Adding app files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.avi filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,190 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+*.DS_Store
+data/
+*.gif
+# *.avi
+*.mp4
+*.mp3
+*.wav
+# *.avi
+*.mp4
+/cricketshot-predictor
+/cricketshot
+
+*.DS_Store
+logs/
+*.log

README.md

@@ -12,3 +12,10 @@ short_description: Classify the cricket shot
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# Installtion
+```bash
+conda create -n hf-cricshot python=3.11
+conda activate hf-cricshot
+pip install -r requirements.txt
+```

app.py

@@ -0,0 +1,123 @@
+import os
+import shutil
+from PIL import Image
+import ffmpeg
+import streamlit as st
+import torch
+from transformers import AutoProcessor, AutoModel
+from src.lstm_model import LSTMNetwork
+from src.frames import extract_frames, convert_to_mp4
+
+# Required dictionary
+idx_to_class = {0: 'cover', 1: 'defense', 2: 'flick', 3: 'hook', 4: 'late_cut', 
+                5: 'lofted', 6: 'pull', 7: 'square_cut', 8: 'straight', 9: 'sweep'}
+
+class_label_mapping = {'cover': 0, 'defense': 1, 'flick': 2, 'hook': 3, 'late_cut': 4,
+                      'lofted': 5, 'pull': 6, 'square_cut': 7, 'straight': 8, 'sweep': 9}
+
+# Definig the paths 
+CLIP_MODEL_PATH = "clip-cricket-classifier.pt"
+SIGLIP_MODEL_PATH = "siglip-cricket-classifier.pt"
+
+CLIP_MODEL_ID = "openai/clip-vit-base-patch32"
+SIGLIP_MODEL_ID = "google/siglip-base-patch16-224"
+
+def embeddings_creators(MODEL_ID):
+    embedding_processor = AutoProcessor.from_pretrained(MODEL_ID)
+    embedding_model = AutoModel.from_pretrained(MODEL_ID)
+    embedding_model.to(device)
+    return embedding_processor, embedding_model
+
+def load_model(MODEL_PATH):
+    if MODEL_PATH == CLIP_MODEL_PATH:
+        input_size = 512
+    elif MODEL_PATH == SIGLIP_MODEL_PATH:
+        input_size = 768
+    else:
+        raise ValueError(f"Invalid model path: {MODEL_PATH}")
+    model = LSTMNetwork(input_size=input_size, hidden_size=256, num_classes=10).to(device)
+    model.load_state_dict(torch.load(MODEL_PATH))
+    return model
+
+# device
+device = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+def app():
+    st.image("assets/banner.png")
+    st.title("Cricket Shot Classifier", anchor=False)
+
+    model_choice = st.radio("Select a model", ["None", "CLIP", "SIGLIP"])
+
+    if model_choice == "None":
+        st.stop()
+        st.write("Please select a model")
+
+    if model_choice == "CLIP":
+        embedding_processor, embedding_model = embeddings_creators(CLIP_MODEL_ID)
+        model = load_model(CLIP_MODEL_PATH)
+
+    elif model_choice == "SIGLIP":
+        embedding_processor, embedding_model = embeddings_creators(SIGLIP_MODEL_ID)
+        model = load_model(SIGLIP_MODEL_PATH)
+    
+    # List sample videos from assets folder
+    sample_videos = [f for f in os.listdir("assets") if f.endswith(('.avi'))]
+    if not sample_videos:
+        st.error("No sample videos found in assets folder")
+        st.stop()
+    
+    selected_video = st.selectbox("Select a sample video", sample_videos)
+    video_path = os.path.join("assets", selected_video)
+    
+    save_directory = './demo'
+    os.makedirs(save_directory, exist_ok=True)
+    new_video_path = f"{save_directory}/{selected_video}"
+    shutil.copy2(video_path, new_video_path)
+        
+
+    final_video_path = f"{save_directory}/{os.path.splitext(os.path.basename(new_video_path))[0]}.mp4"
+    
+    if not new_video_path.lower().endswith('.mp4'):
+        convert_to_mp4(new_video_path, final_video_path)
+    else:
+        final_video_path = new_video_path
+
+    st.video(final_video_path)
+
+    frames_dir = f"{save_directory}/frames"
+    os.makedirs(frames_dir, exist_ok=True)
+    extract_frames(final_video_path, frames_dir)
+    st.write("Frames extracted from the video.")
+
+    inference_paths = [os.path.join(frames_dir, f) for f in os.listdir(frames_dir) if f.endswith(('.jpg', '.jpeg', '.png'))]
+    inference_images = [Image.open(path).convert("RGB") for path in inference_paths]
+    tokens = embedding_processor(
+        text=None,
+        images=inference_images,
+        return_tensors="pt"
+    ).to(device)
+    inference_embeddings = embedding_model.get_image_features(**tokens)
+
+    with torch.no_grad():
+        output = model(inference_embeddings.unsqueeze(0))
+        prob = output.softmax(dim=1)
+        
+        _, indices = torch.sort(prob[0], descending=True)
+        
+        for idx in indices:
+            i = idx.item()
+            st.write(f"Prediction: {idx_to_class[i]}")
+            st.progress(int(prob[0][i].item() * 100))
+
+    try:
+        shutil.rmtree(frames_dir)
+        os.remove(new_video_path)
+        os.remove(final_video_path)
+        print(f"Folder '{frames_dir}' and its contents have been deleted.")
+    except Exception as e:
+        print(f"Error while deleting folder '{frames_dir}': {e}")
+
+
+if __name__ == "__main__":
+    app()
+

