[
    {
      "id": "basic_reduce",
      "description": "Given array x, compute mean of each row.",
      "input": "data/x.npy",
      "output": "basic_reduce.npy"
    },
    {
      "id": "map_square",
      "description": "Apply square function to each element using vectorization.",
      "input": "data/x.npy",
      "output": "map_square.npy"
    },
    {
      "id": "grad_logistic",
      "description": "Compute gradient of logistic loss.",
      "input": "data/logistic.npz",
      "output": "grad_logistic.npy"
    },
    {
      "id": "scan_rnn",
      "description": "Implement RNN forward pass using scan.",
      "input": "data/seq.npz",
      "output": "scan_rnn.npy"
    },
    {
      "id": "jit_mlp",
      "description": "Implement and JIT compile a 2-layer MLP.",
      "input": "data/mlp.npz",
      "output": "jit_mlp.npy"
    }
  ]