From babdb6073f06906b08c5a2cc4305fee8af8b0edf Mon Sep 17 00:00:00 2001
From: tomrink <rink@ssec.wisc.edu>
Date: Mon, 17 Apr 2023 13:26:24 -0500
Subject: [PATCH] initial commit...
---
.../deeplearning/cloud_fraction_fcn_abi.py | 1066 +++++++++++++++++
1 file changed, 1066 insertions(+)
create mode 100644 modules/deeplearning/cloud_fraction_fcn_abi.py
diff --git a/modules/deeplearning/cloud_fraction_fcn_abi.py b/modules/deeplearning/cloud_fraction_fcn_abi.py
new file mode 100644
index 00000000..28f76db1
--- /dev/null
+++ b/modules/deeplearning/cloud_fraction_fcn_abi.py
@@ -0,0 +1,1066 @@
+import glob
+import tensorflow as tf
+
+from util.plot_cm import confusion_matrix_values
+from util.setup import logdir, modeldir, now, ancillary_path
+from util.util import EarlyStop, normalize, denormalize, get_grid_values_all
+import os, datetime
+import numpy as np
+import pickle
+import h5py
+import xarray as xr
+import gc
+
+AUTOTUNE = tf.data.AUTOTUNE
+
+LOG_DEVICE_PLACEMENT = False
+
+PROC_BATCH_SIZE = 4
+PROC_BATCH_BUFFER_SIZE = 5000
+
+NumClasses = 5
+if NumClasses == 2:
+ NumLogits = 1
+else:
+ NumLogits = NumClasses
+
+BATCH_SIZE = 128
+NUM_EPOCHS = 80
+
+TRACK_MOVING_AVERAGE = False
+EARLY_STOP = True
+
+NOISE_TRAINING = False
+NOISE_STDDEV = 0.01
+DO_AUGMENT = False
+
+DO_SMOOTH = False
+SIGMA = 1.0
+DO_ZERO_OUT = False
+
+# setup scaling parameters dictionary
+mean_std_dct = {}
+mean_std_file = ancillary_path+'mean_std_lo_hi_l2.pkl'
+f = open(mean_std_file, 'rb')
+mean_std_dct_l2 = pickle.load(f)
+f.close()
+
+mean_std_file = ancillary_path+'mean_std_lo_hi_l1b.pkl'
+f = open(mean_std_file, 'rb')
+mean_std_dct_l1b = pickle.load(f)
+f.close()
+
+mean_std_dct.update(mean_std_dct_l1b)
+mean_std_dct.update(mean_std_dct_l2)
+
+IMG_DEPTH = 1
+
+label_param = 'cloud_probability'
+
+params = ['temp_11_0um_nom', 'refl_0_65um_nom', label_param]
+params_i = ['temp_11_0um_nom', 'refl_0_65um_nom', label_param]
+data_params_half = ['temp_11_0um_nom']
+data_params_full = ['refl_0_65um_nom']
+
+label_idx_i = params_i.index(label_param)
+label_idx = params.index(label_param)
+
+print('data_params_half: ', data_params_half)
+print('data_params_full: ', data_params_full)
+print('label_param: ', label_param)
+
+KERNEL_SIZE = 3 # target size: (128, 128)
+N_X = N_Y = 1
+X_LEN = Y_LEN = 128
+
+if KERNEL_SIZE == 3:
+ slc_x = slice(1, int((N_X*X_LEN)/2) + 3)
+ slc_y = slice(1, int((N_Y*Y_LEN)/2) + 3)
+ x_128 = slice(4, N_X*X_LEN + 4)
+ y_128 = slice(4, N_Y*Y_LEN + 4)
+# elif KERNEL_SIZE == 5: These no longer apply here
+# slc_x = slice(3, 135)
+# slc_y = slice(3, 135)
+# slc_x_2 = slice(2, 137, 2)
+# slc_y_2 = slice(2, 137, 2)
+# x_128 = slice(5, 133)
+# y_128 = slice(5, 133)
+# t = np.arange(1, 67, 0.5)
+# s = np.arange(1, 67, 0.5)
+# x_2 = np.arange(68)
+# y_2 = np.arange(68)
+# ----------------------------------------
+
+
+def build_residual_conv2d_block(conv, num_filters, block_name, activation=tf.nn.relu, padding='SAME',
+ kernel_initializer='he_uniform', scale=None, kernel_size=3,
+ do_drop_out=True, drop_rate=0.5, do_batch_norm=True):
+
+ with tf.name_scope(block_name):
+ skip = tf.keras.layers.Conv2D(num_filters, kernel_size=kernel_size, padding=padding, kernel_initializer=kernel_initializer, activation=activation)(conv)
+ skip = tf.keras.layers.Conv2D(num_filters, kernel_size=kernel_size, padding=padding, activation=None)(skip)
+
+ if scale is not None:
+ skip = tf.keras.layers.Lambda(lambda x: x * scale)(skip)
+
+ if do_drop_out:
+ skip = tf.keras.layers.Dropout(drop_rate)(skip)
+
+ if do_batch_norm:
+ skip = tf.keras.layers.BatchNormalization()(skip)
+
+ conv = conv + skip
+ print(block_name+':', conv.shape)
+
+ return conv
+
+
+def upsample_mean(grd):
+ bsize, ylen, xlen = grd.shape
+ up = np.zeros((bsize, ylen*2, xlen*2))
+
+ up[:, ::2, ::2] = grd[:, ::2, ::2]
+ up[:, 1::2, ::2] = grd[:, ::2, ::2]
+ up[:, ::2, 1::2] = grd[:, ::2, ::2]
+ up[:, 1::2, 1::2] = grd[:, ::2, ::2]
+
+ return up
+
+
+def get_grid_cell_mean(grd_k):
+ grd_k = np.where(np.isnan(grd_k), 0, grd_k)
+ a = grd_k[:, 0::2, 0::2]
+ b = grd_k[:, 1::2, 0::2]
+ c = grd_k[:, 0::2, 1::2]
+ d = grd_k[:, 1::2, 1::2]
+ mean = np.nanmean([a, b, c, d], axis=0)
+
+ return mean
+
+
+def get_min_max_std(grd_k):
+ grd_k = np.where(np.isnan(grd_k), 0, grd_k)
+ a = grd_k[:, 0::2, 0::2]
+ b = grd_k[:, 1::2, 0::2]
+ c = grd_k[:, 0::2, 1::2]
