Python 如何为TensorFlow中的矢量化参数设置双射体？

我将遵循本教程第3个示例中关于的逻辑。然而，我的设置中的一个不同点是，我的振幅和长度_标度是向量。但是，我很难为矢量化参数设置双射体

# Create a list to save all variables to be iterated.
initial_chain_states = [
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_amp_1"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_scale_1"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_amp_0"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_scale_0"),
    tf.ones([], dtype=tf.float32, name="init_sigma_sq_1"),
    tf.ones([], dtype=tf.float32, name="init_sigma_sq_0")
]

vectorized_sp = tfb.Softplus(hinge_softness=np.ones([1, num_GPs], dtype=np.float32))

unconstrained_bijectors = [
    vectorized_sp,
    vectorized_sp,
    vectorized_sp,
    vectorized_sp,
    tfp.bijectors.Softplus(),
    tfp.bijectors.Softplus()
]

def un_normalized_log_posterior(amplitude_1, length_scale_1,
                                amplitude_0, length_scale_0,
                                noise_var_1, noise_var_0):
    return compute_joint_log_prob_3(
        para_index, delayed_signal, y_type,
        amplitude_1, length_scale_1, amplitude_0, length_scale_0,
        noise_var_1, noise_var_0
    )

num_results = 200
[
    amps_1,
    scales_1,
    amps_0,
    scales_0,
    sigma_sqs_1,
    sigma_sqs_0
], kernel_results = tfp.mcmc.sample_chain(
    num_results=num_results,
    num_burnin_steps=250,
    num_steps_between_results=3,
    current_state=initial_chain_states,
    kernel=tfp.mcmc.TransformedTransitionKernel(
        inner_kernel=tfp.mcmc.HamiltonianMonteCarlo(
            target_log_prob_fn=un_normalized_log_posterior,
            step_size=np.float32(0.1),
            num_leapfrog_steps=3,
            step_size_update_fn=tfp.mcmc.make_simple_step_size_update_policy(
                num_adaptation_steps=100)),
        bijector=unconstrained_bijectors))

我尝试了官方示例教程中的一种方法（并搜索关键字“Batching Bijectors”）

他们用

softplus = tfp.bijectors.Softplus(
  hinge_softness=[1., .5, .1])
print("Hinge softness shape:", softplus.hinge_softness.shape)

为标量参数更改Softplus的形状。但控制台一直显示相同的错误消息

My

compute_joint_log_prob_3

仅输出给定所有数据和参数的标量对数后验概率。我已经测试过这个函数运行得很好。唯一的问题是在存在向量化内核超参数的情况下设置无约束的对象

# Create a list to save all variables to be iterated.
initial_chain_states = [
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_amp_1"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_scale_1"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_amp_0"),
    tf.ones([1, num_GPs], dtype=tf.float32, name="init_scale_0"),
    tf.ones([], dtype=tf.float32, name="init_sigma_sq_1"),
    tf.ones([], dtype=tf.float32, name="init_sigma_sq_0")
]

vectorized_sp = tfb.Softplus(hinge_softness=np.ones([1, num_GPs], dtype=np.float32))

unconstrained_bijectors = [
    vectorized_sp,
    vectorized_sp,
    vectorized_sp,
    vectorized_sp,
    tfp.bijectors.Softplus(),
    tfp.bijectors.Softplus()
]

def un_normalized_log_posterior(amplitude_1, length_scale_1,
                                amplitude_0, length_scale_0,
                                noise_var_1, noise_var_0):
    return compute_joint_log_prob_3(
        para_index, delayed_signal, y_type,
        amplitude_1, length_scale_1, amplitude_0, length_scale_0,
        noise_var_1, noise_var_0
    )

num_results = 200
[
    amps_1,
    scales_1,
    amps_0,
    scales_0,
    sigma_sqs_1,
    sigma_sqs_0
], kernel_results = tfp.mcmc.sample_chain(
    num_results=num_results,
    num_burnin_steps=250,
    num_steps_between_results=3,
    current_state=initial_chain_states,
    kernel=tfp.mcmc.TransformedTransitionKernel(
        inner_kernel=tfp.mcmc.HamiltonianMonteCarlo(
            target_log_prob_fn=un_normalized_log_posterior,
            step_size=np.float32(0.1),
            num_leapfrog_steps=3,
            step_size_update_fn=tfp.mcmc.make_simple_step_size_update_policy(
                num_adaptation_steps=100)),
        bijector=unconstrained_bijectors))

