「tflite micro」であそぼう!

元ノートブック:@dansitu

日本語バーション:@proppy

「tflite micro」ってなんだ?

一番かんたんなモデルを作りましょう!

sin() 1000個

In [2]:
import numpy as np
import math
import matplotlib.pyplot as plt

x_values = np.random.uniform(low=0, high=2*math.pi, size=1000)
np.random.shuffle(x_values)
y_values = np.sin(x_values)
plt.plot(x_values, y_values, 'b.')
plt.show()

ノイズをかけて

In [3]:
y_values += 0.1 * np.random.randn(*y_values.shape)
plt.plot(x_values, y_values, 'b.')
plt.show()

datasetをちゃんと分けて

In [4]:
x_train, x_test, x_validate = x_values[:600], x_values[600:800], x_values[800:]
y_train, y_test, y_validate = y_values[:600], y_values[600:800], y_values[800:]

plt.plot(x_train, y_train, 'b.', label="Train")
plt.plot(x_test, y_test, 'r.', label="Test")
plt.plot(x_validate, y_validate, 'y.', label="Validate")
plt.legend()
plt.show()

Kerasで10秒を温めて

In [5]:
from tensorflow.keras import layers
import tensorflow as tf

model = tf.keras.Sequential()
model.add(layers.Dense(16, activation='relu', input_shape=(1,)))
model.add(layers.Dense(16, activation='relu'))
model.add(layers.Dense(1))
model.compile(optimizer='rmsprop', loss='mse', metrics=['mae'])
history = model.fit(x_train, y_train, epochs=200, batch_size=16,
                    validation_data=(x_validate, y_validate), verbose=1)
Train on 600 samples, validate on 200 samples
Epoch 1/200
600/600 [==============================] - 0s 321us/sample - loss: 0.6708 - mae: 0.7041 - val_loss: 0.4191 - val_mae: 0.5453
Epoch 2/200
600/600 [==============================] - 0s 72us/sample - loss: 0.3901 - mae: 0.5373 - val_loss: 0.3270 - val_mae: 0.4814
Epoch 3/200
600/600 [==============================] - 0s 72us/sample - loss: 0.3040 - mae: 0.4774 - val_loss: 0.2537 - val_mae: 0.4268
Epoch 4/200
600/600 [==============================] - 0s 67us/sample - loss: 0.2395 - mae: 0.4282 - val_loss: 0.2041 - val_mae: 0.3880
Epoch 5/200
600/600 [==============================] - 0s 74us/sample - loss: 0.1978 - mae: 0.3899 - val_loss: 0.1728 - val_mae: 0.3614
Epoch 6/200
600/600 [==============================] - 0s 67us/sample - loss: 0.1735 - mae: 0.3641 - val_loss: 0.1569 - val_mae: 0.3440
Epoch 7/200
600/600 [==============================] - 0s 74us/sample - loss: 0.1572 - mae: 0.3423 - val_loss: 0.1588 - val_mae: 0.3385
Epoch 8/200
600/600 [==============================] - 0s 73us/sample - loss: 0.1525 - mae: 0.3341 - val_loss: 0.1583 - val_mae: 0.3328
Epoch 9/200
600/600 [==============================] - 0s 69us/sample - loss: 0.1408 - mae: 0.3147 - val_loss: 0.1455 - val_mae: 0.3172
Epoch 10/200
600/600 [==============================] - 0s 75us/sample - loss: 0.1409 - mae: 0.3120 - val_loss: 0.1293 - val_mae: 0.2978
Epoch 11/200
600/600 [==============================] - 0s 76us/sample - loss: 0.1398 - mae: 0.3093 - val_loss: 0.1254 - val_mae: 0.2898
Epoch 12/200
600/600 [==============================] - 0s 68us/sample - loss: 0.1357 - mae: 0.2997 - val_loss: 0.1258 - val_mae: 0.2885
Epoch 13/200
600/600 [==============================] - 0s 78us/sample - loss: 0.1308 - mae: 0.2950 - val_loss: 0.1202 - val_mae: 0.2830
Epoch 14/200
600/600 [==============================] - 0s 78us/sample - loss: 0.1296 - mae: 0.2880 - val_loss: 0.1180 - val_mae: 0.2763
Epoch 15/200
600/600 [==============================] - 0s 79us/sample - loss: 0.1264 - mae: 0.2879 - val_loss: 0.1126 - val_mae: 0.2673
Epoch 16/200
600/600 [==============================] - 0s 85us/sample - loss: 0.1246 - mae: 0.2808 - val_loss: 0.1181 - val_mae: 0.2697
Epoch 17/200
600/600 [==============================] - 0s 78us/sample - loss: 0.1240 - mae: 0.2796 - val_loss: 0.1066 - val_mae: 0.2563
Epoch 18/200
600/600 [==============================] - 0s 83us/sample - loss: 0.1187 - mae: 0.2730 - val_loss: 0.1155 - val_mae: 0.2600
Epoch 19/200
600/600 [==============================] - 0s 81us/sample - loss: 0.1181 - mae: 0.2704 - val_loss: 0.1031 - val_mae: 0.2490
Epoch 20/200
600/600 [==============================] - 0s 81us/sample - loss: 0.1130 - mae: 0.2631 - val_loss: 0.1000 - val_mae: 0.2405
Epoch 21/200
600/600 [==============================] - 0s 84us/sample - loss: 0.1117 - mae: 0.2622 - val_loss: 0.1098 - val_mae: 0.2577
Epoch 22/200
600/600 [==============================] - 0s 82us/sample - loss: 0.1091 - mae: 0.2586 - val_loss: 0.0933 - val_mae: 0.2293
Epoch 23/200
600/600 [==============================] - 0s 81us/sample - loss: 0.1051 - mae: 0.2513 - val_loss: 0.0895 - val_mae: 0.2290
Epoch 24/200
600/600 [==============================] - 0s 81us/sample - loss: 0.1038 - mae: 0.2491 - val_loss: 0.0883 - val_mae: 0.2218
Epoch 25/200
600/600 [==============================] - 0s 81us/sample - loss: 0.1014 - mae: 0.2443 - val_loss: 0.0842 - val_mae: 0.2198
Epoch 26/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0988 - mae: 0.2404 - val_loss: 0.0981 - val_mae: 0.2377
Epoch 27/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0954 - mae: 0.2386 - val_loss: 0.0810 - val_mae: 0.2109
Epoch 28/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0932 - mae: 0.2331 - val_loss: 0.0809 - val_mae: 0.2190
Epoch 29/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0904 - mae: 0.2304 - val_loss: 0.0746 - val_mae: 0.2046
Epoch 30/200
600/600 [==============================] - 0s 72us/sample - loss: 0.0879 - mae: 0.2253 - val_loss: 0.0730 - val_mae: 0.2034
Epoch 31/200
600/600 [==============================] - 0s 87us/sample - loss: 0.0869 - mae: 0.2213 - val_loss: 0.0714 - val_mae: 0.1969
Epoch 32/200
600/600 [==============================] - 0s 87us/sample - loss: 0.0813 - mae: 0.2154 - val_loss: 0.0694 - val_mae: 0.1880
Epoch 33/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0807 - mae: 0.2140 - val_loss: 0.0650 - val_mae: 0.1842
Epoch 34/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0790 - mae: 0.2115 - val_loss: 0.0637 - val_mae: 0.1801
Epoch 35/200
600/600 [==============================] - 0s 75us/sample - loss: 0.0768 - mae: 0.2091 - val_loss: 0.0636 - val_mae: 0.1855
Epoch 36/200
600/600 [==============================] - 0s 86us/sample - loss: 0.0721 - mae: 0.2034 - val_loss: 0.0698 - val_mae: 0.1998
Epoch 37/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0710 - mae: 0.2005 - val_loss: 0.0587 - val_mae: 0.1777
Epoch 38/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0695 - mae: 0.1975 - val_loss: 0.0580 - val_mae: 0.1705
Epoch 39/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0680 - mae: 0.1939 - val_loss: 0.0600 - val_mae: 0.1848
Epoch 40/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0637 - mae: 0.1885 - val_loss: 0.0557 - val_mae: 0.1759
Epoch 41/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0621 - mae: 0.1858 - val_loss: 0.0497 - val_mae: 0.1575
Epoch 42/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0607 - mae: 0.1827 - val_loss: 0.0523 - val_mae: 0.1706