+ d = grd_k[:, 1::2, 1::2]
+
+ lo = np.nanmin([a, b, c, d], axis=0)
+ hi = np.nanmax([a, b, c, d], axis=0)
+ std = np.nanstd([a, b, c, d], axis=0)
+ avg = np.nanmean([a, b, c, d], axis=0)
+
+ return lo, hi, std, avg
+
+
+def get_label_data(grd_k):
+ grd_k = np.where(np.isnan(grd_k), 0, grd_k)
+ grd_k = np.where(grd_k < 0.50, 0, 1)
+
+ a = grd_k[:, 0::2, 0::2]
+ b = grd_k[:, 1::2, 0::2]
+ c = grd_k[:, 0::2, 1::2]
+ d = grd_k[:, 1::2, 1::2]
+ s = a + b + c + d
+
+ cat_0 = (s == 0)
+ cat_1 = np.logical_and(s > 0, s < 4)
+ cat_2 = (s == 4)
+ s[cat_0] = 0
+ s[cat_1] = 1
+ s[cat_2] = 2
+
+ return s
+
+
+def get_label_data_5cat(grd_k):
+ grd_k = np.where(np.isnan(grd_k), 0, grd_k)
+ # grd_u = np.where(np.logical_and(grd_k > 0.45, grd_k < 0.55), 1, 0)
+ grd_k = np.where(grd_k < 0.5, 0, 1)
+
+ a = grd_k[:, 0::2, 0::2]
+ b = grd_k[:, 1::2, 0::2]
+ c = grd_k[:, 0::2, 1::2]
+ d = grd_k[:, 1::2, 1::2]
+ s = a + b + c + d
+
+ cat_0 = (s == 0)
+ cat_1 = (s == 1)
+ cat_2 = (s == 2)
+ cat_3 = (s == 3)
+ cat_4 = (s == 4)
+
+ s[cat_0] = 0
+ s[cat_1] = 1
+ s[cat_2] = 2
+ s[cat_3] = 3
+ s[cat_4] = 4
+
+ # a = grd_u[:, 0::2, 0::2]
+ # b = grd_u[:, 1::2, 0::2]
+ # c = grd_u[:, 0::2, 1::2]
+ # d = grd_u[:, 1::2, 1::2]
+ # s_u = a + b + c + d
+ # cat_u = (s_u == 4)
+ # s[cat_u] = 5
+
+ return s
+
+
+class SRCNN:
+
+ def __init__(self):
+
+ self.train_data = None
+ self.train_label = None
+ self.test_data = None
+ self.test_label = None
+ self.test_data_denorm = None
+
+ self.train_dataset = None
+ self.inner_train_dataset = None
+ self.test_dataset = None
+ self.eval_dataset = None
+ self.X_img = None
+ self.X_prof = None
+ self.X_u = None
+ self.X_v = None
+ self.X_sfc = None
+ self.inputs = []
+ self.y = None
+ self.handle = None
+ self.inner_handle = None
+ self.in_mem_batch = None
+
+ self.h5f_l1b_trn = None
+ self.h5f_l1b_tst = None
+ self.h5f_l2_trn = None
+ self.h5f_l2_tst = None
+
+ self.logits = None
+
+ self.predict_data = None
+ self.predict_dataset = None
+ self.mean_list = None
+ self.std_list = None
+
+ self.training_op = None
+ self.correct = None
+ self.accuracy = None
+ self.loss = None
+ self.pred_class = None
+ self.variable_averages = None
+
+ self.global_step = None
+
+ self.writer_train = None
+ self.writer_valid = None
+ self.writer_train_valid_loss = None
+
+ self.OUT_OF_RANGE = False
+
+ self.model = None
+ self.optimizer = None
+ self.ema = None
+ self.train_loss = None
+ self.train_accuracy = None
+ self.test_loss = None
+ self.test_accuracy = None
+ self.test_auc = None
+ self.test_recall = None
+ self.test_precision = None
+ self.test_confusion_matrix = None
+ self.test_true_pos = None
+ self.test_true_neg = None
+ self.test_false_pos = None
+ self.test_false_neg = None
+
+ self.test_labels = []
+ self.test_preds = []
+ self.test_probs = None
+
+ self.learningRateSchedule = None
+ self.num_data_samples = None
+ self.initial_learning_rate = None
+
+ self.data_dct = None
+ self.train_data_files = None
+ self.train_label_files = None
+ self.test_data_files = None
+ self.test_label_files = None
+
+ # self.n_chans = len(data_params_half) + len(data_params_full) + 1
+ self.n_chans = 5
+
+ self.X_img = tf.keras.Input(shape=(None, None, self.n_chans))
+
+ self.inputs.append(self.X_img)
+
+ tf.debugging.set_log_device_placement(LOG_DEVICE_PLACEMENT)
+
+ def get_in_mem_data_batch(self, idxs, is_training):
+ if is_training:
+ data_files = self.train_data_files
+ label_files = self.train_label_files
+ else:
+ data_files = self.test_data_files
+ label_files = self.test_label_files
+
+ data_s = []
+ label_s = []
+ for k in idxs:
+ f = data_files[k]
+ nda = np.load(f)
+ data_s.append(nda)
+
+ f = label_files[k]
+ nda = np.load(f)
+ label_s.append(nda)
+ input_data = np.concatenate(data_s)
+ input_label = np.concatenate(label_s)
+
+ data_norm = []
+ for param in data_params_half:
+ # If next 2 uncommented, take out get_grid_cell_mean
+ # idx = params.index(param)
+ # tmp = input_data[:, idx, :, :]
+ idx = params_i.index(param)
+ tmp = input_label[:, idx, :, :]
+ tmp = get_grid_cell_mean(tmp)
+ tmp = tmp[:, slc_y, slc_x]
+ tmp = normalize(tmp, param, mean_std_dct)
+ data_norm.append(tmp)
+
+ for param in data_params_full:
+ idx = params_i.index(param)
+ tmp = input_label[:, idx, :, :]
+
+ lo, hi, std, avg = get_min_max_std(tmp)
+ lo = normalize(lo, param, mean_std_dct)
