ValueError: Issue with tf.squeeze Expected Dimension in construct_model.py

[mayisme]

Hello,

I hope this message finds you well. I'm reaching out for assistance with an issue I've encountered while running construct_model.py using the autoXRD library.

Initially, I used my own CIF files sourced from the Crystallography Open Database, converting them with pymatgen to match the format of the example CIF files provided by your library. During this process, I encountered a ValueError:

ValueError: Cannot squeeze dim[1], expected a dimension of 1, got 9 for '{{node Squeeze}} = SqueezeT=DT_INT32, squeeze_dims=[-1]' with input shapes: [32,9]. This error suggests there is an issue with the tf.squeeze operation, where it is expecting a size of 1 in the last dimension but is encountering a size of 9. The tensor shape causing the issue is [32, 9].

To rule out the possibility of the error being related to my data, I replaced my CIF files with some of the example CIF files provided by your library. Unfortunately, the same issue occurred.

I have verified that the shapes of the input data conform to the model’s expected input shape, yet the error still occurs. Could you provide some guidance on what might be causing this error, or suggest any debugging steps I could take?

I'm happy to provide additional details or the full stack trace if it would be helpful for diagnosing the problem.

I appreciate any help you can give me on this matter.

Thank you for your time and assistance.

Kind regards

[mayisme]

generated using pymatgen

data_Al(SiO3)2 _symmetry_space_group_name_H-M 'P 1' _cell_length_a 5.14000000 _cell_length_b 8.90000000 _cell_length_c 18.55000000 _cell_angle_alpha 90.00000000 _cell_angle_beta 99.92000000 _cell_angle_gamma 90.00000000 _symmetry_Int_Tables_number 1 _chemical_formula_structural Al(SiO3)2 _chemical_formula_sum 'Al8 Si16 O48' _cell_volume 835.90126980 _cell_formula_unitsZ 8 loop _symmetry_equiv_pos_site_id _symmetry_equiv_pos_asxyz 1 'x, y, z' loop _atom_site_type_symbol _atom_site_label _atom_site_symmetry_multiplicity _atom_site_fract_x _atom_site_fract_y _atom_site_fract_z _atom_site_occupancy Al Al0 1 0.00000000 0.33300000 0.00000000 1.0 Al Al1 1 0.50000000 0.83300000 0.00000000 1.0 Al Al2 1 0.00000000 0.66700000 0.50000000 1.0 Al Al3 1 0.50000000 0.16700000 0.50000000 1.0 Al Al4 1 0.00000000 0.33300000 0.50000000 1.0 Al Al5 1 0.50000000 0.83300000 0.50000000 1.0 Al Al6 1 0.00000000 0.66700000 0.00000000 1.0 Al Al7 1 0.50000000 0.16700000 0.00000000 1.0 Si Si1 1 0.76100000 0.00000000 0.14300000 1.0 Si Si1 1 0.26100000 0.50000000 0.14300000 1.0 Si Si1 1 0.76100000 0.00000000 0.64300000 1.0 Si Si1 1 0.26100000 0.50000000 0.64300000 1.0 Si Si1 1 0.23900000 0.00000000 0.35700000 1.0 Si Si1 1 0.73900000 0.50000000 0.35700000 1.0 Si Si1 1 0.23900000 0.00000000 0.85700000 1.0 Si Si1 1 0.73900000 0.50000000 0.85700000 1.0 Si Si2 1 0.26100000 0.16700000 0.14300000 1.0 Si Si2 1 0.76100000 0.66700000 0.14300000 1.0 Si Si2 1 0.26100000 0.83300000 0.64300000 1.0 Si Si2 1 0.76100000 0.33300000 0.64300000 1.0 Si Si2 1 0.73900000 0.16700000 0.35700000 1.0 Si Si2 1 0.23900000 0.66700000 0.35700000 1.0 Si Si2 1 0.73900000 0.83300000 0.85700000 1.0 Si Si2 1 0.23900000 0.33300000 0.85700000 1.0 O O1 1 0.20300000 0.50000000 0.05800000 1.0 O O1 1 0.70300000 0.00000000 0.05800000 1.0 O O1 1 0.20300000 0.50000000 0.55800000 1.0 O O1 1 0.70300000 0.00000000 0.55800000 1.0 O O1 1 0.79700000 0.50000000 0.44200000 1.0 O O1 1 0.29700000 0.00000000 0.44200000 1.0 O O1 1 0.79700000 0.50000000 0.94200000 1.0 O O1 1 0.29700000 0.00000000 0.94200000 1.0 O O2 1 0.20300000 0.16700000 0.05800000 1.0 O O2 1 0.70300000 0.66700000 0.05800000 1.0 O O2 1 0.20300000 0.83300000 0.55800000 1.0 O O2 1 0.70300000 0.33300000 0.55800000 1.0 O O2 1 0.79700000 0.16700000 0.44200000 1.0 O O2 1 0.29700000 0.66700000 0.44200000 1.0 O O2 1 0.79700000 0.83300000 0.94200000 1.0 O O2 1 0.29700000 0.33300000 0.94200000 1.0 O O-H1 1 0.20300000 0.83300000 0.05800000 1.0 O O-H1 1 0.70300000 0.33300000 0.05800000 1.0 O O-H1 1 0.20300000 0.16700000 0.55800000 1.0 O O-H1 1 0.70300000 0.66700000 0.55800000 1.0 O O-H1 1 0.79700000 0.83300000 0.44200000 1.0 O O-H1 1 0.29700000 0.33300000 0.44200000 1.0 O O-H1 1 0.79700000 0.16700000 0.94200000 1.0 O O-H1 1 0.29700000 0.66700000 0.94200000 1.0 O O3 1 0.02500000 0.08300000 0.17600000 1.0 O O3 1 0.52500000 0.58300000 0.17600000 1.0 O O3 1 0.02500000 0.91700000 0.67600000 1.0 O O3 1 0.52500000 0.41700000 0.67600000 1.0 O O3 1 0.97500000 0.08300000 0.32400000 1.0 O O3 1 0.47500000 0.58300000 0.32400000 1.0 O O3 1 0.97500000 0.91700000 0.82400000 1.0 O O3 1 0.47500000 0.41700000 0.82400000 1.0 O O4 1 0.52500000 0.08300000 0.17600000 1.0 O O4 1 0.02500000 0.58300000 0.17600000 1.0 O O4 1 0.52500000 0.91700000 0.67600000 1.0 O O4 1 0.02500000 0.41700000 0.67600000 1.0 O O4 1 0.47500000 0.08300000 0.32400000 1.0 O O4 1 0.97500000 0.58300000 0.32400000 1.0 O O4 1 0.47500000 0.91700000 0.82400000 1.0 O O4 1 0.97500000 0.41700000 0.82400000 1.0 O O5 1 0.27500000 0.33300000 0.17600000 1.0 O O5 1 0.77500000 0.83300000 0.17600000 1.0 O O5 1 0.27500000 0.66700000 0.67600000 1.0 O O5 1 0.77500000 0.16700000 0.67600000 1.0 O O5 1 0.72500000 0.33300000 0.32400000 1.0 O O5 1 0.22500000 0.83300000 0.32400000 1.0 O O5 1 0.72500000 0.66700000 0.82400000 1.0 O O5 1 0.22500000 0.16700000 0.82400000 1.0 my Own data is like this

[njszym]

Can you provide me with the full error/output? At what line in the code does this ValueError occur?

[mayisme]

Hi，the full error is as follows: Traceback (most recent call last): File "/Users/xiaoyf/Documents/Jupyterlab/XRD-AutoAnalyzer/Novel-Space/construct_model.py", line 107, in cnn.main(xrd_specs, num_epochs, test_fraction, is_pdf=False) File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/autoXRD/cnn/init.py", line 210, in main model = train_model(train_x, train_y, num_phases, num_epochs, is_pdf) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/autoXRD/cnn/init.py", line 184, in train_model model.fit(x_train, y_train, batch_size=32, epochs=num_epochs, File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/utils/traceback_utils.py", line 123, in error_handler raise e.with_traceback(filtered_tb) from None File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/backend/tensorflow/numpy.py", line 1767, in squeeze return tf.squeeze(x, axis=axis) ^^^^^^^^^^^^^^^^^^^^^^^^ ValueError: Can not squeeze dim[1], expected a dimension of 1, got 9 for '{{node Squeeze}} = SqueezeT=DT_INT32, squeeze_dims=[-1]' with input shapes: [32,9].

[mayisme]

Hi, when I try to modify def y(self) as below, the Problem was solved. @property def y(self): """ Target property to predict (one-hot encoded vectors associated with the reference phases) """ n_phases = len(self.xrd) # 假设 self.xrd 是所有不同相位的集合 phase_indices = self.phase_indices one_hot_vectors = [] for index in phase_indices: one_hot_vector = [0] * n_phases one_hot_vector[index] = 1 one_hot_vectors.append(one_hot_vector) return np.array(one_hot_vectors)

however when I try to execute the run_CNN.py with new trained Model.h5, a new issue came up. the full error is as follows: Traceback (most recent call last): File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/ops/operation.py", line 196, in from_config return cls(**config) ^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/layers/core/dense.py", line 87, in init self.activation = activations.get(activation) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/activations/init.py", line 104, in get raise ValueError( ValueError: Could not interpret activation function identifier: {'module': 'builtins', 'class_name': 'function', 'config': 'softmax_v2', 'registered_name': 'function'}

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/Users/xiaoyf/Documents/Jupyterlab/XRD-AutoAnalyzer/Novel-Space/run_CNN1.py", line 33, in model = load_model('Model.h5', compile=False) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/saving/saving_api.py", line 183, in load_model return legacy_h5_format.load_model_from_hdf5(filepath) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/legacy/saving/legacy_h5_format.py", line 133, in load_model_from_hdf5 model = saving_utils.model_from_config( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/legacy/saving/serialization.py", line 495, in deserialize_keras_object deserialized_obj = cls.from_config( ^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/models/sequential.py", line 327, in from_config layer = saving_utils.model_from_config( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/legacy/saving/serialization.py", line 504, in deserialize_keras_object deserialized_obj = cls.from_config(cls_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/xiaoyf/Library/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/keras/src/ops/operation.py", line 198, in from_config raise TypeError( TypeError: Error when deserializing class 'Dense' using config={'name': 'dense_2', 'trainable': True, 'dtype': 'float32', 'units': 9, 'activation': {'module': 'builtins', 'class_name': 'function', 'config': 'softmax_v2', 'registered_name': 'function'}, 'use_bias': True, 'kernel_initializer': {'module': 'keras.initializers', 'class_name': 'GlorotUniform', 'config': {'seed': None}, 'registered_name': None}, 'bias_initializer': {'module': 'keras.initializers', 'class_name': 'Zeros', 'config': {}, 'registered_name': None}, 'kernel_regularizer': None, 'bias_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}.

Exception encountered: Could not interpret activation function identifier: {'module': 'builtins', 'class_name': 'function', 'config': 'softmax_v2', 'registered_name': 'function'}

[njszym]

Can you provide me with a zipped version of your run folder? This would help me determine what's causing the issue.

[mayisme]

my training accuracy is 0, how to improve the model?