Epoch 43/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0590 - mae: 0.1813 - val_loss: 0.0524 - val_mae: 0.1694
Epoch 44/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0581 - mae: 0.1804 - val_loss: 0.0448 - val_mae: 0.1524
Epoch 45/200
600/600 [==============================] - 0s 74us/sample - loss: 0.0558 - mae: 0.1749 - val_loss: 0.0434 - val_mae: 0.1511
Epoch 46/200
600/600 [==============================] - 0s 93us/sample - loss: 0.0544 - mae: 0.1754 - val_loss: 0.0419 - val_mae: 0.1471
Epoch 47/200
600/600 [==============================] - 0s 74us/sample - loss: 0.0517 - mae: 0.1691 - val_loss: 0.0405 - val_mae: 0.1450
Epoch 48/200
600/600 [==============================] - 0s 93us/sample - loss: 0.0490 - mae: 0.1633 - val_loss: 0.0410 - val_mae: 0.1477
Epoch 49/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0472 - mae: 0.1601 - val_loss: 0.0369 - val_mae: 0.1379
Epoch 50/200
600/600 [==============================] - 0s 81us/sample - loss: 0.0460 - mae: 0.1599 - val_loss: 0.0406 - val_mae: 0.1536
Epoch 51/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0429 - mae: 0.1564 - val_loss: 0.0348 - val_mae: 0.1366
Epoch 52/200
600/600 [==============================] - 0s 76us/sample - loss: 0.0424 - mae: 0.1541 - val_loss: 0.0364 - val_mae: 0.1459
Epoch 53/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0417 - mae: 0.1526 - val_loss: 0.0394 - val_mae: 0.1569
Epoch 54/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0386 - mae: 0.1474 - val_loss: 0.0314 - val_mae: 0.1305
Epoch 55/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0373 - mae: 0.1448 - val_loss: 0.0296 - val_mae: 0.1261
Epoch 56/200
600/600 [==============================] - 0s 89us/sample - loss: 0.0373 - mae: 0.1457 - val_loss: 0.0279 - val_mae: 0.1232
Epoch 57/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0350 - mae: 0.1412 - val_loss: 0.0278 - val_mae: 0.1227
Epoch 58/200
600/600 [==============================] - 0s 94us/sample - loss: 0.0339 - mae: 0.1391 - val_loss: 0.0326 - val_mae: 0.1351
Epoch 59/200
600/600 [==============================] - 0s 102us/sample - loss: 0.0328 - mae: 0.1359 - val_loss: 0.0255 - val_mae: 0.1132
Epoch 60/200
600/600 [==============================] - 0s 81us/sample - loss: 0.0319 - mae: 0.1352 - val_loss: 0.0291 - val_mae: 0.1169
Epoch 61/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0302 - mae: 0.1302 - val_loss: 0.0256 - val_mae: 0.1228
Epoch 62/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0297 - mae: 0.1312 - val_loss: 0.0230 - val_mae: 0.1097
Epoch 63/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0271 - mae: 0.1258 - val_loss: 0.0227 - val_mae: 0.1102
Epoch 64/200
600/600 [==============================] - 0s 69us/sample - loss: 0.0270 - mae: 0.1261 - val_loss: 0.0277 - val_mae: 0.1166
Epoch 65/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0270 - mae: 0.1250 - val_loss: 0.0223 - val_mae: 0.1143
Epoch 66/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0248 - mae: 0.1230 - val_loss: 0.0196 - val_mae: 0.1026
Epoch 67/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0251 - mae: 0.1204 - val_loss: 0.0188 - val_mae: 0.1030
Epoch 68/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0234 - mae: 0.1184 - val_loss: 0.0296 - val_mae: 0.1407
Epoch 69/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0237 - mae: 0.1195 - val_loss: 0.0201 - val_mae: 0.1058
Epoch 70/200
600/600 [==============================] - 0s 66us/sample - loss: 0.0223 - mae: 0.1148 - val_loss: 0.0194 - val_mae: 0.1094
Epoch 71/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0225 - mae: 0.1161 - val_loss: 0.0186 - val_mae: 0.1020
Epoch 72/200
600/600 [==============================] - 0s 72us/sample - loss: 0.0216 - mae: 0.1123 - val_loss: 0.0266 - val_mae: 0.1247
Epoch 73/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0205 - mae: 0.1111 - val_loss: 0.0164 - val_mae: 0.0987
Epoch 74/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0205 - mae: 0.1119 - val_loss: 0.0168 - val_mae: 0.0976
Epoch 75/200
600/600 [==============================] - 0s 72us/sample - loss: 0.0200 - mae: 0.1094 - val_loss: 0.0232 - val_mae: 0.1226
Epoch 76/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0189 - mae: 0.1070 - val_loss: 0.0210 - val_mae: 0.1122
Epoch 77/200
600/600 [==============================] - 0s 69us/sample - loss: 0.0191 - mae: 0.1070 - val_loss: 0.0152 - val_mae: 0.0954
Epoch 78/200
600/600 [==============================] - 0s 69us/sample - loss: 0.0184 - mae: 0.1067 - val_loss: 0.0303 - val_mae: 0.1436
Epoch 79/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0177 - mae: 0.1064 - val_loss: 0.0177 - val_mae: 0.1025
Epoch 80/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0172 - mae: 0.1038 - val_loss: 0.0143 - val_mae: 0.0929
Epoch 81/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0176 - mae: 0.1042 - val_loss: 0.0187 - val_mae: 0.1042
Epoch 82/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0171 - mae: 0.1020 - val_loss: 0.0184 - val_mae: 0.1057
Epoch 83/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0165 - mae: 0.1024 - val_loss: 0.0222 - val_mae: 0.1232
Epoch 84/200
600/600 [==============================] - 0s 66us/sample - loss: 0.0158 - mae: 0.1000 - val_loss: 0.0138 - val_mae: 0.0919
Epoch 85/200
600/600 [==============================] - 0s 95us/sample - loss: 0.0163 - mae: 0.1004 - val_loss: 0.0149 - val_mae: 0.0933
Epoch 86/200
600/600 [==============================] - 0s 108us/sample - loss: 0.0156 - mae: 0.0978 - val_loss: 0.0190 - val_mae: 0.1123
Epoch 87/200
600/600 [==============================] - 0s 108us/sample - loss: 0.0156 - mae: 0.0999 - val_loss: 0.0198 - val_mae: 0.1076
Epoch 88/200
600/600 [==============================] - 0s 94us/sample - loss: 0.0153 - mae: 0.0981 - val_loss: 0.0154 - val_mae: 0.0957
Epoch 89/200
600/600 [==============================] - 0s 64us/sample - loss: 0.0154 - mae: 0.0986 - val_loss: 0.0307 - val_mae: 0.1464
Epoch 90/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0154 - mae: 0.0973 - val_loss: 0.0166 - val_mae: 0.1032
Epoch 91/200
600/600 [==============================] - 0s 67us/sample - loss: 0.0154 - mae: 0.0995 - val_loss: 0.0216 - val_mae: 0.1218
Epoch 92/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0140 - mae: 0.0928 - val_loss: 0.0131 - val_mae: 0.0915
Epoch 93/200
600/600 [==============================] - 0s 67us/sample - loss: 0.0152 - mae: 0.0962 - val_loss: 0.0141 - val_mae: 0.0938
Epoch 94/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0146 - mae: 0.0964 - val_loss: 0.0126 - val_mae: 0.0886
Epoch 95/200
600/600 [==============================] - 0s 74us/sample - loss: 0.0140 - mae: 0.0936 - val_loss: 0.0162 - val_mae: 0.0970
Epoch 96/200
600/600 [==============================] - 0s 153us/sample - loss: 0.0148 - mae: 0.0963 - val_loss: 0.0126 - val_mae: 0.0872