+ hi = normalize(hi, param, mean_std_dct)
+ avg = normalize(avg, param, mean_std_dct)
+
+ data_norm.append(lo[:, slc_y, slc_x])
+ data_norm.append(hi[:, slc_y, slc_x])
+ data_norm.append(avg[:, slc_y, slc_x])
+ # ---------------------------------------------------
+ # If next uncommented, take out get_grid_cell_mean
+ # tmp = input_data[:, label_idx, :, :]
+ tmp = input_label[:, label_idx_i, :, :]
+ tmp = get_grid_cell_mean(tmp)
+ tmp = tmp[:, slc_y, slc_x]
+ data_norm.append(tmp)
+ # ---------
+ data = np.stack(data_norm, axis=3)
+ data = data.astype(np.float32)
+
+ # -----------------------------------------------------
+ # -----------------------------------------------------
+ label = input_label[:, label_idx_i, :, :]
+ label = label[:, y_128, x_128]
+ if NumClasses == 5:
+ label = get_label_data_5cat(label)
+ else:
+ label = get_label_data(label)
+
+ label = np.where(np.isnan(label), 0, label)
+ label = np.expand_dims(label, axis=3)
+
+ data = data.astype(np.float32)
+ label = label.astype(np.float32)
+
+ if is_training and DO_AUGMENT:
+ data_ud = np.flip(data, axis=1)
+ label_ud = np.flip(label, axis=1)
+
+ data_lr = np.flip(data, axis=2)
+ label_lr = np.flip(label, axis=2)
+
+ data = np.concatenate([data, data_ud, data_lr])
+ label = np.concatenate([label, label_ud, label_lr])
+
+ return data, label
+
+ def get_in_mem_data_batch_train(self, idxs):
+ return self.get_in_mem_data_batch(idxs, True)
+
+ def get_in_mem_data_batch_test(self, idxs):
+ return self.get_in_mem_data_batch(idxs, False)
+
+ @tf.function(input_signature=[tf.TensorSpec(None, tf.int32)])
+ def data_function(self, indexes):
+ out = tf.numpy_function(self.get_in_mem_data_batch_train, [indexes], [tf.float32, tf.float32])
+ return out
+
+ @tf.function(input_signature=[tf.TensorSpec(None, tf.int32)])
+ def data_function_test(self, indexes):
+ out = tf.numpy_function(self.get_in_mem_data_batch_test, [indexes], [tf.float32, tf.float32])
+ return out
+
+ def get_train_dataset(self, indexes):
+ indexes = list(indexes)
+
+ dataset = tf.data.Dataset.from_tensor_slices(indexes)
+ dataset = dataset.batch(PROC_BATCH_SIZE)
+ dataset = dataset.map(self.data_function, num_parallel_calls=AUTOTUNE)
+ dataset = dataset.cache()
+ if DO_AUGMENT:
+ dataset = dataset.shuffle(PROC_BATCH_BUFFER_SIZE)
+ dataset = dataset.prefetch(buffer_size=AUTOTUNE)
+ self.train_dataset = dataset
+
+ def get_test_dataset(self, indexes):
+ indexes = list(indexes)
+
+ dataset = tf.data.Dataset.from_tensor_slices(indexes)
+ dataset = dataset.batch(PROC_BATCH_SIZE)
+ dataset = dataset.map(self.data_function_test, num_parallel_calls=AUTOTUNE)
+ dataset = dataset.cache()
+ self.test_dataset = dataset
+
+ def setup_pipeline(self, train_data_files, train_label_files, test_data_files, test_label_files, num_train_samples):
+ self.train_data_files = train_data_files
+ self.train_label_files = train_label_files
+ self.test_data_files = test_data_files
+ self.test_label_files = test_label_files
+
+ trn_idxs = np.arange(len(train_data_files))
+ np.random.shuffle(trn_idxs)
+
+ tst_idxs = np.arange(len(test_data_files))
+
+ self.get_train_dataset(trn_idxs)
+ self.get_test_dataset(tst_idxs)
+
+ self.num_data_samples = num_train_samples # approximately
+
+ print('datetime: ', now)
+ print('training and test data: ')
+ print('---------------------------')
+ print('num train samples: ', self.num_data_samples)
+ print('BATCH SIZE: ', BATCH_SIZE)
+ print('num test samples: ', tst_idxs.shape[0])
+ print('setup_pipeline: Done')
+
+ def setup_test_pipeline(self, test_data_files, test_label_files):
+ self.test_data_files = test_data_files
+ self.test_label_files = test_label_files
+ tst_idxs = np.arange(len(test_data_files))
+ self.get_test_dataset(tst_idxs)
+ print('setup_test_pipeline: Done')
+
+ def build_srcnn(self, do_drop_out=False, do_batch_norm=False, drop_rate=0.5, factor=2):
+ print('build_cnn')
+ padding = "SAME"
+
+ # activation = tf.nn.relu
+ # activation = tf.nn.elu
+ activation = tf.nn.relu
+ momentum = 0.99
+
+ num_filters = 64
+
+ input_2d = self.inputs[0]
+ print('input: ', input_2d.shape)
+
+ conv = conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=KERNEL_SIZE, kernel_initializer='he_uniform', activation=activation, padding='VALID')(input_2d)
+ print(conv.shape)
+
+ # if NOISE_TRAINING:
+ # conv = conv_b = tf.keras.layers.GaussianNoise(stddev=NOISE_STDDEV)(conv)
