      - DATA_PARAMS.CLASSES
-    outs:
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-
-  model_training:
-    cmd: python src/EmotionRecognition/pipeline/stage_02_model_training.py
-    deps:
-      - src/EmotionRecognition/pipeline/stage_02_model_training.py
-      - src/EmotionRecognition/components/model_trainer.py
-      - artifacts/data_preparation/train
-      - artifacts/data_preparation/test
-    params:
-      - DATA_PARAMS
-      - TRAINING_PARAMS
-    outs:
-      - artifacts/training/model.keras
-
-  model_evaluation:
-    cmd: python src/EmotionRecognition/pipeline/stage_03_model_evaluation.py
-    deps:
-      - src/EmotionRecognition/components/model_evaluation.py
-      - artifacts/data_preparation/test
-      - artifacts/training/model.keras
-    params:
-      - DATA_PARAMS
-    metrics:
-      - artifacts/evaluation/metrics.json:
-          cache: false

huggingface-space/huggingface-space/huggingface-space/gpuCheck.py

@@ -1,42 +0,0 @@
-import os
-import tensorflow as tf
-
-# --- THE WORKAROUND ---
-# Define the full path to the CUDA bin directory
-cuda_bin_path = r"E:\Nvidia\CUDA\v11.2\bin"
-
-# Add this path to the OS environment's DLL search path
-# This MUST be done BEFORE importing tensorflow
-try:
-    os.add_dll_directory(cuda_bin_path)
-    print(f"Successfully added {cuda_bin_path} to DLL search path.")
-except AttributeError:
-    # This function was added in Python 3.8. For older versions, you might need
-    # to add the path to the system PATH environment variable manually.
-    print("os.add_dll_directory not available. Ensure CUDA bin is in the system PATH.")
-# --- END WORKAROUND ---
-
-
-print(f"TensorFlow Version: {tf.__version__}")
-print("-" * 30)
-
-# Check for GPU devices
-gpu_devices = tf.config.list_physical_devices('GPU')
-print(f"Num GPUs Available: {len(gpu_devices)}")
-print("-" * 30)
-
-if gpu_devices:
-    print("GPU Device Details:")
-    for gpu in gpu_devices:
-        tf.config.experimental.set_memory_growth(gpu, True)
-        print(f"- {gpu.name}, Type: {gpu.device_type}")
-    print("\nSUCCESS: TensorFlow is configured to use the GPU!")
-else:
-    print("\nFAILURE: TensorFlow did not detect a GPU.")
-
-
-import tensorflow as tf
-from tensorflow.python.client import device_lib
-
-print("Verbose device list:")
-print(device_lib.list_local_devices())

huggingface-space/huggingface-space/huggingface-space/huggingface-space/.gitattributes

@@ -1 +0,0 @@
-sota_model/model.safetensors filter=lfs diff=lfs merge=lfs -text

huggingface-space/huggingface-space/huggingface-space/huggingface-space/README.md

@@ -1,42 +0,0 @@
-# 🎭 End-to-End Facial Emotion Recognition
-
- <!-- Replace with a link to your final app screenshot -->
-
-This repository contains a complete, end-to-end MLOps pipeline and a production-ready web application for real-time facial emotion recognition. The project leverages a state-of-the-art Vision Transformer model and is deployed as a user-friendly Gradio application on Hugging Face Spaces.
-
-**Live Demo:** [🚀 Click here to try the application on Hugging Face Spaces!](https://huggingface.co/spaces/YOUR-USERNAME/YOUR-SPACE-NAME) <!-- Replace with your HF Space URL -->
-
----
-
-## ✨ Features
-
--   **Real-time Emotion Detection:** Analyzes your webcam feed to predict emotions in real-time.
--   **High Accuracy:** Powered by a pre-trained Swin Transformer model fine-tuned on the massive AffectNet dataset for superior performance on "in the wild" faces.
--   **Static Image & Video Analysis:** Upload your own images or videos for emotion prediction.
--   **Polished UI:** A professional and responsive user interface with an animated background, built with Gradio.
--   **Reproducible MLOps Pipeline:** The entire model training and data processing workflow is managed by DVC, ensuring 100% reproducibility.
--   **Containerized for Deployment:** The application is packaged with Docker for easy and consistent deployment anywhere.
-
-## 🛠️ Tech Stack
-
--   **Model:** Swin Transformer (`PangPang/affectnet-swin-tiny-patch4-window7-224`)
--   **ML/Ops:** Python, TensorFlow/Keras, DVC, MLflow, Hugging Face `transformers`
--   **Backend & UI:** Gradio
--   **Face Detection:** MTCNN
--   **Deployment:** Hugging Face Spaces, Docker
-
-## 🚀 Getting Started
-
-Follow these steps to run the project locally.
-
-### Prerequisites
-
--   Python 3.10+
--   Git and Git LFS ([installation guide](https://git-lfs.github.com))
--   An NVIDIA GPU with CUDA drivers is recommended for the training pipeline, but the deployed app runs on CPU.
-
-### 1. Clone the Repository
-
-```bash
-git clone https://github.com/YOUR-USERNAME/Emotion-Recognition-MLOps.git
-cd Emotion-Recognition-MLOps

