27 KiB
27 KiB
In [25]:
import tensorflow as tf
import numpy as np
mnist = tf.keras.datasets.mnist #28*28 images of hand-written digits 0-9
(x_train,y_train),(x_test,y_test)=mnist.load_data()
x_train=tf.keras.utils.normalize(x_train,axis=1)
x_test=tf.keras.utils.normalize(x_test,axis=1)
np.reshape(x_train,(len(x_train),1,784))
print("len(x_train):",len(x_train))
model=tf.keras.models.Sequential()
model.add(tf.keras.layers.Flatten())
l=[784,400,200,100,50,25,10]
l=[28,10]
print(l)
#Encoding the data from 128 parameters into 10 parameters
for e in l:
model.add(tf.keras.layers.Dense(e,activation=tf.nn.relu))
l.reverse()
l=l[1:]
print(l)
#Decoding the data from 10 parameters into 128 parameters
for e in l:
model.add(tf.keras.layers.Dense(e,activation=tf.nn.relu))
#model.add(tf.keras.layers.Dense(10,activation=tf.nn.softmax))
#model.add(tf.keras.layers.Flatten())
model.compile(optimizer="adam",
loss="sparse_categorical_crossentropy",
metrics=["accuracy"])
print(model.predict(x_train[0]))
model.fit(x_train,x_train,epochs=3)
len(x_train): 60000 [28, 10] [28] [[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0.01748016 0.02673778 0. 0. 0.00602611 0. 0. 0. 0. 0. 0.03430181 0.023855 0.00654872 0. 0. 0. 0.03825076 0. 0.07328451 0. 0.02354929 0.03025666 0. 0.04835717 0. 0.02556546 0.03385373 0.04659563] [0. 0. 0. 0. 0.06535421 0. 0. 0. 0.05645705 0. 0.13004425 0.01549844 0. 0.04896671 0. 0. 0.1036097 0.07355008 0.03742907 0. 0. 0. 0. 0. 0. 0.12024324 0.14770994 0. ] [0. 0. 0. 0. 0.0027197 0. 0.04302675 0. 0.11189289 0. 0.09072255 0. 0. 0.04395699 0.00933373 0.04975676 0.08487647 0.01378354 0.00572389 0. 0.02748866 0. 0. 0. 0. 0.12651348 0.16967528 0. ] [0. 0. 0. 0. 0. 0. 0.021663 0. 0.0808398 0. 0.07567543 0. 0. 0.03423792 0.00101215 0.0327794 0.08129841 0.00456038 0.02017812 0. 0.03672768 0. 0. 0. 0. 0.11122777 0.15737762 0. ] [0. 0. 0. 0. 0. 0. 0.00828841 0. 0.0364771 0. 0.02291169 0. 0. 0.01047879 0.00830038 0.008518 0.04492189 0. 0.01576747 0. 0.02269142 0. 0. 0. 0. 0.06107318 0.10030939 0.0093521 ] [0.03836422 0. 0.01341372 0. 0. 0.01680471 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.0099496 0. 0.02884961 0. 0.0285176 0.00626036 0. 0.02896094 0. 0.05333598 0.08157708 0.04866493] [0.05666146 0. 0.01858671 0. 0. 0.01886109 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.05889048 0. 0.04217981 0.01229247 0. 0.05158316 0. 0.0669031 0.10159926 0.07421101] [0.04424474 0. 0.00426109 0. 0. 0.00784304 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.06376532 0. 0.03909021 0.00079017 0. 0.04093649 0. 0.05953597 0.09062535 0.06668206] [0.00458736 0. 0. 0. 0. 0. 0.03779557 0. 0.03694625 0. 0. 0. 0.01871822 0. 0. 0.01730741 0. 0. 0.08863883 0. 0. 0. 0. 0. 0. 0.01105253 0.05268183 0.01396778] [0.01045329 0.00026588 0. 0. 0.00459926 0. 0.01541175 0. 0.02767995 0. 0.02607668 0.01160748 0.02473249 0. 0. 0. 0. 0.01378152 0.0738868 0. 0. 0. 0. 0.00621534 0. 0.01667231 0. 0. ] [0.02621182 0.0039462 0. 0. 0. 0. 0. 0. 0. 0. 0.00181005 0. 0. 0. 0. 0. 0. 0. 0.10302293 0. 0. 0. 0. 0.03307459 0. 