if model_name.value == "lenet":
        return {"model_name": model_name, "message": "LeCNN all the images"}

def test_get_enums_resnet():
    response = client.get("/models/resnet")
    assert response.status_code == 200
    assert response.json() == {"model_name": "resnet", "message": "Have some residuals"}


def test_get_enums_invalid():
    response = client.get("/models/foo")
    assert response.status_code == 422
    assert response.json() == IsDict(
        {
            "detail": [
                {

			t.Errorf("parsePAXRecord(%q): got (%q: %q), want (%q: %q)",
				v.in, key, val, v.wantKey, v.wantVal)
		}
		if res != v.wantRes {
			t.Errorf("parsePAXRecord(%q): got residual %q, want residual %q",
				v.in, res, v.wantRes)
		}
	}
}

func TestFormatPAXRecord(t *testing.T) {
	medName := strings.Repeat("CD", 50)
	longName := strings.Repeat("AB", 100)

	vectors := []struct {

	// headers since 0.0 headers were not PAX compliant.
	var sparseMap []string

	paxHdrs := make(map[string]string)
	for len(sbuf) > 0 {
		key, value, residual, err := parsePAXRecord(sbuf)
		if err != nil {
			return nil, ErrHeader
		}
		sbuf = residual

		switch key {
		case paxGNUSparseOffset, paxGNUSparseNumBytes:
			// Validate sparse header order and value.

    *   `Estimator` now supports `Dataset`: `input_fn` can return a `Dataset`
        instead of `Tensor`s.
    *   Add `RevBlock`, a memory-efficient implementation of reversible residual
        layers.
    *   Reduce BFCAllocator internal fragmentation.
    *   Add `cross_entropy` and `kl_divergence` to
        `tf.distributions.Distribution`.