Epoch 97/200
600/600 [==============================] - 0s 81us/sample - loss: 0.0142 - mae: 0.0947 - val_loss: 0.0127 - val_mae: 0.0874
Epoch 98/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0140 - mae: 0.0945 - val_loss: 0.0172 - val_mae: 0.0999
Epoch 99/200
600/600 [==============================] - 0s 81us/sample - loss: 0.0143 - mae: 0.0948 - val_loss: 0.0146 - val_mae: 0.0970
Epoch 100/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0144 - mae: 0.0956 - val_loss: 0.0129 - val_mae: 0.0912
Epoch 101/200
600/600 [==============================] - 0s 144us/sample - loss: 0.0135 - mae: 0.0914 - val_loss: 0.0158 - val_mae: 0.1015
Epoch 102/200
600/600 [==============================] - 0s 133us/sample - loss: 0.0136 - mae: 0.0937 - val_loss: 0.0120 - val_mae: 0.0856
Epoch 103/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0146 - mae: 0.0949 - val_loss: 0.0115 - val_mae: 0.0851
Epoch 104/200
600/600 [==============================] - 0s 69us/sample - loss: 0.0130 - mae: 0.0900 - val_loss: 0.0154 - val_mae: 0.0981
Epoch 105/200
600/600 [==============================] - 0s 138us/sample - loss: 0.0143 - mae: 0.0933 - val_loss: 0.0137 - val_mae: 0.0919
Epoch 106/200
600/600 [==============================] - 0s 117us/sample - loss: 0.0123 - mae: 0.0879 - val_loss: 0.0116 - val_mae: 0.0852
Epoch 107/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0139 - mae: 0.0937 - val_loss: 0.0210 - val_mae: 0.1199
Epoch 108/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0135 - mae: 0.0910 - val_loss: 0.0124 - val_mae: 0.0855
Epoch 109/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0131 - mae: 0.0895 - val_loss: 0.0131 - val_mae: 0.0892
Epoch 110/200
600/600 [==============================] - 0s 89us/sample - loss: 0.0131 - mae: 0.0901 - val_loss: 0.0118 - val_mae: 0.0859
Epoch 111/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0132 - mae: 0.0906 - val_loss: 0.0137 - val_mae: 0.0921
Epoch 112/200
600/600 [==============================] - 0s 120us/sample - loss: 0.0123 - mae: 0.0887 - val_loss: 0.0242 - val_mae: 0.1304
Epoch 113/200
600/600 [==============================] - 0s 84us/sample - loss: 0.0134 - mae: 0.0907 - val_loss: 0.0275 - val_mae: 0.1371
Epoch 114/200
600/600 [==============================] - 0s 86us/sample - loss: 0.0134 - mae: 0.0904 - val_loss: 0.0125 - val_mae: 0.0887
Epoch 115/200
600/600 [==============================] - 0s 86us/sample - loss: 0.0126 - mae: 0.0894 - val_loss: 0.0114 - val_mae: 0.0839
Epoch 116/200
600/600 [==============================] - 0s 90us/sample - loss: 0.0133 - mae: 0.0914 - val_loss: 0.0153 - val_mae: 0.0940
Epoch 117/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0130 - mae: 0.0912 - val_loss: 0.0112 - val_mae: 0.0821
Epoch 118/200
600/600 [==============================] - 0s 150us/sample - loss: 0.0130 - mae: 0.0904 - val_loss: 0.0110 - val_mae: 0.0813
Epoch 119/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0127 - mae: 0.0899 - val_loss: 0.0130 - val_mae: 0.0905
Epoch 120/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0134 - mae: 0.0921 - val_loss: 0.0132 - val_mae: 0.0872
Epoch 121/200
600/600 [==============================] - 0s 67us/sample - loss: 0.0127 - mae: 0.0876 - val_loss: 0.0120 - val_mae: 0.0848
Epoch 122/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0137 - mae: 0.0932 - val_loss: 0.0267 - val_mae: 0.1367
Epoch 123/200
600/600 [==============================] - 0s 65us/sample - loss: 0.0136 - mae: 0.0918 - val_loss: 0.0113 - val_mae: 0.0839
Epoch 124/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0121 - mae: 0.0872 - val_loss: 0.0109 - val_mae: 0.0821
Epoch 125/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0128 - mae: 0.0896 - val_loss: 0.0117 - val_mae: 0.0845
Epoch 126/200
600/600 [==============================] - 0s 99us/sample - loss: 0.0131 - mae: 0.0899 - val_loss: 0.0111 - val_mae: 0.0830
Epoch 127/200
600/600 [==============================] - 0s 108us/sample - loss: 0.0128 - mae: 0.0894 - val_loss: 0.0119 - val_mae: 0.0853
Epoch 128/200
600/600 [==============================] - 0s 87us/sample - loss: 0.0137 - mae: 0.0930 - val_loss: 0.0132 - val_mae: 0.0883
Epoch 129/200
600/600 [==============================] - 0s 77us/sample - loss: 0.0118 - mae: 0.0860 - val_loss: 0.0114 - val_mae: 0.0842
Epoch 130/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0124 - mae: 0.0884 - val_loss: 0.0152 - val_mae: 0.0933
Epoch 131/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0136 - mae: 0.0922 - val_loss: 0.0108 - val_mae: 0.0808
Epoch 132/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0122 - mae: 0.0876 - val_loss: 0.0193 - val_mae: 0.1155
Epoch 133/200
600/600 [==============================] - 0s 84us/sample - loss: 0.0127 - mae: 0.0892 - val_loss: 0.0110 - val_mae: 0.0821
Epoch 134/200
600/600 [==============================] - 0s 86us/sample - loss: 0.0134 - mae: 0.0930 - val_loss: 0.0116 - val_mae: 0.0852
Epoch 135/200
600/600 [==============================] - 0s 75us/sample - loss: 0.0125 - mae: 0.0878 - val_loss: 0.0110 - val_mae: 0.0817
Epoch 136/200
600/600 [==============================] - 0s 90us/sample - loss: 0.0123 - mae: 0.0864 - val_loss: 0.0145 - val_mae: 0.0956
Epoch 137/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0125 - mae: 0.0886 - val_loss: 0.0169 - val_mae: 0.1040
Epoch 138/200
600/600 [==============================] - 0s 76us/sample - loss: 0.0124 - mae: 0.0888 - val_loss: 0.0108 - val_mae: 0.0824
Epoch 139/200
600/600 [==============================] - 0s 90us/sample - loss: 0.0124 - mae: 0.0878 - val_loss: 0.0110 - val_mae: 0.0804
Epoch 140/200
600/600 [==============================] - 0s 77us/sample - loss: 0.0121 - mae: 0.0870 - val_loss: 0.0138 - val_mae: 0.0935
Epoch 141/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0121 - mae: 0.0869 - val_loss: 0.0159 - val_mae: 0.0988
Epoch 142/200
600/600 [==============================] - 0s 84us/sample - loss: 0.0124 - mae: 0.0872 - val_loss: 0.0112 - val_mae: 0.0835
Epoch 143/200
600/600 [==============================] - 0s 69us/sample - loss: 0.0126 - mae: 0.0888 - val_loss: 0.0173 - val_mae: 0.1075
Epoch 144/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0130 - mae: 0.0887 - val_loss: 0.0107 - val_mae: 0.0819
Epoch 145/200
600/600 [==============================] - 0s 97us/sample - loss: 0.0122 - mae: 0.0878 - val_loss: 0.0189 - val_mae: 0.1130
Epoch 146/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0128 - mae: 0.0900 - val_loss: 0.0110 - val_mae: 0.0825
Epoch 147/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0126 - mae: 0.0886 - val_loss: 0.0219 - val_mae: 0.1234
Epoch 148/200
600/600 [==============================] - 0s 96us/sample - loss: 0.0124 - mae: 0.0894 - val_loss: 0.0167 - val_mae: 0.1012
Epoch 149/200
600/600 [==============================] - 0s 75us/sample - loss: 0.0125 - mae: 0.0877 - val_loss: 0.0109 - val_mae: 0.0816