0.04530415 0.04354399 0.04714802] [0.02551171 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.01098537 0. 0. 0. 0. 0. 0.0165434 0. 0.09723056 0. 0. 0. 0. 0.01548411 0. 0.07768489 0.0828326 0.04813599] [0.01116125 0. 0. 0. 0.02727178 0. 0. 0. 0. 0. 0.04559217 0. 0. 0.00793533 0. 0. 0.03210801 0.03480392 0.07089017 0. 0. 0. 0. 0. 0. 0.10177225 0.07893473 0.0061889 ] [0.02698384 0. 0. 0. 0. 0. 0.00097598 0. 0.02877774 0. 0.02568768 0. 0. 0. 0. 0. 0.00186984 0. 0.12629043 0. 0. 0. 0. 0. 0. 0.09394979 0.08351476 0.02893183] [0.01762724 0. 0. 0. 0. 0. 0.03111188 0. 0.05560888 0. 0.01469027 0. 0. 0. 0. 0.03756982 0. 0. 0.14676633 0. 0.0004858 0. 0. 0. 0. 0.11704206 0.12784247 0.03102896] [0.01644074 0. 0. 0. 0. 0. 0.00085254 0. 0.01004637 0. 0. 0. 0. 0. 0. 0.0299817 0. 0. 0.09267729 0. 0.01315921 0. 0. 0. 0. 0.07092497 0.10192703 0.04632806] [0. 0. 0. 0. 0. 0. 0.01564954 0. 0.04468219 0. 0.02532769 0. 0. 0. 0. 0.04668204 0.03436052 0. 0.06634824 0. 0.04312603 0. 0. 0. 0. 0.09648092 0.13965487 0.02023894] [0. 0. 0. 0. 0.0534489 0. 0.03701807 0. 0.11353236 0. 0.10966916 0. 0. 0.04975216 0.04810934 0.01796302 0.0962729 0.02871083 0. 0. 0.0064699 0. 0. 0. 0. 0.11479117 0.12461423 0. ] [0.0585979 0. 0. 0. 0.07174546 0. 0. 0. 0.02940173 0. 0.11493647 0. 0. 0.01027856 0.01944368 0. 0.1047096 0. 0.01168374 0. 0.01006486 0. 0. 0. 0. 0.14723466 0.08145888 0. ] [0.09485141 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.01003646 0. 0. 0. 0. 0. 0.06193022 0. 0.11450395 0. 0.05824487 0. 0. 0.08244261 0. 0.14939535 0.15929016 0.08277234] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ]]
Epoch 1/3
[0;31m---------------------------------------------------------------------------[0m
[0;31mInvalidArgumentError[0m Traceback (most recent call last)
[0;32m<ipython-input-25-1fb9cbeb0304>[0m in [0;36m<module>[0;34m[0m
[1;32m 39[0m [0mprint[0m[0;34m([0m[0mmodel[0m[0;34m.[0m[0mpredict[0m[0;34m([0m[0mx_train[0m[0;34m[[0m[0;36m0[0m[0;34m][0m[0;34m)[0m[0;34m)[0m[0;34m[0m[0;34m[0m[0m
[1;32m 40[0m [0;34m[0m[0m
[0;32m---> 41[0;31m [0mmodel[0m[0;34m.[0m[0mfit[0m[0;34m([0m[0mx_train[0m[0;34m,[0m[0mx_train[0m[0;34m,[0m[0mepochs[0m[0;34m=[0m[0;36m3[0m[0;34m)[0m[0;34m[0m[0;34m[0m[0m
[0m[1;32m 42[0m [0;34m[0m[0m
[1;32m 43[0m [0;34m[0m[0m
[0;32m/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py[0m in [0;36mfit[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, max_queue_size, workers, use_multiprocessing, **kwargs)[0m
[1;32m 878[0m [0minitial_epoch[0m[0;34m=[0m[0minitial_epoch[0m[0;34m,[0m[0;34m[0m[0;34m[0m[0m
[1;32m 879[0m [0msteps_per_epoch[0m[0;34m=[0m[0msteps_per_epoch[0m[0;34m,[0m[0;34m[0m[0;34m[0m[0m
[0;32m--> 880[0;31m validation_steps=validation_steps)
[0m[1;32m 881[0m [0;34m[0m[0m
[1;32m 882[0m def evaluate(self,
[0;32m/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training_arrays.py[0m in [0;36mmodel_iteration[0;34m(model, inputs, targets, sample_weights, batch_size, epochs, verbose, callbacks, val_inputs, val_targets, val_sample_weights, shuffle, initial_epoch, steps_per_epoch, validation_steps, mode, validation_in_fit, **kwargs)[0m
[1;32m 327[0m [0;34m[0m[0m
