在張量流中，我制作了一個將哈希作為輸入的常規網絡。作為一個例子，我使用了內置的 python hash()函數（是的，它在每個會話中都改變了鹽，但這是一個例子）代碼是這樣的：from time import timest = time()import tensorflow as tfprint(time() - st)import numpy as npimport chessimport atexitfrom numpy import shapedata = open("data.data", "r").readlines()[:10000]targets = open("targets.data", "r").readlines()[:10000]boards_data = []new_targets = []for i in data:    boards_data.append(hash(i))for i in targets:    new_targets.append(float(i))print(len(new_targets))print(len(boards_data))print(np.array(new_targets))print(np.array(boards_data))def create_model():   model = tf.keras.models.Sequential()   model.add(tf.keras.layers.Reshape((1,1,1)))   model.add(tf.keras.layers.Dense(1000, activation="tanh"))   model.add(tf.keras.layers.Flatten())   model.add(tf.keras.layers.Dense(1, activation='tanh'))   model.compile(loss="mse", optimizer="adam", metrics=['accuracy'])   return modelmodel = create_model()model.fit(np.array(boards_data), np.array(new_targets), epochs=10)model.predict(np.array(hash("8/6P1/5k1K/6r1/8/8/8/8 b - - 0 83")))錯誤在預測中。我在如何修復 Tensorflow 中的“IndexError：列表索引超出范圍”中看到了 conv2d 示例 ，但事實并非如此......和回溯：Traceback (most recent call last):  File "/Volumes/POOPOO USB/lichess-bot/engines/engine2/nn_evaluation/nn_evaluation2.py", line 36, in <module>    model.predict(np.array(hash("8/6P1/5k1K/6r1/8/8/8/8 b - - 0 83")))  File "/Users/ofek/Library/Python/3.8/lib/python/site-packages/tensorflow/python/keras/engine/training.py", line 130, in _method_wrapper    return method(self, *args, **kwargs)  File "/Users/ofek/Library/Python/3.8/lib/python/site-packages/tensorflow/python/keras/engine/training.py", line 1569, in predict    data_handler = data_adapter.DataHandler(  File "/Users/ofek/Library/Python/3.8/lib/python/site-packages/tensorflow/python/keras/engine/data_adapter.py", line 1105, in __init__    self._adapter = adapter_cls(