它应该会起作用，模型会抽取这些参数的样本。相反，我收到了一堆错误信息，上面说

Traceback (most recent call last):
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1659, in _create_c_op
    c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Requires start <= limit when delta > 0: 1/0 for 'mcmc_sample_chain/transformed_kernel_bootstrap_results/mh_bootstrap_results/hmc_kernel_bootstrap_results/maybe_call_fn_and_grads/value_and_gradients/softplus_10/forward_log_det_jacobian/range' (op: 'Range') with input shapes: [], [], [] and with computed input tensors: input[0] = <1>, input[1] = <0>, input[2] = <1>.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py", line 183, in _run_module_as_main
    mod_name, mod_spec, code = _get_module_details(mod_name, _Error)
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py", line 109, in _get_module_details
    __import__(pkg_name)
  File "/MMAR_q/MMAR_q.py", line 237, in <module>
    bijector=unconstrained_bijectors))
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/sample.py", line 235, in sample_chain
    previous_kernel_results = kernel.bootstrap_results(current_state)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/transformed_kernel.py", line 344, in bootstrap_results
    transformed_init_state))
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/hmc.py", line 518, in bootstrap_results
    kernel_results = self._impl.bootstrap_results(init_state)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/metropolis_hastings.py", line 264, in bootstrap_results
    pkr = self.inner_kernel.bootstrap_results(init_state)
  File "/MAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/hmc.py", line 687, in bootstrap_results
    ] = mcmc_util.maybe_call_fn_and_grads(self.target_log_prob_fn, init_state)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/util.py", line 237, in maybe_call_fn_and_grads
    result, grads = _value_and_gradients(fn, fn_arg_list, result, grads)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/util.py", line 185, in _value_and_gradients
    result = fn(*fn_arg_list)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/transformed_kernel.py", line 204, in new_target_log_prob
    event_ndims=event_ndims)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/transformed_kernel.py", line 51, in fn
    for b, e, sp in zip(bijector, event_ndims, transformed_state_parts)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/mcmc/transformed_kernel.py", line 51, in <listcomp>
    for b, e, sp in zip(bijector, event_ndims, transformed_state_parts)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/bijectors/bijector.py", line 1205, in forward_log_det_jacobian
    return self._call_forward_log_det_jacobian(x, event_ndims, name)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/bijectors/bijector.py", line 1177, in _call_forward_log_det_jacobian
    kwargs=kwargs)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/bijectors/bijector.py", line 982, in _compute_inverse_log_det_jacobian_with_caching
    event_ndims)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/bijectors/bijector.py", line 1272, in _reduce_jacobian_det_over_event
    axis=self._get_event_reduce_dims(min_event_ndims, event_ndims))
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow_probability/python/bijectors/bijector.py", line 1284, in _get_event_reduce_dims
    return tf.range(-reduce_ndims, 0)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py", line 1199, in range
    return gen_math_ops._range(start, limit, delta, name=name)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/ops/gen_math_ops.py", line 6746, in _range
    "Range", start=start, limit=limit, delta=delta, name=name)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
    op_def=op_def)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
    return func(*args, **kwargs)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 3300, in create_op
    op_def=op_def)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1823, in __init__
    control_input_ops)
  File "/MMAR_q/venv/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1662, in _create_c_op
    raise ValueError(str(e))
ValueError: Requires start <= limit when delta > 0: 1/0 for 'mcmc_sample_chain/transformed_kernel_bootstrap_results/mh_bootstrap_results/hmc_kernel_bootstrap_results/maybe_call_fn_and_grads/value_and_gradients/softplus_10/forward_log_det_jacobian/range' (op: 'Range') with input shapes: [], [], [] and with computed input tensors: input[0] = <1>, input[1] = <0>, input[2] = <1>.

