facial_emotion / config.json
sreenathsree1578's picture
Update config.json
be71ec4 verified
{
"model_name": "facial_emotion_simplecnn",
"model_type": "simple_cnn",
"framework": "pytorch",
"task": "image-classification",
"num_classes": 7,
"labels": {
"0": "angry",
"1": "disgust",
"2": "fear",
"3": "happy",
"4": "sad",
"5": "surprise",
"6": "neutral"
},
"in_channels": 1,
"input_size": [48, 48],
"preprocessing": {
"resize": [48, 48],
"normalize_mean": [0.5],
"normalize_std": [0.5],
"color_mode": "grayscale"
},
"training_dataset": "FER2013 (or similar facial emotion dataset)",
"author": "sreenathsree1578",
"license": "mit",
"example_usage": "from facial_emotion import SimpleCNN\nfrom huggingface_hub import hf_hub_download\nimport torch\n\n# Load config\nimport json\nconfig = json.load(open('config.json'))\n\n# Build model\nmodel = SimpleCNN(num_classes=config['num_classes'], in_channels=config['in_channels'])\n\n# Load weights from hub\ncheckpoint = hf_hub_download(repo_id='sreenathsree1578/facial_emotion', filename='pytorch_model.bin')\nmodel.load_state_dict(torch.load(checkpoint, map_location='cpu'))\nmodel.eval()\n\n# Example inference\ntensor = torch.randn(1, 1, 48, 48) # dummy grayscale image\nwith torch.no_grad():\n output = model(tensor)\n pred = torch.argmax(output, dim=1).item()\n print('Predicted class:', config['labels'][str(pred)])"
}