vivit-b-16x2-kinetics400-UCF-Crime

Necesita más información.

Hyperparámetros de entrenamiento:

{
  "learning_rate": 5e-05,
  "train_batch_size": 8,
  "eval_batch_size": 8,
  "seed": 42,
  "optimizer": "Adam con betas=(0.9,0.999) y epsilon=1e-08",
  "lr_scheduler_type": "lineal",
  "lr_scheduler_warmup_ratio": 0.1,
  "training_steps": 3132
}

Resultados de entrenamiento:

[
  { "Epoch": 1,  "Step": 196,  "Validation Loss": 1.6400, "Accuracy": 0.5518 },
  { "Epoch": 2,  "Step": 392,  "Validation Loss": 1.4988, "Accuracy": 0.5634 },
  { "Epoch": 3,  "Step": 588,  "Validation Loss": 1.5328, "Accuracy": 0.5861 },
  { "Epoch": 4,  "Step": 784,  "Validation Loss": 1.3932, "Accuracy": 0.6178 },
  { "Epoch": 5,  "Step": 980,  "Validation Loss": 1.5449, "Accuracy": 0.6060 },
  { "Epoch": 6,  "Step": 1176, "Validation Loss": 1.3075, "Accuracy": 0.6287 },
  { "Epoch": 7,  "Step": 1372, "Validation Loss": 1.6743, "Accuracy": 0.6001 },
  { "Epoch": 8,  "Step": 1568, "Validation Loss": 1.5287, "Accuracy": 0.6172 },
  { "Epoch": 9,  "Step": 1764, "Validation Loss": 1.7749, "Accuracy": 0.6089 },
  { "Epoch": 10, "Step": 1960, "Validation Loss": 1.9027, "Accuracy": 0.6076 },
  { "Epoch": 11, "Step": 2156, "Validation Loss": 1.9622, "Accuracy": 0.6085 },
  { "Epoch": 12, "Step": 2352, "Validation Loss": 2.0872, "Accuracy": 0.6152 },
  { "Epoch": 13, "Step": 2548, "Validation Loss": 2.0095, "Accuracy": 0.6094 },
  { "Epoch": 14, "Step": 2744, "Validation Loss": 1.9398, "Accuracy": 0.6232 },
  { "Epoch": 15, "Step": 2940, "Validation Loss": 1.9994, "Accuracy": 0.6134 },
  { "Epoch": 16, "Step": 3132, "Validation Loss": 1.9757, "Accuracy": 0.6149 }
]

Versiones de los frameworks:

• Transformers 4.33.2
• Pytorch 1.10.0+cu113
• Datasets 2.14.5
• Tokenizers 0.13.3