回溯（最近一次呼叫最后一次）：
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/framework/ops.py”，第1659行，在“创建”c_op中
c_op=c_api.TF_FinishOperation（操作说明）
tensorflow.python.framework.errors\u impl.invalidargumeinterror:需要0:1/0开始“mcmc\u样本链/转换的\u内核\u引导\u结果/mh\u引导\u结果/hmc\u内核\u引导\u结果/可能\u调用\u fn\u和\u梯度/softplus\u 10/向前\u日志\u det\u雅可比数/range”（op:“range”），输入形状：[]、[]、[]、[]使用计算的输入张量：输入[0]=，输入[1]=，输入[2]=。
在处理上述异常期间，发生了另一个异常：
回溯（最近一次呼叫最后一次）：
文件“/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py”，第183行，在运行模块中
模块名称，模块规格，代码=\u获取模块详细信息（模块名称，错误）
文件“/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py”，第109行，在获取模块详细信息中
__导入（包装名称）
文件“/MMAR_q/MMAR_q.py”，第237行，在
双喷射器=无约束（双喷射器）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/sample.py”，第235行，在sample_链中
先前的\u内核\u结果=内核.bootstrap\u结果（当前状态）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/transformed_kernel.py”，第344行，在引导结果中
转换的（初始状态）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/hmc.py”，第518行，在bootstrap_结果中
内核结果=自执行引导结果（初始状态）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/metropolis_hastings.py”，第264行，在引导结果中
pkr=self.internal\u kernel.bootstrap\u结果（init\u状态）
bootstrap结果中的文件“/MAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmcmc/hmc.py”，第687行
]=mcmc_util.maybe_call_fn_和_grads（self.target_log_prob_fn，初始状态）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/util.py”，第237行，在maybe_call_fn_和_grads中
结果，梯度=_值和_梯度（fn，fn_arg_列表，结果，梯度）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/util.py”，第185行，输入值和梯度
结果=fn（*fn_参数_列表）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/transformed_kernel.py”，第204行，在new_target_log_prob中
事件（ndims=事件（ndims）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/transformed_kernel.py”，第51行，fn
对于zip中的b、e、sp（对象、事件、状态、转换部分）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/mcmc/transformed_kernel.py”，第51行，在
对于zip中的b、e、sp（对象、事件、状态、转换部分）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/bijectors/bijector.py”，第1205行，在forward_log_det_jacobian中
返回self.\u调用\u转发\u日志\u det\u jacobian（x，事件ndims，名称）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/bijectors/bijector.py”，第1177行，在调用转发日志中
kwargs=kwargs）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/bijectors/bijector.py”，第982行，在带有缓存的计算逆计算日志中
事件（ndims）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/bijectors/bijector.py”，第1272行，在“reduce_jacobian_det_over_”事件中
轴=自身。\获取\事件\减少\ dims（最小\事件\ ndims，事件\ ndims））
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow_probability/python/bijectors/bijector.py”，第1284行，在“获取事件”中
返回tf.范围（-reduce_ndims，0）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/ops/math_ops.py”，第1199行，范围内
返回发电机数学运算范围（起始、极限、增量、名称=名称）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/ops/gen_math_ops.py”，第6746行，在_范围内
“范围”，开始=开始，限制=限制，增量=增量，名称=名称）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/framework/op_def_library.py”，第788行，在“应用”op_helper中
op_def=op_def）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/util/deprecation.py”，第507行，在new_func中
返回函数（*args，**kwargs）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/framework/ops.py”，第3300行，在create_op中
op_def=op_def）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/framework/ops.py”，第1823行，在__
控制（输入操作）
文件“/MMAR_q/venv/lib/python3.7/site packages/tensorflow/python/framework/ops.py”，第1662行，在_create_c_op
提升值错误（str（e））
ValueError:Re