huggingface-space/huggingface-space/huggingface-space/main.py

@@ -1,70 +0,0 @@
-from EmotionRecognition import logger
-from EmotionRecognition.pipeline.stage_01_data_ingestion import DataIngestionTrainingPipeline
-from EmotionRecognition.pipeline.stage_02_data_validation import DataValidationTrainingPipeline
-from EmotionRecognition.pipeline.stage_01_data_preparation import DataPreparationPipeline
-from EmotionRecognition.pipeline.stage_02_model_training import ModelTrainingPipeline
-from EmotionRecognition.pipeline.stage_03_model_evaluation import ModelEvaluationPipeline 
-
-# Data Ingestion Stage
-STAGE_NAME = "Data Ingestion Stage"
-try:
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-    obj = DataIngestionTrainingPipeline()
-    obj.main()
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-except Exception as e:
-    logger.exception(e)
-    raise e
-
-# Data Validation Stage
-STAGE_NAME = "Data Validation Stage"
-try:
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-    obj = DataValidationTrainingPipeline()
-    obj.main()
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-except Exception as e:
-    logger.exception(e)
-    raise e
-
-# Data Preprocessing Stage
-#STAGE_NAME = "Data Preprocessing Stage"
-#try:
-#    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-#    obj = DataPreprocessingTrainingPipeline()
-#    obj.main()
-#    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-#except Exception as e:
-#    logger.exception(e)
-#    raise e
-
-STAGE_NAME = "Data Preparation Stage"
-try:
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-    obj = DataPreparationPipeline()
-    obj.main()
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-except Exception as e:
-    logger.exception(e)
-    raise e
-
-STAGE_NAME = "Model Training Stage"
-try:
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-    obj = ModelTrainingPipeline()
-    obj.main()
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-except Exception as e:
-    logger.exception(e)
-    raise e
-
-# Model Evaluation Stage
-STAGE_NAME = "Model Evaluation Stage"
-try:
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' started <<<<<<")
-    obj = ModelEvaluationPipeline()
-    obj.main()
-    logger.info(f">>>>>> Stage '{STAGE_NAME}' completed successfully <<<<<<\n\nx==========x")
-except Exception as e:
-    logger.exception(e)
-    raise e

huggingface-space/huggingface-space/huggingface-space/params.yaml

@@ -1,15 +0,0 @@
-DATA_PARAMS:
-  IMAGE_SIZE: [224, 224]
-  CHANNELS: 3
-  BATCH_SIZE: 32
-  # Our final 7 classes (Contempt is merged into Disgust)
-  CLASSES: ['angry', 'disgust', 'fear', 'happy', 'neutral', 'sad', 'surprise']
-  NUM_CLASSES: 7
-
-TRAINING_PARAMS:
-  EPOCHS: 50 # A solid number for a baseline run
-  LEARNING_RATE: 0.0001 # A small, stable learning rate
-  OPTIMIZER: Adam
-  LOSS_FUNCTION: CategoricalCrossentropy
-  METRICS: ['accuracy']
-  DROPOUT_RATE: 0.5 # Strong regularization is good

huggingface-space/huggingface-space/huggingface-space/requirements.txt

@@ -1,44 +0,0 @@
-# -----------------------------------------------------------
-# MLOps Pipeline & Data Versioning
-# -----------------------------------------------------------
-dvc[s3]               # For data versioning. Change [s3] to your remote type or remove.
-kaggle                # For downloading datasets from Kaggle.
-
-
-# -----------------------------------------------------------
-# Core Machine Learning & Deep Learning
-# -----------------------------------------------------------
-# Using the stable TensorFlow 2.10 for GPU support on native Windows
-tensorflow==2.10.0
-scikit-learn          # For evaluation metrics.
-
-# Hugging Face SOTA Model Dependencies
-transformers          # For loading models and processors from the Hub.
-torch                 # PyTorch is a backend dependency for many HF vision models.
-torchvision
-timm                  # Another common dependency for HF vision transformers.
-
-# Computer Vision
-opencv-python         # For image/video processing and drawing.
-mtcnn                 # For fast and effective face detection.
-Pillow                # For basic image manipulation.
-
-
-# -----------------------------------------------------------
-# Web Application & User Interface
-# -----------------------------------------------------------
-gradio==3.50.2        # Locked to a stable version for consistent UI behavior.
-
-
-# -----------------------------------------------------------
-# Utilities
-# -----------------------------------------------------------
-numpy
-pandas                # For data manipulation in the preparation stage.
-PyYAML                # For reading .yaml configuration files.
-python-box            # For dot-notation access to config dictionaries.
-tqdm                  # For progress bars in scripts.
-ensure                # For runtime type checking.
-matplotlib            # For plotting (useful in research).
-seaborn               # For advanced plotting (useful in research).
-notebook              # For running Jupyter notebooks.