+
+ scale = 0.2
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_1', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_2', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_3', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_4', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_5', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_6', kernel_size=KERNEL_SIZE, scale=scale)
+
+ conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=3, strides=1, activation=activation, kernel_initializer='he_uniform', padding=padding)(conv_b)
+
+ # conv = conv + conv_b
+ conv = conv_b
+ print(conv.shape)
+
+ if NumClasses == 2:
+ final_activation = tf.nn.sigmoid # For binary
+ else:
+ final_activation = tf.nn.softmax # For multi-class
+
+ # This is effectively a Dense layer
+ self.logits = tf.keras.layers.Conv2D(NumLogits, kernel_size=1, strides=1, padding=padding, activation=final_activation)(conv)
+ print(self.logits.shape)
+
+ def build_training(self):
+ if NumClasses == 2:
+ self.loss = tf.keras.losses.BinaryCrossentropy(from_logits=False) # for two-class only
+ else:
+ self.loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False) # For multi-class
+ # self.loss = tf.keras.losses.MeanAbsoluteError() # Regression
+
+ # decayed_learning_rate = learning_rate * decay_rate ^ (global_step / decay_steps)
+ initial_learning_rate = 0.002
+ decay_rate = 0.95
+ steps_per_epoch = int(self.num_data_samples/BATCH_SIZE) # one epoch
+ decay_steps = int(steps_per_epoch) * 4
+ print('initial rate, decay rate, steps/epoch, decay steps: ', initial_learning_rate, decay_rate, steps_per_epoch, decay_steps)
+
+ self.learningRateSchedule = tf.keras.optimizers.schedules.ExponentialDecay(initial_learning_rate, decay_steps, decay_rate)
+
+ optimizer = tf.keras.optimizers.Adam(learning_rate=self.learningRateSchedule)
+
+ if TRACK_MOVING_AVERAGE:
+ # Not sure that this works properly (from tfa)
+ # optimizer = tfa.optimizers.MovingAverage(optimizer)
+ self.ema = tf.train.ExponentialMovingAverage(decay=0.9999)
+
+ self.optimizer = optimizer
+ self.initial_learning_rate = initial_learning_rate
+
+ def build_evaluation(self):
+ self.train_loss = tf.keras.metrics.Mean(name='train_loss')
+ self.test_loss = tf.keras.metrics.Mean(name='test_loss')
+
+ if NumClasses == 2:
+ self.train_accuracy = tf.keras.metrics.BinaryAccuracy(name='train_accuracy')
+ self.test_accuracy = tf.keras.metrics.BinaryAccuracy(name='test_accuracy')
+ self.test_auc = tf.keras.metrics.AUC(name='test_auc')
+ self.test_recall = tf.keras.metrics.Recall(name='test_recall')
+ self.test_precision = tf.keras.metrics.Precision(name='test_precision')
+ self.test_true_neg = tf.keras.metrics.TrueNegatives(name='test_true_neg')
+ self.test_true_pos = tf.keras.metrics.TruePositives(name='test_true_pos')
+ self.test_false_neg = tf.keras.metrics.FalseNegatives(name='test_false_neg')
+ self.test_false_pos = tf.keras.metrics.FalsePositives(name='test_false_pos')
+ else:
+ self.train_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='train_accuracy')
+ self.test_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='test_accuracy')
+
+ @tf.function(input_signature=[tf.TensorSpec(None, tf.float32), tf.TensorSpec(None, tf.float32)])
+ def train_step(self, inputs, labels):
+ labels = tf.squeeze(labels, axis=[3])
+ with tf.GradientTape() as tape:
+ pred = self.model([inputs], training=True)
+ loss = self.loss(labels, pred)
+ total_loss = loss
+ if len(self.model.losses) > 0:
+ reg_loss = tf.math.add_n(self.model.losses)
+ total_loss = loss + reg_loss
+ gradients = tape.gradient(total_loss, self.model.trainable_variables)
+ self.optimizer.apply_gradients(zip(gradients, self.model.trainable_variables))
+ if TRACK_MOVING_AVERAGE:
+ self.ema.apply(self.model.trainable_variables)
+
+ self.train_loss(loss)
+ self.train_accuracy(labels, pred)
+
+ return loss
+
+ @tf.function(input_signature=[tf.TensorSpec(None, tf.float32), tf.TensorSpec(None, tf.float32)])
+ def test_step(self, inputs, labels):
+ labels = tf.squeeze(labels, axis=[3])
+ pred = self.model([inputs], training=False)
+ t_loss = self.loss(labels, pred)
+
+ self.test_loss(t_loss)
+ self.test_accuracy(labels, pred)
+
+ # @tf.function(input_signature=[tf.TensorSpec(None, tf.float32), tf.TensorSpec(None, tf.float32)])
+ # decorator commented out because pred.numpy(): pred not evaluated yet.