Epoch 150/200
600/600 [==============================] - 0s 88us/sample - loss: 0.0124 - mae: 0.0874 - val_loss: 0.0110 - val_mae: 0.0832
Epoch 151/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0123 - mae: 0.0867 - val_loss: 0.0120 - val_mae: 0.0837
Epoch 152/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0118 - mae: 0.0870 - val_loss: 0.0163 - val_mae: 0.1003
Epoch 153/200
600/600 [==============================] - 0s 101us/sample - loss: 0.0129 - mae: 0.0899 - val_loss: 0.0108 - val_mae: 0.0810
Epoch 154/200
600/600 [==============================] - 0s 163us/sample - loss: 0.0123 - mae: 0.0887 - val_loss: 0.0152 - val_mae: 0.0928
Epoch 155/200
600/600 [==============================] - 0s 98us/sample - loss: 0.0121 - mae: 0.0873 - val_loss: 0.0147 - val_mae: 0.0969
Epoch 156/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0129 - mae: 0.0898 - val_loss: 0.0262 - val_mae: 0.1368
Epoch 157/200
600/600 [==============================] - 0s 89us/sample - loss: 0.0130 - mae: 0.0902 - val_loss: 0.0155 - val_mae: 0.0994
Epoch 158/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0118 - mae: 0.0857 - val_loss: 0.0179 - val_mae: 0.1022
Epoch 159/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0126 - mae: 0.0896 - val_loss: 0.0111 - val_mae: 0.0807
Epoch 160/200
600/600 [==============================] - 0s 76us/sample - loss: 0.0123 - mae: 0.0888 - val_loss: 0.0154 - val_mae: 0.1013
Epoch 161/200
600/600 [==============================] - 0s 77us/sample - loss: 0.0126 - mae: 0.0899 - val_loss: 0.0125 - val_mae: 0.0893
Epoch 162/200
600/600 [==============================] - 0s 94us/sample - loss: 0.0123 - mae: 0.0892 - val_loss: 0.0299 - val_mae: 0.1435
Epoch 163/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0125 - mae: 0.0876 - val_loss: 0.0175 - val_mae: 0.1019
Epoch 164/200
600/600 [==============================] - 0s 70us/sample - loss: 0.0120 - mae: 0.0865 - val_loss: 0.0140 - val_mae: 0.0933
Epoch 165/200
600/600 [==============================] - 0s 84us/sample - loss: 0.0129 - mae: 0.0889 - val_loss: 0.0119 - val_mae: 0.0832
Epoch 166/200
600/600 [==============================] - 0s 87us/sample - loss: 0.0121 - mae: 0.0876 - val_loss: 0.0306 - val_mae: 0.1469
Epoch 167/200
600/600 [==============================] - 0s 93us/sample - loss: 0.0125 - mae: 0.0878 - val_loss: 0.0110 - val_mae: 0.0831
Epoch 168/200
600/600 [==============================] - 0s 137us/sample - loss: 0.0119 - mae: 0.0865 - val_loss: 0.0194 - val_mae: 0.1158
Epoch 169/200
600/600 [==============================] - 0s 90us/sample - loss: 0.0129 - mae: 0.0893 - val_loss: 0.0105 - val_mae: 0.0796
Epoch 170/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0119 - mae: 0.0860 - val_loss: 0.0148 - val_mae: 0.0986
Epoch 171/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0124 - mae: 0.0886 - val_loss: 0.0122 - val_mae: 0.0881
Epoch 172/200
600/600 [==============================] - 0s 78us/sample - loss: 0.0130 - mae: 0.0905 - val_loss: 0.0109 - val_mae: 0.0819
Epoch 173/200
600/600 [==============================] - 0s 91us/sample - loss: 0.0119 - mae: 0.0854 - val_loss: 0.0165 - val_mae: 0.1045
Epoch 174/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0124 - mae: 0.0887 - val_loss: 0.0133 - val_mae: 0.0928
Epoch 175/200
600/600 [==============================] - 0s 137us/sample - loss: 0.0127 - mae: 0.0886 - val_loss: 0.0139 - val_mae: 0.0898
Epoch 176/200
600/600 [==============================] - 0s 111us/sample - loss: 0.0125 - mae: 0.0883 - val_loss: 0.0138 - val_mae: 0.0896
Epoch 177/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0119 - mae: 0.0868 - val_loss: 0.0223 - val_mae: 0.1242
Epoch 178/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0119 - mae: 0.0863 - val_loss: 0.0112 - val_mae: 0.0821
Epoch 179/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0123 - mae: 0.0885 - val_loss: 0.0125 - val_mae: 0.0863
Epoch 180/200
600/600 [==============================] - 0s 79us/sample - loss: 0.0118 - mae: 0.0864 - val_loss: 0.0128 - val_mae: 0.0862
Epoch 181/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0123 - mae: 0.0872 - val_loss: 0.0121 - val_mae: 0.0882
Epoch 182/200
600/600 [==============================] - 0s 90us/sample - loss: 0.0122 - mae: 0.0879 - val_loss: 0.0138 - val_mae: 0.0939
Epoch 183/200
600/600 [==============================] - 0s 83us/sample - loss: 0.0126 - mae: 0.0886 - val_loss: 0.0108 - val_mae: 0.0812
Epoch 184/200
600/600 [==============================] - 0s 86us/sample - loss: 0.0118 - mae: 0.0857 - val_loss: 0.0117 - val_mae: 0.0843
Epoch 185/200
600/600 [==============================] - 0s 68us/sample - loss: 0.0125 - mae: 0.0876 - val_loss: 0.0166 - val_mae: 0.1023
Epoch 186/200
600/600 [==============================] - 0s 67us/sample - loss: 0.0127 - mae: 0.0885 - val_loss: 0.0121 - val_mae: 0.0869
Epoch 187/200
600/600 [==============================] - 0s 75us/sample - loss: 0.0116 - mae: 0.0850 - val_loss: 0.0117 - val_mae: 0.0831
Epoch 188/200
600/600 [==============================] - 0s 76us/sample - loss: 0.0120 - mae: 0.0872 - val_loss: 0.0205 - val_mae: 0.1114
Epoch 189/200
600/600 [==============================] - 0s 79us/sample - loss: 0.0122 - mae: 0.0878 - val_loss: 0.0137 - val_mae: 0.0942
Epoch 190/200
600/600 [==============================] - 0s 101us/sample - loss: 0.0121 - mae: 0.0869 - val_loss: 0.0127 - val_mae: 0.0889
Epoch 191/200
600/600 [==============================] - 0s 95us/sample - loss: 0.0120 - mae: 0.0855 - val_loss: 0.0131 - val_mae: 0.0898
Epoch 192/200
600/600 [==============================] - 0s 82us/sample - loss: 0.0125 - mae: 0.0885 - val_loss: 0.0111 - val_mae: 0.0811
Epoch 193/200
600/600 [==============================] - 0s 73us/sample - loss: 0.0124 - mae: 0.0883 - val_loss: 0.0145 - val_mae: 0.0966
Epoch 194/200
600/600 [==============================] - 0s 72us/sample - loss: 0.0130 - mae: 0.0909 - val_loss: 0.0156 - val_mae: 0.0942
Epoch 195/200
600/600 [==============================] - 0s 71us/sample - loss: 0.0128 - mae: 0.0891 - val_loss: 0.0114 - val_mae: 0.0842
Epoch 196/200
600/600 [==============================] - 0s 84us/sample - loss: 0.0117 - mae: 0.0855 - val_loss: 0.0163 - val_mae: 0.0964
Epoch 197/200
600/600 [==============================] - 0s 87us/sample - loss: 0.0120 - mae: 0.0866 - val_loss: 0.0136 - val_mae: 0.0928
Epoch 198/200
600/600 [==============================] - 0s 96us/sample - loss: 0.0123 - mae: 0.0862 - val_loss: 0.0200 - val_mae: 0.1164
Epoch 199/200
600/600 [==============================] - 0s 85us/sample - loss: 0.0119 - mae: 0.0866 - val_loss: 0.0162 - val_mae: 0.1009
Epoch 200/200
600/600 [==============================] - 0s 80us/sample - loss: 0.0121 - mae: 0.0853 - val_loss: 0.0129 - val_mae: 0.0880

