Add files via upload

Author: AashiDutt

MNIST-Recognition/hand_digit_recognition_pycharm/MNIST-data/t10k-images-idx3-ubyte.gz

@@ -0,0 +1,53 @@
+import tensorflow as tf
+import numpy as np
+
+# getting dataset and training ,Testing data
+mnist_data = tf.contrib.learn.datasets.load_dataset('mnist')
+x_train = mnist_data.train.images
+y_train = np.asarray(mnist_data.train.labels, dtype=np.int32)
+x_eval = mnist_data.test.images
+y_eval = np.asarray(mnist_data.test.labels, dtype=np.int32)
+
+x_predict =x_eval[:1]
+
+
+# creating linear regression model y = Wx+b
+def model_fn(features, labels, mode):
+    x = tf.reshape(features['x'],[-1,784])
+    W = tf.get_variable(name='W', shape=[784, 10], dtype=tf.float32)
+    b = tf.get_variable(name='b', shape=[10], dtype=tf.float32)
+    y = tf.add(tf.matmul(x, W), b)
+
+    if mode == tf.estimator.ModeKeys.PREDICT:
+        return tf.estimator.EstimatorSpec(mode=mode, predictions=tf.nn.softmax(logits=y))
+
+    onehot_labels = tf.one_hot(indices=tf.cast(labels, tf.int32), depth=10)
+    loss = tf.losses.softmax_cross_entropy(onehot_labels=onehot_labels, logits=y)
+
+#training
+    if mode == tf.estimator.ModeKeys.TRAIN:
+        optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01)
+        train_step = optimizer.minimize(loss=loss, global_step=tf.train.get_global_step())
+
+        return tf.estimator.EstimatorSpec(mode=mode, loss=loss, train_op=train_step)
+
+#testing and evaluation
+    eval_metric_ops={'accuracy':tf.metrics.accuracy(labels=labels,predictions=tf.argmax(y, 1))}
+    return tf.estimator.EstimatorSpec(mode=mode, loss=loss, eval_metric_ops=eval_metric_ops)
+
+
+
+
+
+estimator= tf.estimator.Estimator(model_fn= model_fn)
+
+train_input_fn = tf.estimator.inputs.numpy_input_fn(x={'x':x_train}, y=y_train, batch_size=100, num_epochs=None, shuffle=True)
+
+eval_input_fn = tf.estimator.inputs.numpy_input_fn(x={'x':x_eval}, y=y_eval, num_epochs=1, shuffle=False)
+
+predict_input_fn = tf.estimator.inputs.numpy_input_fn(x={'x':x_predict},  num_epochs=1, shuffle=False)
+
+estimator.train(input_fn=train_input_fn, steps=20000)
+
+print(estimator.evaluate(input_fn=eval_input_fn))
+print(list(estimator.predict(input_fn=predict_input_fn)))