+ def predict(self, inputs, labels):
+ pred = self.model([inputs], training=False)
+ # t_loss = self.loss(tf.squeeze(labels, axis=[3]), pred)
+ t_loss = self.loss(labels, pred)
+
+ self.test_labels.append(labels)
+ self.test_preds.append(pred.numpy())
+
+ self.test_loss(t_loss)
+ self.test_accuracy(labels, pred)
+
+ def reset_test_metrics(self):
+ self.test_loss.reset_states()
+ self.test_accuracy.reset_states()
+
+ def get_metrics(self):
+ recall = self.test_recall.result()
+ precsn = self.test_precision.result()
+ f1 = 2 * (precsn * recall) / (precsn + recall)
+
+ tn = self.test_true_neg.result()
+ tp = self.test_true_pos.result()
+ fn = self.test_false_neg.result()
+ fp = self.test_false_pos.result()
+
+ mcc = ((tp * tn) - (fp * fn)) / np.sqrt((tp + fp) * (tp + fn) * (tn + fp) * (tn + fn))
+ return f1, mcc
+
+ def do_training(self, ckpt_dir=None):
+
+ if ckpt_dir is None:
+ if not os.path.exists(modeldir):
+ os.mkdir(modeldir)
+ ckpt = tf.train.Checkpoint(step=tf.Variable(1), model=self.model)
+ ckpt_manager = tf.train.CheckpointManager(ckpt, modeldir, max_to_keep=3)
+ else:
+ ckpt = tf.train.Checkpoint(step=tf.Variable(1), model=self.model)
+ ckpt_manager = tf.train.CheckpointManager(ckpt, ckpt_dir, max_to_keep=3)
+ ckpt.restore(ckpt_manager.latest_checkpoint)
+
+ self.writer_train = tf.summary.create_file_writer(os.path.join(logdir, 'plot_train'))
+ self.writer_valid = tf.summary.create_file_writer(os.path.join(logdir, 'plot_valid'))
+ self.writer_train_valid_loss = tf.summary.create_file_writer(os.path.join(logdir, 'plot_train_valid_loss'))
+
+ step = 0
+ total_time = 0
+ best_test_loss = np.finfo(dtype=np.float).max
+
+ if EARLY_STOP:
+ es = EarlyStop()
+
+ for epoch in range(NUM_EPOCHS):
+ self.train_loss.reset_states()
+ self.train_accuracy.reset_states()
+
+ t0 = datetime.datetime.now().timestamp()
+
+ proc_batch_cnt = 0
+ n_samples = 0
+
+ for data, label in self.train_dataset:
+ trn_ds = tf.data.Dataset.from_tensor_slices((data, label))
+ trn_ds = trn_ds.batch(BATCH_SIZE)
+ for mini_batch in trn_ds:
+ if self.learningRateSchedule is not None:
+ loss = self.train_step(mini_batch[0], mini_batch[1])
+
+ if (step % 100) == 0:
+
+ with self.writer_train.as_default():
+ tf.summary.scalar('loss_trn', loss.numpy(), step=step)
+ tf.summary.scalar('learning_rate', self.optimizer._decayed_lr('float32').numpy(), step=step)
+ tf.summary.scalar('num_train_steps', step, step=step)
+ tf.summary.scalar('num_epochs', epoch, step=step)
+
+ self.reset_test_metrics()
+ for data_tst, label_tst in self.test_dataset:
+ tst_ds = tf.data.Dataset.from_tensor_slices((data_tst, label_tst))
+ tst_ds = tst_ds.batch(BATCH_SIZE)
+ for mini_batch_test in tst_ds:
+ self.test_step(mini_batch_test[0], mini_batch_test[1])
+
+ with self.writer_valid.as_default():
+ tf.summary.scalar('loss_val', self.test_loss.result(), step=step)
+ tf.summary.scalar('acc_val', self.test_accuracy.result(), step=step)
+
+ with self.writer_train_valid_loss.as_default():
+ tf.summary.scalar('loss_trn', loss.numpy(), step=step)
+ tf.summary.scalar('loss_val', self.test_loss.result(), step=step)
+
+ print('****** test loss, acc, lr: ', self.test_loss.result().numpy(), self.test_accuracy.result().numpy(),
+ self.optimizer._decayed_lr('float32').numpy())
+
+ step += 1
+ print('train loss: ', loss.numpy())
+
+ proc_batch_cnt += 1
+ n_samples += data.shape[0]
+ print('proc_batch_cnt: ', proc_batch_cnt, n_samples)
+
+ t1 = datetime.datetime.now().timestamp()
+ print('End of Epoch: ', epoch+1, 'elapsed time: ', (t1-t0))
+ total_time += (t1-t0)
+
+ self.reset_test_metrics()
+ for data, label in self.test_dataset:
+ ds = tf.data.Dataset.from_tensor_slices((data, label))
+ ds = ds.batch(BATCH_SIZE)
+ for mini_batch in ds:
+ self.test_step(mini_batch[0], mini_batch[1])
+
+ print('loss, acc: ', self.test_loss.result().numpy(), self.test_accuracy.result().numpy())
+ print('------------------------------------------------------')
+
+ tst_loss = self.test_loss.result().numpy()
+ if tst_loss < best_test_loss:
+ best_test_loss = tst_loss