huggingface-space/huggingface-space/huggingface-space/setup.py

@@ -1,28 +0,0 @@
-import setuptools
-
-with open("README.md", "r", encoding="utf-8") as f:
-    long_description = f.read()
-
-__version__ = "0.0.1"
-
-REPO_NAME = "Emotion-Recognition-MLOps"
-AUTHOR_USER_NAME = "AlyyanAhmed21"
-SRC_REPO = "EmotionRecognition"
-AUTHOR_EMAIL = "alyyanawan19@gmail.com" 
-
-
-setuptools.setup(
-    name=SRC_REPO,
-    version=__version__,
-    author=AUTHOR_USER_NAME,
-    author_email=AUTHOR_EMAIL,
-    description="A small python package for MLOps based facial emotion detection app",
-    long_description=long_description,
-    long_description_content_type="text/markdown",
-    url=f"https://github.com/{AUTHOR_USER_NAME}/{REPO_NAME}",
-    project_urls={
-        "Bug Tracker": f"https://github.com/{AUTHOR_USER_NAME}/{REPO_NAME}/issues",
-    },
-    package_dir={"": "src"},
-    packages=setuptools.find_packages(where="src")
-)

huggingface-space/huggingface-space/huggingface-space/src/EmotionRecognition/__init__.py

@@ -1,25 +0,0 @@
-import os
-import sys
-import logging
-
-# Define the logging format
-logging_str = "[%(asctime)s: %(levelname)s: %(module)s: %(message)s]"
-
-# Define the directory for log files
-log_dir = "logs"
-log_filepath = os.path.join(log_dir, "running_logs.log")
-os.makedirs(log_dir, exist_ok=True)
-
-# Configure the logging
-logging.basicConfig(
-    level=logging.INFO,
-    format=logging_str,
-
-    handlers=[
-        logging.FileHandler(log_filepath),  # Log to a file
-        logging.StreamHandler(sys.stdout)   # Log to the console
-    ]
-)
-
-# Create a logger object that can be imported by other modules
-logger = logging.getLogger("EmotionRecognitionLogger")

huggingface-space/huggingface-space/huggingface-space/src/EmotionRecognition/components/data_ingestion.py

@@ -1,27 +0,0 @@
-# File: src/EmotionRecognition/components/data_ingestion.py
-import os
-from EmotionRecognition import logger
-from EmotionRecognition.entity.config_entity import DataIngestionConfig
-
-class DataIngestion:
-    def __init__(self, config: DataIngestionConfig):
-        self.config = config
-
-    def validate_source_data(self):
-        """
-        Validates the existence of all raw source data files and folders.
-        """
-        logger.info("Validating source data files and folders...")
-        
-        all_paths = [
-            self.config.root_dir,
-            self.config.ferplus_pixels_csv,
-            self.config.ferplus_labels_csv,
-            self.config.ckplus_dir
-        ]
-
-        for path in all_paths:
-            if not os.path.exists(path):
-                raise FileNotFoundError(f"Missing required raw data source: {path}")
-
-        logger.info("All raw data sources found successfully.")

huggingface-space/huggingface-space/huggingface-space/src/EmotionRecognition/components/data_preparation.py