assets/cover_0006.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cda892e8d464a6f78eda93331d65d2f92a8b8de185a404730586d5db5e8de55a
+size 1160320

assets/defense_0007.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1b386f7c57c1031a7b253be4d810ca05764461d8dbb58a7be4ae336f351bfd0
+size 2545010

assets/flick_0008.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b19dc6d3dcef0b153e0b72b7b89f09dae52eada5d87c42ecbfb85e5d83d27a17
+size 1303944

assets/hook_0009.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa88e98e553ecd73b7c6a8d260f0347aceb524ecb591baffc6f61072a2b379bb
+size 1038106

assets/late_cut_0010.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:477e6b970bb1ed9b279a4e10524e77ba5945d1c372aec2aa6d2c077863163196
+size 1222750

assets/lofted_0011.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:665a02766349596622e916616492f6a3abef98808b84bd6540fbc4e6e53b5d7e
+size 3914000

assets/pull_0010.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f8e5d088e6d863542d057358ec5a8aecda8670380158ece462fc003fb785b7
+size 1239450

assets/square_cut_0011.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dab8de01051da67ce94d81f5cfedfc1c89dab9cb190af9aa585f9414570d6d9
+size 1535956

assets/straight_0012.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76592c312ced400c1a123e7931d644348f52ec05be63b983929df652c366326d
+size 2034822

assets/sweep_0013.avi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82fa15ab99021da4e24ccaf26672c79ff66c5990ddc62d44adca6ed939a5a7b8
+size 2164312

clip-cricket-classifier.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb97f8950a29bd41e8af501cc53a503b50bced86314701fa505554c33d1f66c8
+size 3167016

requirements.txt

@@ -0,0 +1,14 @@
+transformers
+datasets
+evaluate
+imageio
+huggingface-hub
+git+https://github.com/facebookresearch/pytorchvideo
+accelerate>=0.26.0
+scikit-learn
+python-dotenv
+sentencepiece
+protobuf
+torch
+torchvision
+ffmpeg-python

siglip-cricket-classifier.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42d8f0ff10ce0a7a40d860a8d5c3a96f46a07f6bd6fb4f832ede61660d919a97
+size 4215612

src/frames.py

@@ -0,0 +1,14 @@
+
+import os
+import ffmpeg
+
+def extract_frames(video_path: str, frames_dir: str) -> None:
+    """Extracts frames from a video file using FFmpeg."""
+    output_pattern = os.path.join(frames_dir, "video_frame_%04d.jpg")
+    ffmpeg.input(video_path).output(output_pattern, vf='fps=5', loglevel='quiet').run()
+
+def convert_to_mp4(input_path: str, output_path: str) -> None:
+    """Converts a video file to MP4 using FFmpeg."""
+    ffmpeg.input(input_path).output(output_path).run()
+
+

src/lstm_model.py

@@ -0,0 +1,12 @@
+import torch.nn as nn
+
+class LSTMNetwork(nn.Module):
+    def __init__(self, input_size=768, hidden_size=256, num_classes=4):
+        super(LSTMNetwork, self).__init__()
+        self.lstm = nn.LSTM(input_size=input_size, hidden_size=hidden_size, num_layers=1, batch_first=True)
+        self.fc = nn.Linear(hidden_size, num_classes)
+
+    def forward(self, x):
+        x, _ = self.lstm(x)
+        x = self.fc(x[:, -1, :])  
+        return x