+ ckpt_manager.save()
+
+ if EARLY_STOP and es.check_stop(tst_loss):
+ break
+
+ print('total time: ', total_time)
+ self.writer_train.close()
+ self.writer_valid.close()
+ self.writer_train_valid_loss.close()
+
+ def build_model(self):
+ self.build_srcnn()
+ self.model = tf.keras.Model(self.inputs, self.logits)
+
+ def restore(self, ckpt_dir):
+
+ ckpt = tf.train.Checkpoint(step=tf.Variable(1), model=self.model)
+ ckpt_manager = tf.train.CheckpointManager(ckpt, ckpt_dir, max_to_keep=3)
+ ckpt.restore(ckpt_manager.latest_checkpoint)
+
+ self.reset_test_metrics()
+
+ for data, label in self.test_dataset:
+ ds = tf.data.Dataset.from_tensor_slices((data, label))
+ ds = ds.batch(BATCH_SIZE)
+ for mini_batch_test in ds:
+ self.predict(mini_batch_test[0], mini_batch_test[1])
+
+ print('loss, acc: ', self.test_loss.result().numpy(), self.test_accuracy.result().numpy())
+
+ labels = np.concatenate(self.test_labels)
+ preds = np.concatenate(self.test_preds)
+ print(labels.shape, preds.shape)
+
+ return labels, preds
+
+ def do_evaluate(self, inputs, ckpt_dir):
+
+ ckpt = tf.train.Checkpoint(step=tf.Variable(1), model=self.model)
+ ckpt_manager = tf.train.CheckpointManager(ckpt, ckpt_dir, max_to_keep=3)
+ ckpt.restore(ckpt_manager.latest_checkpoint)
+
+ self.reset_test_metrics()
+
+ pred = self.model([inputs], training=False)
+ self.test_probs = pred
+ pred = pred.numpy()
+
+ return pred
+
+ def run(self, directory, ckpt_dir=None, num_data_samples=50000):
+ train_data_files = glob.glob(directory+'train*mres*.npy')
+ valid_data_files = glob.glob(directory+'valid*mres*.npy')
+ train_label_files = glob.glob(directory+'train*ires*.npy')
+ valid_label_files = glob.glob(directory+'valid*ires*.npy')
+ self.setup_pipeline(train_data_files, train_label_files, valid_data_files, valid_label_files, num_data_samples)
+
+ self.build_model()
+ self.build_training()
+ self.build_evaluation()
+ self.do_training(ckpt_dir=ckpt_dir)
+
+ def run_restore(self, directory, ckpt_dir):
+ self.num_data_samples = 1000
+
+ valid_data_files = glob.glob(directory + 'valid*mres*.npy')
+ valid_label_files = glob.glob(directory + 'valid*ires*.npy')
+ self.setup_test_pipeline(valid_data_files, valid_label_files)
+
+ self.build_model()
+ self.build_training()
+ self.build_evaluation()
+ return self.restore(ckpt_dir)
+
+ def run_evaluate(self, data, ckpt_dir):
+ # data = tf.convert_to_tensor(data, dtype=tf.float32)
+ self.num_data_samples = 80000
+ self.build_model()
+ self.build_training()
+ self.build_evaluation()
+ return self.do_evaluate(data, ckpt_dir)
+
+
+def run_restore_static(directory, ckpt_dir, out_file=None):
+ nn = SRCNN()
+ labels, preds = nn.run_restore(directory, ckpt_dir)
+ if out_file is not None:
+ np.save(out_file,
+ [np.squeeze(labels), preds.argmax(axis=3)])
+
+
+def run_evaluate_static(in_file, out_file, ckpt_dir):
+ gc.collect()
+
+ h5f = h5py.File(in_file, 'r')
+
+ bt = get_grid_values_all(h5f, 'orig/temp_11_0um')
+ y_len, x_len = bt.shape[0], bt.shape[1]
+ lons = get_grid_values_all(h5f, 'orig/longitude')
+ lats = get_grid_values_all(h5f, 'orig/latitude')
+ bt = np.where(np.isnan(bt), 0, bt)
+ bt = normalize(bt, 'temp_11_0um_nom', mean_std_dct)
+
+ refl = get_grid_values_all(h5f, 'super/refl_0_65um')
+ refl = np.where(np.isnan(refl), 0, refl)
+ refl = np.expand_dims(refl, axis=0)
+ refl_lo, refl_hi, refl_std, refl_avg = get_min_max_std(refl)
+ refl_lo = normalize(refl_lo, 'refl_0_65um_nom', mean_std_dct)
+ refl_hi = normalize(refl_hi, 'refl_0_65um_nom', mean_std_dct)
+ refl_avg = normalize(refl_avg, 'refl_0_65um_nom', mean_std_dct)
+ refl_lo = np.squeeze(refl_lo)
+ refl_hi = np.squeeze(refl_hi)
+ refl_avg = np.squeeze(refl_avg)
+
+ cp = get_grid_values_all(h5f, 'orig/'+label_param)
+ cp = np.where(np.isnan(cp), 0, cp)
+
+ data = np.stack([bt, refl_lo, refl_hi, refl_avg, cp], axis=2)
+ data = np.expand_dims(data, axis=0)
+
+ h5f.close()
+
+ nn = SRCNN()
+ probs = nn.run_evaluate(data, ckpt_dir)
+ cld_frac = probs.argmax(axis=3)
+ cld_frac = cld_frac.astype(np.int8)
