Create app.py

app.py

@@ -0,0 +1,84 @@
+import subprocess
+
+# Install necessary packages
+subprocess.run(["pip", "install", "-U", "git+https://github.com/huggingface/transformers.git"])
+subprocess.run(["pip", "install", "-U", "git+https://github.com/huggingface/accelerate.git"])
+subprocess.run(["pip", "install", "datasets"])
+subprocess.run(["pip", "install", "evaluate"])
+subprocess.run(["pip", "install", "torchvision"])
+subprocess.run(["pip", "install", "scikit-learn"])
+
+# Load the necessary libraries
+from datasets import load_dataset
+from transformers import AutoModelForImageClassification, AutoImageProcessor, TrainingArguments, Trainer
+import torch
+import torchvision.transforms as transforms
+import numpy as np
+from evaluate import load
+
+# Load the dataset from Hugging Face Hub
+dataset = load_dataset("DamarJati/Face-Mask-Detection")
+
+# Define the labels
+labels = dataset["train"].features["label"].names
+label2id, id2label = dict(), dict()
+for i, label in enumerate(labels):
+    label2id[label] = i
+    id2label[i] = label
+
+# Load the pre-trained model and processor
+model_checkpoint = "microsoft/resnet-50"
+model = AutoModelForImageClassification.from_pretrained(model_checkpoint, num_labels=len(labels))
+image_processor = AutoImageProcessor.from_pretrained(model_checkpoint)
+
+# Define the image transformations
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=image_processor.image_mean, std=image_processor.image_std)
+])
+
+# Preprocess the dataset
+def preprocess(example_batch):
+    example_batch["pixel_values"] = [transform(image.convert("RGB")) for image in example_batch["image"]]
+    return example_batch
+
+dataset = dataset.with_transform(preprocess)
+
+# Split the dataset into training and validation sets
+splits = dataset["train"].train_test_split(test_size=0.3)
+train_ds = splits['train']
+val_ds = splits['test']
+
+# Define the evaluation metric
+metric = load("accuracy")
+
+def compute_metrics(eval_pred):
+    predictions = np.argmax(eval_pred.predictions, axis=1)
+    return metric.compute(predictions=predictions, references=eval_pred.label_ids)
+
+# Define the data collator
+def collate_fn(examples):
+    pixel_values = torch.stack([example["pixel_values"] for example in examples])
+    labels = torch.tensor([example["label"] for example in examples])
+    return {"pixel_values": pixel_values, "labels": labels}
+
+# Define the training arguments
+args = TrainingArguments(
+    output_dir="./results",
+    per_device_eval_batch_size=128,
+    remove_unused_columns=False,
+)
+
+# Initialize the Trainer
+trainer = Trainer(
+    model=model,
+    args=args,
+    eval_dataset=val_ds,
+    compute_metrics=compute_metrics,
+    data_collator=collate_fn,
+)
+
+# Evaluate the pre-trained model
+metrics = trainer.evaluate()
+print(metrics)