[1;32m 328[0m [0;31m# Get outputs.[0m[0;34m[0m[0;34m[0m[0;34m[0m[0m
[0;32m--> 329[0;31m [0mbatch_outs[0m [0;34m=[0m [0mf[0m[0;34m([0m[0mins_batch[0m[0;34m)[0m[0;34m[0m[0;34m[0m[0m
[0m[1;32m 330[0m [0;32mif[0m [0;32mnot[0m [0misinstance[0m[0;34m([0m[0mbatch_outs[0m[0;34m,[0m [0mlist[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0;34m[0m[0m
[1;32m 331[0m [0mbatch_outs[0m [0;34m=[0m [0;34m[[0m[0mbatch_outs[0m[0;34m][0m[0;34m[0m[0;34m[0m[0m
[0;32m/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/backend.py[0m in [0;36m__call__[0;34m(self, inputs)[0m
[1;32m 3074[0m [0;34m[0m[0m
[1;32m 3075[0m fetched = self._callable_fn(*array_vals,
[0;32m-> 3076[0;31m run_metadata=self.run_metadata)
[0m[1;32m 3077[0m [0mself[0m[0;34m.[0m[0m_call_fetch_callbacks[0m[0;34m([0m[0mfetched[0m[0;34m[[0m[0;34m-[0m[0mlen[0m[0;34m([0m[0mself[0m[0;34m.[0m[0m_fetches[0m[0;34m)[0m[0;34m:[0m[0;34m][0m[0;34m)[0m[0;34m[0m[0;34m[0m[0m
[1;32m 3078[0m return nest.pack_sequence_as(self._outputs_structure,
[0;32m/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py[0m in [0;36m__call__[0;34m(self, *args, **kwargs)[0m
[1;32m 1437[0m ret = tf_session.TF_SessionRunCallable(
[1;32m 1438[0m [0mself[0m[0;34m.[0m[0m_session[0m[0;34m.[0m[0m_session[0m[0;34m,[0m [0mself[0m[0;34m.[0m[0m_handle[0m[0;34m,[0m [0margs[0m[0;34m,[0m [0mstatus[0m[0;34m,[0m[0;34m[0m[0;34m[0m[0m
[0;32m-> 1439[0;31m run_metadata_ptr)
[0m[1;32m 1440[0m [0;32mif[0m [0mrun_metadata[0m[0;34m:[0m[0;34m[0m[0;34m[0m[0m
[1;32m 1441[0m [0mproto_data[0m [0;34m=[0m [0mtf_session[0m[0;34m.[0m[0mTF_GetBuffer[0m[0;34m([0m[0mrun_metadata_ptr[0m[0;34m)[0m[0;34m[0m[0;34m[0m[0m
[0;32m/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/errors_impl.py[0m in [0;36m__exit__[0;34m(self, type_arg, value_arg, traceback_arg)[0m
[1;32m 526[0m [0;32mNone[0m[0;34m,[0m [0;32mNone[0m[0;34m,[0m[0;34m[0m[0;34m[0m[0m
[1;32m 527[0m [0mcompat[0m[0;34m.[0m[0mas_text[0m[0;34m([0m[0mc_api[0m[0;34m.[0m[0mTF_Message[0m[0;34m([0m[0mself[0m[0;34m.[0m[0mstatus[0m[0;34m.[0m[0mstatus[0m[0;34m)[0m[0;34m)[0m[0;34m,[0m[0;34m[0m[0;34m[0m[0m
[0;32m--> 528[0;31m c_api.TF_GetCode(self.status.status))
[0m[1;32m 529[0m [0;31m# Delete the underlying status object from memory otherwise it stays alive[0m[0;34m[0m[0;34m[0m[0;34m[0m[0m
[1;32m 530[0m [0;31m# as there is a reference to status from this from the traceback due to[0m[0;34m[0m[0;34m[0m[0;34m[0m[0m
[0;31mInvalidArgumentError[0m: Can not squeeze dim[2], expected a dimension of 1, got 28
[[{{node metrics_16/acc/Squeeze}}]]In [26]:
for y in model.predict(x_train[0]):
print(y)[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0.01748016 0.02673778 0. 0. 0.00602611 0. 0. 0. 0. 0. 0.03430181 0.023855 0.00654872 0. 0. 0. 0.03825076 0. 0.07328451 0. 0.02354929 0.03025666 0. 0.04835717 0. 0.02556546 0.03385373 0.04659563] [0. 0. 0. 0. 0.06535421 0. 0. 0. 0.05645705 0. 0.13004425 0.01549844 0. 0.04896671 0. 0. 0.1036097 0.07355008 0.03742907 0. 0. 0. 0. 0. 0. 0.12024324 0.14770994 0. ] [0. 0. 0. 0. 0.0027197 0. 0.04302675 0. 0.11189289 0. 0.09072255 0. 0. 