+ cld_frac_out = np.zeros((y_len, x_len), dtype=np.int8)
+ border = int((KERNEL_SIZE - 1)/2)
+ cld_frac_out[border:y_len - border, border:x_len - border] = cld_frac[0, :, :]
+
+ bt = denormalize(bt, 'temp_11_0um_nom', mean_std_dct)
+ refl_avg = denormalize(refl_avg, 'refl_0_65um_nom', mean_std_dct)
+
+ var_names = ['cloud_fraction', 'temp_11_0um', 'refl_0_65um']
+ dims = ['num_params', 'y', 'x']
+
+ da = xr.DataArray(np.stack([cld_frac_out, bt, refl_avg], axis=0), dims=dims)
+ da.assign_coords({
+ 'num_params': var_names,
+ 'lat': (['y', 'x'], lats),
+ 'lon': (['y', 'x'], lons)
+ })
+
+ if out_file is not None:
+ np.save(out_file, (cld_frac_out, bt, refl_avg, cp, lons, lats))
+ else:
+ return [cld_frac_out, bt, refl_avg, cp, lons, lats]
+
+
+def analyze_3cat(file):
+
+ tup = np.load(file, allow_pickle=True)
+ lbls = tup[0]
+ pred = tup[1]
+
+ lbls = lbls.flatten()
+ pred = pred.flatten()
+ print(np.sum(lbls == 0), np.sum(lbls == 1), np.sum(lbls == 2))
+
+ msk_0_1 = lbls != 2
+ msk_1_2 = lbls != 0
+ msk_0_2 = lbls != 1
+
+ lbls_0_1 = lbls[msk_0_1]
+
+ pred_0_1 = pred[msk_0_1]
+ pred_0_1 = np.where(pred_0_1 == 2, 1, pred_0_1)
+
+ # ----
+ lbls_1_2 = lbls[msk_1_2]
+ lbls_1_2 = np.where(lbls_1_2 == 1, 0, lbls_1_2)
+ lbls_1_2 = np.where(lbls_1_2 == 2, 1, lbls_1_2)
+
+ pred_1_2 = pred[msk_1_2]
+ pred_1_2 = np.where(pred_1_2 == 0, -9, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == 1, 0, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == 2, 1, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == -9, 1, pred_1_2)
+
+ # ----
+ lbls_0_2 = lbls[msk_0_2]
+ lbls_0_2 = np.where(lbls_0_2 == 2, 1, lbls_0_2)
+
+ pred_0_2 = pred[msk_0_2]
+ pred_0_2 = np.where(pred_0_2 == 2, 1, pred_0_2)
+
+ cm_0_1 = confusion_matrix_values(lbls_0_1, pred_0_1)
+ cm_1_2 = confusion_matrix_values(lbls_1_2, pred_1_2)
+ cm_0_2 = confusion_matrix_values(lbls_0_2, pred_0_2)
+
+ true_0_1 = (lbls_0_1 == 0) & (pred_0_1 == 0)
+ false_0_1 = (lbls_0_1 == 1) & (pred_0_1 == 0)
+
+ true_no_0_1 = (lbls_0_1 == 1) & (pred_0_1 == 1)
+ false_no_0_1 = (lbls_0_1 == 0) & (pred_0_1 == 1)
+
+ true_0_2 = (lbls_0_2 == 0) & (pred_0_2 == 0)
+ false_0_2 = (lbls_0_2 == 1) & (pred_0_2 == 0)
+
+ true_no_0_2 = (lbls_0_2 == 1) & (pred_0_2 == 1)
+ false_no_0_2 = (lbls_0_2 == 0) & (pred_0_2 == 1)
+
+ true_1_2 = (lbls_1_2 == 0) & (pred_1_2 == 0)
+ false_1_2 = (lbls_1_2 == 1) & (pred_1_2 == 0)
+
+ true_no_1_2 = (lbls_1_2 == 1) & (pred_1_2 == 1)
+ false_no_1_2 = (lbls_1_2 == 0) & (pred_1_2 == 1)
+
+ tp_0 = np.sum(true_0_1).astype(np.float64)
+ tp_1 = np.sum(true_1_2).astype(np.float64)
+ tp_2 = np.sum(true_0_2).astype(np.float64)
+
+ tn_0 = np.sum(true_no_0_1).astype(np.float64)
+ tn_1 = np.sum(true_no_1_2).astype(np.float64)
+ tn_2 = np.sum(true_no_0_2).astype(np.float64)
+
+ fp_0 = np.sum(false_0_1).astype(np.float64)
+ fp_1 = np.sum(false_1_2).astype(np.float64)
+ fp_2 = np.sum(false_0_2).astype(np.float64)
+
+ fn_0 = np.sum(false_no_0_1).astype(np.float64)
+ fn_1 = np.sum(false_no_1_2).astype(np.float64)
+ fn_2 = np.sum(false_no_0_2).astype(np.float64)
+
+ recall_0 = tp_0 / (tp_0 + fn_0)
+ recall_1 = tp_1 / (tp_1 + fn_1)
+ recall_2 = tp_2 / (tp_2 + fn_2)
+
+ precision_0 = tp_0 / (tp_0 + fp_0)
+ precision_1 = tp_1 / (tp_1 + fp_1)
+ precision_2 = tp_2 / (tp_2 + fp_2)
+
+ mcc_0 = ((tp_0 * tn_0) - (fp_0 * fn_0)) / np.sqrt((tp_0 + fp_0) * (tp_0 + fn_0) * (tn_0 + fp_0) * (tn_0 + fn_0))
+ mcc_1 = ((tp_1 * tn_1) - (fp_1 * fn_1)) / np.sqrt((tp_1 + fp_1) * (tp_1 + fn_1) * (tn_1 + fp_1) * (tn_1 + fn_1))
+ mcc_2 = ((tp_2 * tn_2) - (fp_2 * fn_2)) / np.sqrt((tp_2 + fp_2) * (tp_2 + fn_2) * (tn_2 + fp_2) * (tn_2 + fn_2))
+
+ acc_0 = np.sum(lbls_0_1 == pred_0_1)/pred_0_1.size
+ acc_1 = np.sum(lbls_1_2 == pred_1_2)/pred_1_2.size
+ acc_2 = np.sum(lbls_0_2 == pred_0_2)/pred_0_2.size
+
+ print(acc_0, recall_0, precision_0, mcc_0)
+ print(acc_1, recall_1, precision_1, mcc_1)
+ print(acc_2, recall_2, precision_2, mcc_2)
+
+ return cm_0_1, cm_1_2, cm_0_2, [acc_0, acc_1, acc_2], [recall_0, recall_1, recall_2],\
+ [precision_0, precision_1, precision_2], [mcc_0, mcc_1, mcc_2]