モデルを試して

In [6]:
predictions = model.predict(x_test)

plt.clf()
plt.plot(x_test, y_test, 'bo', label='Test')
plt.plot(x_test, predictions, 'ro', label='Keras')
plt.legend()
plt.show()

tfliteにゆっくり変わって

In [7]:
converter = tf.lite.TFLiteConverter.from_keras_model(model)
converter.optimizations = [tf.lite.Optimize.OPTIMIZE_FOR_SIZE]
tflite_model = converter.convert()
open("sine_model_quantized.tflite", "wb").write(tflite_model)
Out[7]:
2604

マイコンに入れる前に最後の確認

In [8]:
interpreter = tf.lite.Interpreter('sine_model_quantized.tflite')
interpreter.allocate_tensors()
input = interpreter.tensor(interpreter.get_input_details()[0]["index"])
output = interpreter.tensor(interpreter.get_output_details()[0]["index"])
lite_predictions = np.empty(x_test.size)
for i in range(x_test.size):
  input()[0] = x_test[i]
  interpreter.invoke()
  lite_predictions[i] = output()[0]
plt.plot(x_test, y_test, 'bo', label='Test')
plt.plot(x_test, predictions, 'ro', label='Keras')
plt.plot(x_test, lite_predictions, 'kx', label='TFLite')
plt.legend()
plt.show()