0.04395699 0.00933373 0.04975676 0.08487647 0.01378354 0.00572389 0. 0.02748866 0. 0. 0. 0. 0.12651348 0.16967528 0. ] [0. 0. 0. 0. 0. 0. 0.021663 0. 0.0808398 0. 0.07567543 0. 0. 0.03423792 0.00101215 0.0327794 0.08129841 0.00456038 0.02017812 0. 0.03672768 0. 0. 0. 0. 0.11122777 0.15737762 0. ] [0. 0. 0. 0. 0. 0. 0.00828841 0. 0.0364771 0. 0.02291169 0. 0. 0.01047879 0.00830038 0.008518 0.04492189 0. 0.01576747 0. 0.02269142 0. 0. 0. 0. 0.06107318 0.10030939 0.0093521 ] [0.03836422 0. 0.01341372 0. 0. 0.01680471 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.0099496 0. 0.02884961 0. 0.0285176 0.00626036 0. 0.02896094 0. 0.05333598 0.08157708 0.04866493] [0.05666146 0. 0.01858671 0. 0. 0.01886109 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.05889048 0. 0.04217981 0.01229247 0. 0.05158316 0. 0.0669031 0.10159926 0.07421101] [0.04424474 0. 0.00426109 0. 0. 0.00784304 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.06376532 0. 0.03909021 0.00079017 0. 0.04093649 0. 0.05953597 0.09062535 0.06668206] [0.00458736 0. 0. 0. 0. 0. 0.03779557 0. 0.03694625 0. 0. 0. 0.01871822 0. 0. 0.01730741 0. 0. 0.08863883 0. 0. 0. 0. 0. 0. 0.01105253 0.05268183 0.01396778] [0.01045329 0.00026588 0. 0. 0.00459926 0. 0.01541175 0. 0.02767995 0. 0.02607668 0.01160748 0.02473249 0. 0. 0. 0. 0.01378152 0.0738868 0. 0. 0. 0. 0.00621534 0. 0.01667231 0. 0. ] [0.02621182 0.0039462 0. 0. 0. 0. 0. 0. 0. 0. 0.00181005 0. 0. 0. 0. 0. 0. 0. 0.10302293 0. 0. 0. 0. 0.03307459 0. 0.04530415 0.04354399 0.04714802] [0.02551171 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.01098537 0. 0. 0. 0. 0. 0.0165434 0. 0.09723056 0. 0. 0. 0. 0.01548411 0. 0.07768489 0.0828326 0.04813599] [0.01116125 0. 0. 0. 0.02727178 0. 0. 0. 0. 0. 0.04559217 0. 0. 0.00793533 0. 0. 0.03210801 0.03480392 0.07089017 0. 0. 0. 0. 0. 0. 0.10177225 0.07893473 0.0061889 ] [0.02698384 0. 0. 0. 0. 0. 0.00097598 0. 0.02877774 0. 0.02568768 0. 0. 0. 0. 0. 0.00186984 0. 0.12629043 0. 0. 0. 0. 0. 0. 0.09394979 0.08351476 0.02893183] [0.01762724 0. 0. 0. 0. 0. 0.03111188 0. 0.05560888 0. 0.01469027 0. 0. 0. 0. 0.03756982 0. 0. 0.14676633 0. 0.0004858 0. 0. 0. 0. 0.11704206 0.12784247 0.03102896] [0.01644074 0. 0. 0. 0. 0. 0.00085254 0. 0.01004637 0. 0. 0. 0. 0. 0. 0.0299817 0. 0. 0.09267729 0. 0.01315921 0. 0. 0. 0. 0.07092497 0.10192703 0.04632806] [0. 0. 0. 0. 0. 0. 0.01564954 0. 0.04468219 0. 0.02532769 0. 0. 0. 0. 0.04668204 0.03436052 0. 0.06634824 0. 0.04312603 0. 0. 0. 0. 0.09648092 0.13965487 0.02023894] [0. 0. 0. 0. 0.0534489 0. 0.03701807 0. 0.11353236 0. 0.10966916 0. 0. 0.04975216 0.04810934 0.01796302 0.0962729 0.02871083 0. 0. 0.0064699 0. 0. 0. 0. 0.11479117 0.12461423 0. ] [0.0585979 0. 0. 0. 0.07174546 0. 0. 0. 0.02940173 0. 0.11493647 0. 0. 0.01027856 0.01944368 0. 0.1047096 0. 0.01168374 0. 0.01006486 0. 0. 0. 0. 0.14723466 0.08145888 0. ] [0.09485141 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.01003646 0. 0. 0. 0. 0. 0.06193022 0. 0.11450395 0. 0.05824487 0. 0. 0.08244261 0. 0.14939535 0.15929016 0.08277234] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
In [ ]: