so i sarted developing a ml algo.
i stated and downloaded all the replays from one of my algos.
what i did with them doesnt matter, but if someone is interested just ask.
i then traind a model with dense, and flatten layers , and at the last layer i used a dense layer with softmax, every other layer was relu.i then just let the game produce the same 3 dimensional np array as i used for training.i then tried to add the model into it, but i get the error below:
if anyone who tried ml before gotr a similair issue, i would love to know your fix.
SAPlayer 1 algo_0: Random seed: 6529019161501586364
SAPlayer 1 algo_0: ---------------- Starting Your Algo --------------------
SAPlayer 1 algo_0: Configuring your custom algo strategy…
SAPlayer 1 algo_0: 2021-05-13 15:11:16.833650: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
SAPlayer 1 algo_0: 2021-05-13 15:11:16.845944: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2299995000 Hz
SAPlayer 1 algo_0: 2021-05-13 15:11:16.846263: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x20b52f0 executing computations on platform Host. Devices:
SAPlayer 1 algo_0: 2021-05-13 15:11:16.846324: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): ,
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedSequential.call of <tensorflow.python.keras.saving.saved_model.Sequential object at 0x7efc1031f278>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedSequential.call of <tensorflow.python.keras.saving.saved_model.Sequential object at 0x7efc1031f278>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Flatten object at 0x7efc10318d30>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Flatten object at 0x7efc10318d30>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc10318a20>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc10318a20>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031f588>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031f588>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031f898>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031f898>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: WARNING:tensorflow:Entity <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031fba8>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, export AUTOGRAPH_VERBOSITY=10) and attach the full output. Cause: converting <bound method RevivedLayer.call of <tensorflow.python.keras.saving.saved_model.Dense object at 0x7efc1031fba8>>: AssertionError: Bad argument number for Name: 3, expecting 4
SAPlayer 1 algo_0: Traceback (most recent call last):
SAPlayer 1 algo_0: File “//algo_strategy.py”, line 357, in
SAPlayer 1 algo_0: algo.start()
SAPlayer 1 algo_0: File “/gamelib/algocore.py”, line 65, in start
SAPlayer 1 algo_0: self.on_game_start(parsed_config)
SAPlayer 1 algo_0: File “//algo_strategy.py”, line 57, in on_game_start
SAPlayer 1 algo_0: self.model_damage=tf.keras.models.load_model(’/mls/predict_damage/time1620909342.7028215acc0.9935111999511719loss0.02684471756219864.model’)
SAPlayer 1 algo_0: File “/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/saving/save.py”, line 141, in load_model
SAPlayer 1 algo_0: return saved_model.load_from_saved_model_v2(filepath, compile)
SAPlayer 1 algo_0: File “/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/saving/saved_model.py”, line 1225, in load_from_saved_model_v2
SAPlayer 1 algo_0: model._training_config)) # pylint: disable=protected-access
SAPlayer 1 algo_0: File “/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/saving/saving_utils.py”, line 230, in compile_args_from_training_config
SAPlayer 1 algo_0: sample_weight_mode = training_config[‘sample_weight_mode’]
SAPlayer 1 algo_0: KeyError: ‘sample_weight_mode’