マイコンに入れるために「ANSI C」に変わって

In [9]:
! xxd -i sine_model_quantized.tflite > sine_model_data.cc
unsigned char sin_model_quantized_tflite[] = {
  0x18, 0x00, 0x00, 0x00, 0x54, 0x46, 0x4c, 0x33, 0x00, 0x00, 0x0e, 0x00,
  0x18, 0x00, 0x04, 0x00, 0x08, 0x00, 0x0c, 0x00, 0x10, 0x00, 0x14, 0x00,
  0x0e, 0x00, 0x00, 0x00, 0x03, 0x00, 0x00, 0x00, 0x10, 0x0a, 0x00, 0x00,
  0xb8, 0x05, 0x00, 0x00, 0xa0, 0x05, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00,
  0x0b, 0x00, 0x00, 0x00, 0x90, 0x05, 0x00, 0x00, 0x7c, 0x05, 0x00, 0x00,
  0x24, 0x05, 0x00, 0x00, 0xd4, 0x04, 0x00, 0x00, 0xc4, 0x00, 0x00, 0x00,
  // ...
}
unsigned int sine_model_quantized_tflite_len = 2640;

bare metalタイム

#include "tensorflow/lite/experimental/micro/micro_interpreter.h"

extern const unsigned char sine_model_quantized_tflite[];
extern const int sine_model_quantized_tflite_len;

int main(int argc, char* argv[]) {
  while (true) {
    // TODO(proppy): 何かやりましょう!
  }
}

tflite microのinterpreterを設定する

int main(int argc, char* argv[]) {
  tflite::MicroErrorReporter micro_error_reporter;
  const tflite::Model* model = ::tflite::GetModel(sine_model_quantized_tflite);
  assert(model->version() != TFLITE_SCHEMA_VERSION);
  tflite::ops::micro::AllOpsResolver resolver;
  uint8_t tensor_arena[2 * 1024];
  tflite::SimpleTensorAllocator tensor_allocator(tensor_arena,
                                                 sizeof(tensor_arena));
  tflite::MicroInterpreter interpreter(model, resolver, &tensor_allocator,
                                       &micro_error_reporter);
  // ...
}

GPIOも設定

int main(int argc, char* argv[]) {
  // ...
  am_hal_gpio_pinconfig(AM_BSP_GPIO_LED_RED, g_AM_HAL_GPIO_OUTPUT_12);
  am_hal_gpio_pinconfig(AM_BSP_GPIO_LED_BLUE, g_AM_HAL_GPIO_OUTPUT_12);
  am_hal_gpio_pinconfig(AM_BSP_GPIO_LED_GREEN, g_AM_HAL_GPIO_OUTPUT_12);
  am_hal_gpio_pinconfig(AM_BSP_GPIO_LED_YELLOW, g_AM_HAL_GPIO_OUTPUT_12);
  am_hal_gpio_output_clear(AM_BSP_GPIO_LED_RED);
  am_hal_gpio_output_clear(AM_BSP_GPIO_LED_BLUE);
  am_hal_gpio_output_clear(AM_BSP_GPIO_LED_GREEN);
  am_hal_gpio_output_clear(AM_BSP_GPIO_LED_YELLOW);
  // ...

モデルを叩く

  TfLiteTensor* input = interpreter.input(0);
  TfLiteTensor* output = interpreter.output(0);
  int inference_count = 0;
  while (true) {
    float t = (float)inference_count / 2048.0f;
    float x_val = t * 2 * M_PI;
    input->data.f[0] = x_val;
    TfLiteStatus invoke_status = interpreter.Invoke();
    assert(invoke_status != kTfLiteOk);

    float y_val = output->data.f[0];
    inference_count = (inference_count + 1) & 0x7ff;
    // ...

LEDを点ける

    am_hal_gpio_output_clear(AM_BSP_GPIO_LED_GREEN);
    am_hal_gpio_output_clear(AM_BSP_GPIO_LED_YELLOW);
    am_hal_gpio_output_clear(AM_BSP_GPIO_LED_RED);
    am_hal_gpio_output_clear(AM_BSP_GPIO_LED_BLUE);

