From 555cacad999fb24f52803f28eb221ac04b7607d7 Mon Sep 17 00:00:00 2001
From: Jan <jan.mandel@gmail.com>
Date: Sat, 18 Feb 2023 22:16:55 -0700
Subject: [PATCH] in fmda_kf_rnn_orig call moisture_rnn.create_RNN_2 ->
 moisture_rnn.create_RNN_2 print x_train y_train shape in fmda_kf_rnn_orig and
 fmda_rnn_raine verbose=1 in model.fit still different, same as before

---
 fmda/fmda_kf_rnn_orig.ipynb | 27 +++++++++++++++++++++------
 fmda/fmda_rnn_rain.ipynb    | 18 ++++++++++++++++--
 2 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/fmda/fmda_kf_rnn_orig.ipynb b/fmda/fmda_kf_rnn_orig.ipynb
index dfc29eb..316825e 100644
--- a/fmda/fmda_kf_rnn_orig.ipynb
+++ b/fmda/fmda_kf_rnn_orig.ipynb
@@ -1500,7 +1500,9 @@
     "h0 = tf.convert_to_tensor(datat[:samples],dtype=tf.float32)\n",
     "# print('initial state=',h0)\n",
     "# statefull model version for traning\n",
-    "model_fit=create_RNN_2(hidden_units=hidden_units, \n",
+    "import moisture_rnn\n",
+    "# model_fit=create_RNN_2(hidden_units=hidden_units, \n",
+    "model_fit=moisture_rnn.create_RNN_2(hidden_units=hidden_units, \n",
     "                        dense_units=dense_units, \n",
     "                        batch_shape=(samples,timesteps,features),\n",
     "                        stateful=True,\n",
@@ -2694,7 +2696,8 @@
     "h0 = tf.convert_to_tensor(datat[:samples],dtype=tf.float32)\n",
     "# print('initial state=',h0)\n",
     "# statefull model version for traning\n",
-    "model_fit=create_RNN_2(hidden_units=hidden_units, \n",
+    "import moisture_rnn\n",
+    "model_fit=moisture_rnn.create_RNN_2(hidden_units=hidden_units, \n",
     "                        dense_units=dense_units, \n",
     "                        batch_shape=(samples,timesteps,features),\n",
     "                        stateful=True,\n",
@@ -2957,7 +2960,8 @@
     "h0 = tf.convert_to_tensor(datat[:samples],dtype=tf.float32)\n",
     "# print('initial state=',h0)\n",
     "# statefull model version for traning\n",
-    "model_fit=create_RNN_2(hidden_units=hidden_units, \n",
+    "import moisture_rnn\n",
+    "model_fit=moisture_rnn.create_RNN_2(hidden_units=hidden_units, \n",
     "                        dense_units=dense_units, \n",
     "                        batch_shape=(samples,timesteps,features),\n",
     "                        stateful=True,\n",
@@ -2968,7 +2972,7 @@
     "# same model stateless for prediction on the entire dataset - to start onlg\n",
     "# the real application will switch to prediction after training data end\n",
     "# and start from the state there\n",
-    "print('model_fit input shape',x_train.shape,'output shape',model_fit(x_train).shape)\n",
+    "print('model_fit input shape',x_train.shape,'output shape',model_fit(y_train).shape)\n",
     "from keras.utils.vis_utils import plot_model\n",
     "plot_model(model_fit, to_file='model_plot.png', \n",
     "           show_shapes=True, show_layer_names=True)"
@@ -3006,7 +3010,7 @@
    },
    "outputs": [],
    "source": [
-    "model_predict=create_RNN_2(hidden_units=hidden_units, dense_units=dense_units,  \n",
+    "model_predict=moisture_rnn.create_RNN_2(hidden_units=hidden_units, dense_units=dense_units,  \n",
     "                        input_shape=(hours,features),stateful = False,\n",
     "                        return_sequences=True,\n",
     "                        activation=activation,dense_layers=dense_layers)\n",
@@ -3074,6 +3078,17 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "print('model_fit input shape',x_train.shape,'output shape',y_train.shape)\n",
+    "print('x_train',x_train)\n",
+    "print('y_train',y_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "reproducibility.set_seed()"
    ]
   },
@@ -3092,7 +3107,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model_fit.fit(x_train, y_train, epochs=5000, verbose=0,batch_size=samples)\n",
+    "model_fit.fit(x_train, y_train, epochs=5000, verbose=1,batch_size=samples)\n",
     "w_fitted=model_fit.get_weights()\n",
     "for i in range(len(w)):\n",
     "  print('weight',i,' exact:',w_exact[i],':  initial:',w_initial[i],' fitted:',w_fitted[i])"
diff --git a/fmda/fmda_rnn_rain.ipynb b/fmda/fmda_rnn_rain.ipynb
index ca321d2..43628c0 100644
--- a/fmda/fmda_rnn_rain.ipynb
+++ b/fmda/fmda_rnn_rain.ipynb
@@ -487,8 +487,10 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "print(rnn_dat)\n",
     "x_train = rnn_dat['x_train']\n",
     "y_train = rnn_dat['y_train']\n",
+    "type(x_train)\n",
     "\n",
     "# fitting\n",
     "DeltaE = 0\n",
@@ -524,6 +526,18 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "id": "33d6b35c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print('model_fit input shape',x_train.shape,'output shape',y_train.shape)\n",
+    "print('x_train',x_train)\n",
+    "print('y_train',y_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
    "id": "7a2be797",
    "metadata": {},
    "outputs": [],
@@ -534,7 +548,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "adc56d77",
+   "id": "c7857fb4",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -548,7 +562,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model_fit.fit(x_train, y_train, epochs=5000, verbose=0, batch_size=samples)\n",
+    "model_fit.fit(x_train, y_train, epochs=5000, verbose=1, batch_size=samples)\n",
     "w_fitted=model_fit.get_weights()\n",
     "for i in range(len(w)):\n",
     "    vprint('weight',i,' exact:',w_exact[i],':  initial:',w_initial[i],' fitted:',w_fitted[i])\n",
-- 
2.11.4.GIT