+
+
+def analyze_5cat(file):
+
+ tup = np.load(file, allow_pickle=True)
+ lbls = tup[0]
+ pred = tup[1]
+
+ lbls = lbls.flatten()
+ pred = pred.flatten()
+ np.histogram(lbls, bins=5)
+ np.histogram(pred, bins=5)
+
+ new_lbls = np.zeros(lbls.size, dtype=np.int32)
+ new_pred = np.zeros(pred.size, dtype=np.int32)
+
+ new_lbls[lbls == 0] = 0
+ new_lbls[lbls == 1] = 1
+ new_lbls[lbls == 2] = 1
+ new_lbls[lbls == 3] = 1
+ new_lbls[lbls == 4] = 2
+
+ new_pred[pred == 0] = 0
+ new_pred[pred == 1] = 1
+ new_pred[pred == 2] = 1
+ new_pred[pred == 3] = 1
+ new_pred[pred == 4] = 2
+
+ np.histogram(new_lbls, bins=3)
+ np.histogram(new_pred, bins=3)
+
+ lbls = new_lbls
+ pred = new_pred
+
+ print(np.sum(lbls == 0), np.sum(lbls == 1), np.sum(lbls == 2))
+
+ msk_0_1 = lbls != 2
+ msk_1_2 = lbls != 0
+ msk_0_2 = lbls != 1
+
+ lbls_0_1 = lbls[msk_0_1]
+
+ pred_0_1 = pred[msk_0_1]
+ pred_0_1 = np.where(pred_0_1 == 2, 1, pred_0_1)
+
+ # ----------------------------------------------
+ lbls_1_2 = lbls[msk_1_2]
+ lbls_1_2 = np.where(lbls_1_2 == 1, 0, lbls_1_2)
+ lbls_1_2 = np.where(lbls_1_2 == 2, 1, lbls_1_2)
+
+ pred_1_2 = pred[msk_1_2]
+ pred_1_2 = np.where(pred_1_2 == 0, -9, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == 1, 0, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == 2, 1, pred_1_2)
+ pred_1_2 = np.where(pred_1_2 == -9, 1, pred_1_2)
+
+ # -----------------------------------------------
+ lbls_0_2 = lbls[msk_0_2]
+ lbls_0_2 = np.where(lbls_0_2 == 2, 1, lbls_0_2)
+
+ pred_0_2 = pred[msk_0_2]
+ pred_0_2 = np.where(pred_0_2 == 2, 1, pred_0_2)
+
+ cm_0_1 = confusion_matrix_values(lbls_0_1, pred_0_1)
+ cm_1_2 = confusion_matrix_values(lbls_1_2, pred_1_2)
+ cm_0_2 = confusion_matrix_values(lbls_0_2, pred_0_2)
+
+ true_0_1 = (lbls_0_1 == 0) & (pred_0_1 == 0)
+ false_0_1 = (lbls_0_1 == 1) & (pred_0_1 == 0)
+
+ true_no_0_1 = (lbls_0_1 == 1) & (pred_0_1 == 1)
+ false_no_0_1 = (lbls_0_1 == 0) & (pred_0_1 == 1)
+
+ true_0_2 = (lbls_0_2 == 0) & (pred_0_2 == 0)
+ false_0_2 = (lbls_0_2 == 1) & (pred_0_2 == 0)
+
+ true_no_0_2 = (lbls_0_2 == 1) & (pred_0_2 == 1)
+ false_no_0_2 = (lbls_0_2 == 0) & (pred_0_2 == 1)
+
+ true_1_2 = (lbls_1_2 == 0) & (pred_1_2 == 0)
+ false_1_2 = (lbls_1_2 == 1) & (pred_1_2 == 0)
+
+ true_no_1_2 = (lbls_1_2 == 1) & (pred_1_2 == 1)
+ false_no_1_2 = (lbls_1_2 == 0) & (pred_1_2 == 1)
+
+ tp_0 = np.sum(true_0_1).astype(np.float64)
+ tp_1 = np.sum(true_1_2).astype(np.float64)
+ tp_2 = np.sum(true_0_2).astype(np.float64)
+
+ tn_0 = np.sum(true_no_0_1).astype(np.float64)
+ tn_1 = np.sum(true_no_1_2).astype(np.float64)
+ tn_2 = np.sum(true_no_0_2).astype(np.float64)
+
+ fp_0 = np.sum(false_0_1).astype(np.float64)
+ fp_1 = np.sum(false_1_2).astype(np.float64)
+ fp_2 = np.sum(false_0_2).astype(np.float64)
+
+ fn_0 = np.sum(false_no_0_1).astype(np.float64)
+ fn_1 = np.sum(false_no_1_2).astype(np.float64)
+ fn_2 = np.sum(false_no_0_2).astype(np.float64)
+
+ recall_0 = tp_0 / (tp_0 + fn_0)
+ recall_1 = tp_1 / (tp_1 + fn_1)
+ recall_2 = tp_2 / (tp_2 + fn_2)
+
+ precision_0 = tp_0 / (tp_0 + fp_0)
+ precision_1 = tp_1 / (tp_1 + fp_1)
+ precision_2 = tp_2 / (tp_2 + fp_2)
+
+ mcc_0 = ((tp_0 * tn_0) - (fp_0 * fn_0)) / np.sqrt((tp_0 + fp_0) * (tp_0 + fn_0) * (tn_0 + fp_0) * (tn_0 + fn_0))
+ mcc_1 = ((tp_1 * tn_1) - (fp_1 * fn_1)) / np.sqrt((tp_1 + fp_1) * (tp_1 + fn_1) * (tn_1 + fp_1) * (tn_1 + fn_1))
+ mcc_2 = ((tp_2 * tn_2) - (fp_2 * fn_2)) / np.sqrt((tp_2 + fp_2) * (tp_2 + fn_2) * (tn_2 + fp_2) * (tn_2 + fn_2))
+
+ acc_0 = np.sum(lbls_0_1 == pred_0_1)/pred_0_1.size
+ acc_1 = np.sum(lbls_1_2 == pred_1_2)/pred_1_2.size
+ acc_2 = np.sum(lbls_0_2 == pred_0_2)/pred_0_2.size
+
+ print(acc_0, recall_0, precision_0, mcc_0)
+ print(acc_1, recall_1, precision_1, mcc_1)
+ print(acc_2, recall_2, precision_2, mcc_2)
+
+ return cm_0_1, cm_1_2, cm_0_2, [acc_0, acc_1, acc_2], [recall_0, recall_1, recall_2],\
+ [precision_0, precision_1, precision_2], [mcc_0, mcc_1, mcc_2], lbls, pred
+
+
+if __name__ == "__main__":
+ nn = SRCNN()
+ nn.run('matchup_filename')
--
GitLab