Training data shape: (74, 4501, 1) Training labels shape: (74, 92) Epoch 1/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 3s 1s/step - categorical_accuracy: 0.0000e+00 - loss: 7.0845 - val_categorical_accuracy: 0.0000e+00 - val_loss: 14.5905 Epoch 2/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 400ms/step - categorical_accuracy: 0.0217 - loss: 7.3922 - val_categorical_accuracy: 0.1333 - val_loss: 46.1377 Epoch 3/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 404ms/step - categorical_accuracy: 0.0434 - loss: 6.9763 - val_categorical_accuracy: 0.0667 - val_loss: 73.0654 Epoch 4/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 414ms/step - categorical_accuracy: 0.0547 - loss: 7.3809 - val_categorical_accuracy: 0.0000e+00 - val_loss: 90.2825 Epoch 5/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 412ms/step - categorical_accuracy: 0.0113 - loss: 6.6131 - val_categorical_accuracy: 0.0000e+00 - val_loss: 93.7274 Epoch 6/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 419ms/step - categorical_accuracy: 0.0217 - loss: 6.9738 - val_categorical_accuracy: 0.0000e+00 - val_loss: 79.6283 Epoch 7/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 409ms/step - categorical_accuracy: 0.0226 - loss: 7.0533 - val_categorical_accuracy: 0.0000e+00 - val_loss: 76.9755 Epoch 8/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 407ms/step - categorical_accuracy: 0.0000e+00 - loss: 6.7895 - val_categorical_accuracy: 0.0000e+00 - val_loss: 64.8061 Epoch 9/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 404ms/step - categorical_accuracy: 0.0000e+00 - loss: 7.0489 - val_categorical_accuracy: 0.0000e+00 - val_loss: 59.4355 Epoch 10/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 398ms/step - categorical_accuracy: 0.0547 - loss: 6.4233 - val_categorical_accuracy: 0.0000e+00 - val_loss: 66.9305 Epoch 11/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.0217 - loss: 6.1931 - val_categorical_accuracy: 0.0000e+00 - val_loss: 54.7749 Epoch 12/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 421ms/step - categorical_accuracy: 0.0651 - loss: 5.7802 - val_categorical_accuracy: 0.0000e+00 - val_loss: 57.8014 Epoch 13/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.0000e+00 - loss: 5.9988 - val_categorical_accuracy: 0.0667 - val_loss: 52.5752 Epoch 14/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 408ms/step - categorical_accuracy: 0.0443 - loss: 6.3365 - val_categorical_accuracy: 0.0000e+00 - val_loss: 69.5286 Epoch 15/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 416ms/step - categorical_accuracy: 0.0764 - loss: 5.9680 - val_categorical_accuracy: 0.0667 - val_loss: 62.5044 Epoch 16/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 430ms/step - categorical_accuracy: 0.0990 - loss: 4.8516 - val_categorical_accuracy: 0.0667 - val_loss: 50.8224 Epoch 17/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 413ms/step - categorical_accuracy: 0.0660 - loss: 5.5496 - val_categorical_accuracy: 0.0000e+00 - val_loss: 48.6654 Epoch 18/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 404ms/step - categorical_accuracy: 0.1199 - loss: 4.8782 - val_categorical_accuracy: 0.0000e+00 - val_loss: 45.4010 Epoch 19/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 441ms/step - categorical_accuracy: 0.1642 - loss: 5.0994 - val_categorical_accuracy: 0.0000e+00 - val_loss: 48.2345 Epoch 20/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 407ms/step - categorical_accuracy: 0.1103 - loss: 5.6798 - val_categorical_accuracy: 0.0000e+00 - val_loss: 41.5763 Epoch 21/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 397ms/step - categorical_accuracy: 0.1086 - loss: 4.6253 - val_categorical_accuracy: 0.0000e+00 - val_loss: 44.2621 Epoch 22/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 404ms/step - categorical_accuracy: 0.1642 - loss: 4.8132 - val_categorical_accuracy: 0.0000e+00 - val_loss: 40.7891 Epoch 23/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 420ms/step - categorical_accuracy: 0.0660 - loss: 4.8918 - val_categorical_accuracy: 0.0000e+00 - val_loss: 42.6678 Epoch 24/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 402ms/step - categorical_accuracy: 0.1199 - loss: 4.9616 - val_categorical_accuracy: 0.0000e+00 - val_loss: 38.0936 Epoch 25/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 406ms/step - categorical_accuracy: 0.1651 - loss: 4.3112 - val_categorical_accuracy: 0.0000e+00 - val_loss: 43.2405 Epoch 26/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 425ms/step - categorical_accuracy: 0.1755 - loss: 4.8295 - val_categorical_accuracy: 0.0000e+00 - val_loss: 38.3020 Epoch 27/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 416ms/step - categorical_accuracy: 0.1416 - loss: 4.7973 - val_categorical_accuracy: 0.0000e+00 - val_loss: 37.0159 Epoch 28/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 412ms/step - categorical_accuracy: 0.1199 - loss: 4.9667 - val_categorical_accuracy: 0.0000e+00 - val_loss: 40.4837 Epoch 29/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.0877 - loss: 4.5363 - val_categorical_accuracy: 0.0000e+00 - val_loss: 35.5006 Epoch 30/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 412ms/step - categorical_accuracy: 0.1329 - loss: 4.2568 - val_categorical_accuracy: 0.0000e+00 - val_loss: 37.3278 Epoch 31/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 409ms/step - categorical_accuracy: 0.1095 - loss: 4.6527 - val_categorical_accuracy: 0.0000e+00 - val_loss: 32.4564 Epoch 32/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 403ms/step - categorical_accuracy: 0.1529 - loss: 4.0403 - val_categorical_accuracy: 0.0667 - val_loss: 29.2192 Epoch 33/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 411ms/step - categorical_accuracy: 0.1538 - loss: 4.0359 - val_categorical_accuracy: 0.0000e+00 - val_loss: 25.5436 Epoch 34/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 419ms/step - categorical_accuracy: 0.2198 - loss: 3.7955 - val_categorical_accuracy: 0.0000e+00 - val_loss: 23.9804 Epoch 35/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 419ms/step - categorical_accuracy: 0.1972 - loss: 4.4588 - val_categorical_accuracy: 0.0000e+00 - val_loss: 20.5819 Epoch 36/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.1981 - loss: 3.6381 - val_categorical_accuracy: 0.0667 - val_loss: 16.0094 Epoch 37/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 416ms/step - categorical_accuracy: 0.1886 - loss: 3.6445 - val_categorical_accuracy: 0.0000e+00 - val_loss: 17.8915 Epoch 38/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 413ms/step - categorical_accuracy: 0.1651 - loss: 4.0391 - val_categorical_accuracy: 0.0000e+00 - val_loss: 16.6869 Epoch 39/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.2311 - loss: 3.7243 - val_categorical_accuracy: 0.0000e+00 - val_loss: 19.8661 Epoch 40/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 419ms/step - categorical_accuracy: 0.2198 - loss: 3.5134 - val_categorical_accuracy: 0.0000e+00 - val_loss: 23.7789 Epoch 41/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 408ms/step - categorical_accuracy: 0.0782 - loss: 4.0584 - val_categorical_accuracy: 0.0000e+00 - val_loss: 22.0357 Epoch 42/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 418ms/step - categorical_accuracy: 0.2841 - loss: 3.4871 - val_categorical_accuracy: 0.0000e+00 - val_loss: 25.1013 Epoch 43/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 412ms/step - categorical_accuracy: 0.1538 - loss: 3.7292 - val_categorical_accuracy: 0.0667 - val_loss: 19.0040 Epoch 44/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 417ms/step - categorical_accuracy: 0.3076 - loss: 3.2854 - val_categorical_accuracy: 0.0000e+00 - val_loss: 18.3936 Epoch 45/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.3293 - loss: 3.1678 - val_categorical_accuracy: 0.0000e+00 - val_loss: 15.6123 Epoch 46/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 407ms/step - categorical_accuracy: 0.2311 - loss: 3.1164 - val_categorical_accuracy: 0.0000e+00 - val_loss: 18.2801 Epoch 47/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 410ms/step - categorical_accuracy: 0.2754 - loss: 3.1783 - val_categorical_accuracy: 0.0000e+00 - val_loss: 16.1246 Epoch 48/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 415ms/step - categorical_accuracy: 0.1547 - loss: 3.7555 - val_categorical_accuracy: 0.0000e+00 - val_loss: 14.5329 Epoch 49/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 438ms/step - categorical_accuracy: 0.1999 - loss: 3.7122 - val_categorical_accuracy: 0.0000e+00 - val_loss: 11.7346 Epoch 50/50 2/2 ━━━━━━━━━━━━━━━━━━━━ 1s 416ms/step - categorical_accuracy: 0.1981 - loss: 3.4944 - val_categorical_accuracy: 0.0000e+00 - val_loss: 12.3622 WARNING:absl:You are saving your model as an HDF5 file via model.save() or keras.saving.save_model(model). This file format is considered legacy. We recommend using instead the native Keras format, e.g. model.save('my_model.keras') or keras.saving.save_model(model, 'my_model.keras'). 1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 80ms/step - categorical_accuracy: 0.0000e+00 - loss: 12.1257 Test Accuracy: 0.0%

[njszym]

This is likely due to your modification of the y() function. If you can provide me with a zipped version of your run folder, I'd be happy to help debug.

[mayisme]

Hi, Nathan, Attached please find the folder, thanks for your time! Very appreciated!

2024年3月10日 08:39，Nathan Szymanski @.***> 写道：

This is likely due to your modification of the y() function. If you can provide me with a zipped version of your run folder, I'd be happy to help debug.

— Reply to this email directly, view it on GitHub https://github.com/njszym/XRD-AutoAnalyzer/issues/5#issuecomment-1987025373, or unsubscribe https://github.com/notifications/unsubscribe-auth/AU3PTGLTGNM7TMBOB3CWZ63YXOTUDAVCNFSM6AAAAABEOACS2OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOBXGAZDKMZXGM. You are receiving this because you authored the thread.

[mayisme]

Novel-Space.zip

[mayisme]

autoXRD.zip

[mayisme]

Hi, the Novel-space.zip is my training folder, and the autoXRD.zip is the folder from side-packages, thanks for your time

[njszym]

I was able to reproduce your errors after upgrading tensorflow to the latest version. Turns out this modifies how the input should be shaped before passing it to the CNN. I've modified the relevant parts of the package to accommodate these changes. Please try downloading the latest version and re-training your model.

Thanks for bringing this issue to my attention! Let me know if you encounter any other problems.

[mayisme]

Great! it Works! Thank you very much!