@@ -1,118 +0,0 @@
-# File: src/EmotionRecognition/components/data_preparation.py
-import os
-import pandas as pd
-import numpy as np
-from PIL import Image
-from tqdm import tqdm
-import shutil
-from EmotionRecognition import logger
-from EmotionRecognition.entity.config_entity import DataPreparationConfig
-from pathlib import Path
-import glob
-
-class DataPreparation:
-    def __init__(self, config: DataPreparationConfig, params: dict):
-        self.config = config
-        self.params = params.DATA_PARAMS
-
-    def _process_and_save(self, best_emotion_name, usage, index, pixels, dataset_prefix):
-        """Helper function to handle the merging logic and save images."""
-        
-        # --- MERGING LOGIC ---
-        # If the emotion is 'contempt', we re-label it as 'disgust'.
-        if best_emotion_name == 'contempt':
-            final_emotion_name = 'disgust'
-        else:
-            final_emotion_name = best_emotion_name
-        # --- END MERGING LOGIC ---
-            
-        # Check if this emotion is one of our final target classes
-        if final_emotion_name in self.params.CLASSES:
-            if usage == 'Training':
-                output_dir = self.config.combined_train_dir
-            elif usage == 'PublicTest':
-                output_dir = self.config.ferplus_test_dir
-            else:
-                return # Skip other usages like PrivateTest
-
-            image = Image.fromarray(pixels)
-            emotion_folder = Path(output_dir) / final_emotion_name
-            emotion_folder.mkdir(parents=True, exist_ok=True)
-            image.save(emotion_folder / f"{dataset_prefix}_{index}.png")
-
-    def _prepare_ferplus(self):
-        logger.info("Starting preparation of FER+ dataset...")
-        pixels_df = pd.read_csv(self.config.ferplus_pixels_csv)
-        labels_df = pd.read_csv(self.config.ferplus_labels_csv)
-        
-        ferplus_emotion_columns = ['neutral', 'happiness', 'surprise', 'sadness', 'anger', 'disgust', 'fear', 'contempt']
-        
-        for index, row in tqdm(pixels_df.iterrows(), total=len(pixels_df), desc="Processing FER+ Images"):
-            label_votes = labels_df.iloc[index][ferplus_emotion_columns].values
-            source_emotion_name = ferplus_emotion_columns[np.argmax(label_votes)]
-
-            # --- STANDARDIZE THE NAME ---
-            # Default to the source name
-            our_emotion_name = source_emotion_name
-            if source_emotion_name == 'happiness': our_emotion_name = 'happy'
-            if source_emotion_name == 'sadness': our_emotion_name = 'sad'
-            if source_emotion_name == 'anger': our_emotion_name = 'angry'
-            if source_emotion_name == 'contempt': our_emotion_name = 'disgust' # MERGE
-            
-            if our_emotion_name in self.params.CLASSES:
-                usage = row['Usage']
-                if usage == 'Training': output_dir = self.config.combined_train_dir
-                elif usage == 'PublicTest': output_dir = self.config.ferplus_test_dir
-                else: continue
-
-                pixels = np.array(row['pixels'].split(), 'uint8').reshape((48, 48))
-                image = Image.fromarray(pixels)
-                emotion_folder = Path(output_dir) / our_emotion_name
-                emotion_folder.mkdir(parents=True, exist_ok=True)
-                image.save(emotion_folder / f"ferplus_{index}.png")
-                
-        logger.info("FER+ dataset preparation complete.")
-
-    def _prepare_ckplus(self):
-        logger.info("Starting preparation of CK+ dataset...")
-        
-        for ckplus_folder_name in tqdm(os.listdir(self.config.ckplus_dir), desc="Processing CK+ Folders"):
-            source_emotion_dir = Path(self.config.ckplus_dir) / ckplus_folder_name
-            
-            # --- STANDARDIZE THE NAME ---
-            our_emotion_name = ckplus_folder_name # Default
-            if ckplus_folder_name == 'contempt': our_emotion_name = 'disgust' # MERGE
-            
-            if our_emotion_name in self.params.CLASSES and source_emotion_dir.is_dir():
-                dest_emotion_dir = Path(self.config.combined_train_dir) / our_emotion_name
-                dest_emotion_dir.mkdir(parents=True, exist_ok=True)
-                
-                for img_file in os.listdir(source_emotion_dir):
-                    shutil.copy(source_emotion_dir / img_file, dest_emotion_dir / f"ckplus_{img_file}")
-        
-        logger.info("CK+ dataset preparation complete.")
-        
-    def _log_dataset_statistics(self):
-        logger.info("--- Final Dataset Statistics ---")
-        logger.info("Training Set:")
-        for emotion in sorted(self.params.CLASSES):
-            count = len(glob.glob(str(self.config.combined_train_dir / emotion / '*.png')))
-            logger.info(f"- {emotion}: {count} images")
-        
-        logger.info("\nTest Set:")
-        for emotion in sorted(self.params.CLASSES):
-            count = len(glob.glob(str(self.config.ferplus_test_dir / emotion / '*.png')))
-            logger.info(f"- {emotion}: {count} images")
-        logger.info("---------------------------------")
-
-    def combine_and_prepare_data(self):
-        logger.info("--- Starting Data Preparation Stage ---")
-        if os.path.exists(self.config.combined_train_dir): shutil.rmtree(self.config.combined_train_dir)
-        if os.path.exists(self.config.ferplus_test_dir): shutil.rmtree(self.config.ferplus_test_dir)
-        os.makedirs(self.config.combined_train_dir, exist_ok=True)
-        os.makedirs(self.config.ferplus_test_dir, exist_ok=True)
-        
-        self._prepare_ferplus()
-        self._prepare_ckplus()
-        self._log_dataset_statistics()
-        logger.info("--- Data Preparation Stage Complete ---")