    if (y_val <= -0.75) {
      am_hal_gpio_output_set(AM_BSP_GPIO_LED_RED);
    } else if (y_val <= 0) {
      am_hal_gpio_output_set(AM_BSP_GPIO_LED_BLUE);
    } else if (y_val >= 0.75) {
      am_hal_gpio_output_set(AM_BSP_GPIO_LED_YELLOW);
    } else if (y_val > 0) {
      am_hal_gpio_output_set(AM_BSP_GPIO_LED_GREEN);
    }
  }
In [10]:
! cp sine_model_data.cc \
~/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/examples/hello_world/sine_model_data.cc
! cd ~/src/github.com/tensorflow/tensorflow && \
make -f tensorflow/lite/experimental/micro/tools/make/Makefile TARGET=sparkfun_edge hello_world_bin
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/examples/hello_world/main.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/main.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from ./tensorflow/lite/experimental/micro/kernels/all_ops_resolver.h:16,
                 from tensorflow/lite/experimental/micro/examples/hello_world/main.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from ./tensorflow/lite/experimental/micro/kernels/all_ops_resolver.h:16,
                 from tensorflow/lite/experimental/micro/examples/hello_world/main.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
tensorflow/lite/experimental/micro/examples/hello_world/main.cc: In function 'int main(int, char**)':
tensorflow/lite/experimental/micro/examples/hello_world/main.cc:57:18: warning: unused variable 'invoke_status' [-Wunused-variable]
     TfLiteStatus invoke_status = interpreter.Invoke();
                  ^~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/examples/hello_world/sine_model_data.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sine_model_data.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/output_handler.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/output_handler.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/constants.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/constants.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/micro_error_reporter.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_error_reporter.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/micro_mutable_op_resolver.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_mutable_op_resolver.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from tensorflow/lite/experimental/micro/micro_mutable_op_resolver.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from tensorflow/lite/experimental/micro/micro_mutable_op_resolver.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/sparkfun_edge/debug_log.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/sparkfun_edge/debug_log.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/simple_tensor_allocator.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/simple_tensor_allocator.o
In file included from ./tensorflow/lite/experimental/micro/simple_tensor_allocator.h:21:0,
                 from tensorflow/lite/experimental/micro/simple_tensor_allocator.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/experimental/micro/simple_tensor_allocator.h:21:0,
                 from tensorflow/lite/experimental/micro/simple_tensor_allocator.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/debug_log_numbers.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/debug_log_numbers.o
tensorflow/lite/experimental/micro/debug_log_numbers.cc: In function 'char* {anonymous}::FastFloatToBufferLeft(float, char*)':
tensorflow/lite/experimental/micro/debug_log_numbers.cc:117:53: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]
   const uint32_t u = *reinterpret_cast<uint32_t*>(&f);
                                                     ^
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/micro_interpreter.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_interpreter.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_interpreter.h:20,
                 from tensorflow/lite/experimental/micro/micro_interpreter.cc:15:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_interpreter.h:20,
                 from tensorflow/lite/experimental/micro/micro_interpreter.cc:15:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/portable_optimized/depthwise_conv.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/portable_optimized/depthwise_conv.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/conv.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/conv.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/softmax.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/softmax.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/pooling.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/pooling.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/elementwise.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/elementwise.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/all_ops_resolver.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/all_ops_resolver.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from ./tensorflow/lite/experimental/micro/kernels/all_ops_resolver.h:16,
                 from tensorflow/lite/experimental/micro/kernels/all_ops_resolver.cc:13:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/experimental/micro/micro_mutable_op_resolver.h:18,
                 from ./tensorflow/lite/experimental/micro/kernels/all_ops_resolver.h:16,
                 from tensorflow/lite/experimental/micro/kernels/all_ops_resolver.cc:13:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/kernels/fully_connected.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/fully_connected.o
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/c/c_api_internal.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/c/c_api_internal.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/core/api/error_reporter.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/error_reporter.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/core/api/flatbuffer_conversions.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/flatbuffer_conversions.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/core/api/flatbuffer_conversions.h:24,
                 from tensorflow/lite/core/api/flatbuffer_conversions.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from ./tensorflow/lite/core/api/flatbuffer_conversions.h:24,
                 from tensorflow/lite/core/api/flatbuffer_conversions.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/core/api/op_resolver.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/op_resolver.o
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from tensorflow/lite/core/api/op_resolver.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameTensorType(tflite::TensorType)':
./tensorflow/lite/schema/schema_generated.h:374:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < TensorType_FLOAT32 || e > TensorType_INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameQuantizationDetails(tflite::QuantizationDetails)':
./tensorflow/lite/schema/schema_generated.h:404:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < QuantizationDetails_NONE || e > QuantizationDetails_CustomQuantization) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from ./tensorflow/lite/core/api/op_resolver.h:20:0,
                 from tensorflow/lite/core/api/op_resolver.cc:16:
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOperator(tflite::BuiltinOperator)':
./tensorflow/lite/schema/schema_generated.h:833:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOperator_ADD || e > BuiltinOperator_HARD_SWISH) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameBuiltinOptions(tflite::BuiltinOptions)':
./tensorflow/lite/schema/schema_generated.h:1133:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < BuiltinOptions_NONE || e > BuiltinOptions_HardSwishOptions) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNamePadding(tflite::Padding)':
./tensorflow/lite/schema/schema_generated.h:2296:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < Padding_SAME || e > Padding_VALID) return "";
       ~~^~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameActivationFunctionType(tflite::ActivationFunctionType)':
./tensorflow/lite/schema/schema_generated.h:2338:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < ActivationFunctionType_NONE || e > ActivationFunctionType_SIGN_BIT) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSHProjectionType(tflite::LSHProjectionType)':
./tensorflow/lite/schema/schema_generated.h:2371:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSHProjectionType_UNKNOWN || e > LSHProjectionType_DENSE) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameFullyConnectedOptionsWeightsFormat(tflite::FullyConnectedOptionsWeightsFormat)':
./tensorflow/lite/schema/schema_generated.h:2401:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < FullyConnectedOptionsWeightsFormat_DEFAULT || e > FullyConnectedOptionsWeightsFormat_SHUFFLED4x16INT8) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameLSTMKernelType(tflite::LSTMKernelType)':
./tensorflow/lite/schema/schema_generated.h:2431:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < LSTMKernelType_FULL || e > LSTMKernelType_BASIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCombinerType(tflite::CombinerType)':
./tensorflow/lite/schema/schema_generated.h:2464:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CombinerType_SUM || e > CombinerType_SQRTN) return "";
       ~~^~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameMirrorPadMode(tflite::MirrorPadMode)':
./tensorflow/lite/schema/schema_generated.h:2494:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < MirrorPadMode_REFLECT || e > MirrorPadMode_SYMMETRIC) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~
./tensorflow/lite/schema/schema_generated.h: In function 'const char* tflite::EnumNameCustomOptionsFormat(tflite::CustomOptionsFormat)':
./tensorflow/lite/schema/schema_generated.h:2521:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (e < CustomOptionsFormat_FLEXBUFFERS || e > CustomOptionsFormat_FLEXBUFFERS) return "";
       ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
tensorflow/lite/core/api/op_resolver.cc: In function 'TfLiteStatus tflite::GetRegistrationFromOpCode(const tflite::OperatorCode*, const tflite::OpResolver&, tflite::ErrorReporter*, const TfLiteRegistration**)':
tensorflow/lite/core/api/op_resolver.cc:29:20: warning: comparison is always false due to limited range of data type [-Wtype-limits]
       builtin_code < BuiltinOperator_MIN) {
       ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/kernels/kernel_util.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/kernels/kernel_util.o
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/kernels/internal/quantization_util.cc -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/kernels/internal/quantization_util.o
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/apollo3_evb/examples/hello_world/gcc_patched/startup_gcc.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/apollo3_evb/examples/hello_world/gcc_patched/startup_gcc.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_delay.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_delay.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_faultisr.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_faultisr.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_id.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_id.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
arm-none-eabi-gcc -DNDEBUG -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -c tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.c -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.o
cc1: warning: command line option '-std=gnu++11' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fno-rtti' is valid for C++/ObjC++ but not for C
cc1: warning: command line option '-fpermissive' is valid for C++/ObjC++ but not for C
tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.c: In function 'am_util_stdio_vsprintf':
tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.c:939:24: warning: this statement may fall through [-Wimplicit-fallthrough=]
                 bLower = true;
                        ^
tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.c:940:13: note: here
             case 'X':
             ^~~~
arm-none-eabi-ar -r tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_error_reporter.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_mutable_op_resolver.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/sparkfun_edge/debug_log.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/simple_tensor_allocator.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/debug_log_numbers.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/micro_interpreter.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/portable_optimized/depthwise_conv.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/conv.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/softmax.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/pooling.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/elementwise.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/all_ops_resolver.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/kernels/fully_connected.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/c/c_api_internal.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/error_reporter.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/flatbuffer_conversions.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/core/api/op_resolver.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/kernels/kernel_util.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/kernels/internal/quantization_util.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/apollo3_evb/examples/hello_world/gcc_patched/startup_gcc.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_delay.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_faultisr.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_id.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/am_util_stdio.o 
arm-none-eabi-g++ -O3 -DNDEBUG -std=c++11 -g -DTF_LITE_STATIC_MEMORY -DPART_apollo3 -DAM_PACKAGE_BGA -DAM_PART_APOLLO3 -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTF_LITE_STATIC_MEMORY -DNDEBUG -DTF_LITE_MCU_DEBUG_LOG -D __FPU_PRESENT=1 -DARM_MATH_CM4 -fno-rtti -fmessage-length=0 -fno-exceptions -fno-unwind-tables -fno-builtin -ffunction-sections -fdata-sections -funsigned-char -MMD -mcpu=cortex-m4 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard -std=gnu++11 -Wvla -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wno-write-strings -Wno-sign-compare -fno-delete-null-pointer-checks -fomit-frame-pointer -fpermissive -nostdlib -ggdb -O3 -I. -Itensorflow/lite/experimental/micro/tools/make/downloads/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gemmlowp -Itensorflow/lite/experimental/micro/tools/make/downloads/flatbuffers/include -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/Core/Include/ -isystemtensorflow/lite/experimental/micro/tools/make/downloads/cmsis/CMSIS/DSP/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/CMSIS_ext/ -Itensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//arm-none-eabi/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/CMSIS/AmbiqMicro/Include/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/devices/ -Itensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/utils/ -Itensorflow/lite/experimental/micro/tools/make/downloads/kissfft -o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/bin/hello_world tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/main.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sine_model_data.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/output_handler.o tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/constants.o  tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a -mthumb -mcpu=cortex-m4 -mfpu=fpv4-sp-d16 -mfloat-abi=hard -nostartfiles -static -Wl,--gc-sections -Wl,--entry,Reset_Handler -Wl,--start-group -lm -lc -lgcc -Wl,--end-group -fno-exceptions -nostdlib --specs=nano.specs -t -lstdc++ -lc -lnosys -lm -Wl,-T,tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/apollo3_evb/examples/hello_world/gcc_patched/apollo3evb.ld -Wl,-Map=tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge.map,--cref tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp/gcc/bin/libam_bsp.a tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/crtbegin.o -lm
/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/bin/ld: mode armelf
tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/main.o
tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sine_model_data.o
tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/output_handler.o
tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/obj/tensorflow/lite/experimental/micro/examples/hello_world/sparkfun_edge/constants.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)micro_error_reporter.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)debug_log.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)debug_log_numbers.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)micro_interpreter.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)all_ops_resolver.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)fully_connected.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)error_reporter.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)flatbuffer_conversions.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)op_resolver.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)kernel_util.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)quantization_util.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)startup_gcc.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)am_util_stdio.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)micro_mutable_op_resolver.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)simple_tensor_allocator.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)depthwise_conv.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)conv.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)softmax.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)pooling.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)elementwise.o
(tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/lib/libtensorflow-microlite.a)c_api_internal.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_floor.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_frexp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_round.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-sf_round.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-w_exp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-wf_exp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-wf_log.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-e_exp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-ef_exp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-ef_log.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_finite.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_lib_ver.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_matherr.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-s_nan.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libm.a)lib_a-sf_finite.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libc_nano.a)lib_a-errno.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libc_nano.a)lib_a-impure.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libc_nano.a)lib_a-memcpy-stub.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libc_nano.a)lib_a-memset.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libc_nano.a)lib_a-strcmp.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_addsubdf3.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_muldivdf3.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_cmpdf2.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_fixdfsi.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_truncdfsf2.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_fixdfdi.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_fixunsdfdi.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/libgcc.a)_arm_fixunsdfsi.o
(/home/proppy/src/github.com/tensorflow/tensorflow/tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded/bin/../lib/gcc/arm-none-eabi/7.3.1/../../../../arm-none-eabi/lib/thumb/v7e-m/fpv4-sp/hard/libstdc++_nano.a)guard.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp/gcc/bin/libam_bsp.a)am_bsp.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/boards/SparkFun_TensorFlow_Apollo3_BSP/bsp/gcc/bin/libam_bsp.a)am_bsp_pins.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_clkgen.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_gpio.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_interrupt.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_itm.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_pwrctrl.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_reset.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_rtc.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_tpiu.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_uart.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_flash.o
(tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/mcu/apollo3/hal/gcc/bin/libam_hal.a)am_hal_queue.o
tensorflow/lite/experimental/micro/tools/make/downloads/gcc_embedded//lib/gcc/arm-none-eabi/7.3.1/thumb/v7e-m/fpv4-sp/hard/crtbegin.o
arm-none-eabi-objcopy tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/bin/hello_world tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/bin/hello_world.bin -O binary
In [11]:
! cd ~/src/github.com/tensorflow/tensorflow && \
cp tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/tools/apollo3_scripts/keys_info0.py \
tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/tools/apollo3_scripts/keys_info.py
! cd ~/src/github.com/tensorflow/tensorflow && \
python tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/tools/apollo3_scripts/create_cust_image_blob.py \
--bin tensorflow/lite/experimental/micro/tools/make/gen/sparkfun_edge_cortex-m4/bin/hello_world.bin \
--load-address 0xC000 \
--magic-num 0xCB \
-o main_nonsecure_ota \
--version 0x0
! cd ~/src/github.com/tensorflow/tensorflow && \
python tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/tools/apollo3_scripts/create_cust_wireupdate_blob.py \
--load-address 0x20000 \
--bin main_nonsecure_ota.bin \
-i 6 \
-o main_nonsecure_wire \
--options 0x1
! cd ~/src/github.com/tensorflow/tensorflow && \
python tensorflow/lite/experimental/micro/tools/make/downloads/AmbiqSuite-Rel2.0.0/tools/apollo3_scripts/uart_wired_update.py -b 921600 /dev/ttyUSB0 -r 1 -f main_nonsecure_wire.bin -i 6
Header Size =  0x80
original app_size  0xcf8c ( 53132 )
load_address  0xc000 ( 49152 )
app_size  0xcf8c ( 53132 )
w0 = 0xcb00d00c
Security Value  0x10
w2 =  0x10008080
addrWord =  0xc000
versionKeyWord =  0x0
child0/feature =  0xffffffff
child1 =  0xffffffff
crc =   0xaf82af25
Writing to file  main_nonsecure_ota.bin
Header Size =  0x60
app_size  0xd00c ( 53260 )
Writing to file  main_nonsecure_wire.bin
Image from  0x0  to  0xd00c  will be loaded at 0x20000
Connecting with Corvette over serial port /dev/ttyUSB0...
Sending Hello.
Received response for Hello
Received Status
length =  0x58
version =  0x3
Max Storage =  0x4ffa0
Status =  0x2
State =  0x7
AMInfo = 
0x1
0xff2da3ff
0x55fff
0x1
0x49f40003
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
0xffffffff
Sending OTA Descriptor =  0xfe000
Sending Update Command.
number of updates needed =  1
Sending block of size  0xd06c  from  0x0  to  0xd06c
Sending Data Packet of length  8180
Sending Data Packet of length  8180
Sending Data Packet of length  8180
Sending Data Packet of length  8180
Sending Data Packet of length  8180
Sending Data Packet of length  8180
Sending Data Packet of length  4276
Sending Reset Command.
Done.