modified notebook

_proc/04_mini_batch_training.ipynb

@@ -499,18 +499,48 @@
     "```"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/markdown": [
+       "---\n",
+       "\n",
+       "### accuracy\n",
+       "\n",
+       ">      accuracy (out, yb)"
+      ],
+      "text/plain": [
+       "---\n",
+       "\n",
+       "### accuracy\n",
+       "\n",
+       ">      accuracy (out, yb)"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#| echo: false\n",
+    "#| output: asis\n",
+    "show_doc(accuracy)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "517b1844",
+   "id": "dcca7a8d",
    "metadata": {
     "language": "python"
    },
    "outputs": [],
    "source": [
-    "def accuracy(out, yb):\n",
-    "    return (out.argmax(1)==yb).float().mean()\n",
-    "\n",
     "loss_func = F.cross_entropy"
    ]
   },
@@ -590,15 +620,35 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "ee7b70dc",
-   "metadata": {
-    "language": "python"
-   },
-   "outputs": [],
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/markdown": [
+       "---\n",
+       "\n",
+       "### report\n",
+       "\n",
+       ">      report (loss, preds, yb)"
+      ],
+      "text/plain": [
+       "---\n",
+       "\n",
+       "### report\n",
+       "\n",
+       ">      report (loss, preds, yb)"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "def report(loss, preds, yb):\n",
-    "    print(f\"loss: {loss:.2f}, accuracy: {accuracy(preds, yb):.2f}\")"
+    "#| echo: false\n",
+    "#| output: asis\n",
+    "show_doc(report)"
    ]
   },
   {
@@ -1422,7 +1472,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
@@ -1446,7 +1496,7 @@
        "Initialize self.  See help(type(self)) for accurate signature."
       ]
      },
-     "execution_count": 1,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2196,7 +2246,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -2206,17 +2256,17 @@
        "\n",
        "### fit\n",
        "\n",
-       ">      fit (epochs, model, loss_func, opt, train_dl, valid_ld)"
+       ">      fit (epochs, model, loss_func, opt, train_dl, valid_dl)"
       ],
       "text/plain": [
        "---\n",
        "\n",
        "### fit\n",
        "\n",
-       ">      fit (epochs, model, loss_func, opt, train_dl, valid_ld)"
+       ">      fit (epochs, model, loss_func, opt, train_dl, valid_dl)"
       ]
      },
-     "execution_count": 2,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2229,7 +2279,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -2249,7 +2299,7 @@
        ">      get_dls (train_ds, valid_ds, bs, **kwargs)"
       ]
      },
-     "execution_count": 3,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }

_proc/_docs/sitemap.xml

@@ -10,6 +10,6 @@
   </url>
   <url>
     <loc>https://arun477.github.io/practice_deep_learning/mini_batch_training.html</loc>
-    <lastmod>2024-02-19T15:03:49.698Z</lastmod>
+    <lastmod>2024-02-19T15:14:43.526Z</lastmod>
   </url>
 </urlset>

nbs/.ipynb_checkpoints/04_mini_batch_training-checkpoint.ipynb

@@ -489,9 +489,19 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "def accuracy(out, yb):\n",
-    "    return (out.argmax(1)==yb).float().mean()\n",
+    "#|export\n",
     "\n",
+    "def accuracy(out, yb):\n",
+    "    return (out.argmax(1)==yb).float().mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "dcca7a8d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "loss_func = F.cross_entropy"
    ]
   },
@@ -570,6 +580,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "#|export\n",
+    "\n",
     "def report(loss, preds, yb):\n",
     "    print(f\"loss: {loss:.2f}, accuracy: {accuracy(preds, yb):.2f}\")"
    ]
@@ -2029,7 +2041,7 @@
    "source": [
     "#|export\n",
     "\n",
-    "def fit(epochs, model, loss_func, opt, train_dl, valid_ld):\n",
+    "def fit(epochs, model, loss_func, opt, train_dl, valid_dl):\n",
     "    for epoch in range(epochs):\n",
     "        model.train()\n",
     "        for xb, yb in train_dl:\n",

nbs/04_mini_batch_training.ipynb

@@ -489,9 +489,19 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "def accuracy(out, yb):\n",
-    "    return (out.argmax(1)==yb).float().mean()\n",
+    "#|export\n",
     "\n",
+    "def accuracy(out, yb):\n",
+    "    return (out.argmax(1)==yb).float().mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "dcca7a8d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "loss_func = F.cross_entropy"
    ]
   },
@@ -570,6 +580,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "#|export\n",
+    "\n",
     "def report(loss, preds, yb):\n",
     "    print(f\"loss: {loss:.2f}, accuracy: {accuracy(preds, yb):.2f}\")"
    ]
@@ -2029,7 +2041,7 @@
    "source": [
     "#|export\n",
     "\n",
-    "def fit(epochs, model, loss_func, opt, train_dl, valid_ld):\n",
+    "def fit(epochs, model, loss_func, opt, train_dl, valid_dl):\n",
     "    for epoch in range(epochs):\n",
     "        model.train()\n",
     "        for xb, yb in train_dl:\n",

practice_deep_learning/_modidx.py

@@ -13,7 +13,11 @@
                                                                                                        'practice_deep_learning/training.py'),
                                                  'practice_deep_learning.training.Dataset.__len__': ( 'mini_batch_training.html#dataset.__len__',
                                                                                                       'practice_deep_learning/training.py'),
+                                                 'practice_deep_learning.training.accuracy': ( 'mini_batch_training.html#accuracy',
+                                                                                               'practice_deep_learning/training.py'),
                                                  'practice_deep_learning.training.fit': ( 'mini_batch_training.html#fit',
                                                                                           'practice_deep_learning/training.py'),
                                                  'practice_deep_learning.training.get_dls': ( 'mini_batch_training.html#get_dls',
-                                                                                              'practice_deep_learning/training.py')}}}
+                                                                                              'practice_deep_learning/training.py'),
+                                                 'practice_deep_learning.training.report': ( 'mini_batch_training.html#report',
+                                                                                             'practice_deep_learning/training.py')}}}

practice_deep_learning/training.py

@@ -1,15 +1,23 @@
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/04_mini_batch_training.ipynb.
 
 # %% auto 0
-__all__ = ['Dataset', 'fit', 'get_dls']
+__all__ = ['accuracy', 'report', 'Dataset', 'fit', 'get_dls']
 
 # %% ../nbs/04_mini_batch_training.ipynb 1
 import torch, torch.nn as nn
 import torch.nn.functional as F
 from pathlib import Path
 import gzip, pickle, matplotlib.pyplot as plt
 
-# %% ../nbs/04_mini_batch_training.ipynb 74
+# %% ../nbs/04_mini_batch_training.ipynb 29
+def accuracy(out, yb):
+    return (out.argmax(1)==yb).float().mean()
+
+# %% ../nbs/04_mini_batch_training.ipynb 34
+def report(loss, preds, yb):
+    print(f"loss: {loss:.2f}, accuracy: {accuracy(preds, yb):.2f}")
+
+# %% ../nbs/04_mini_batch_training.ipynb 75
 class Dataset:
     def __init__(self, x, y):
         self.x, self.y = x, y
@@ -20,11 +28,11 @@ def __len__(self):
     def __getitem__(self, i):
         return self.x[i], self.y[i]
 
-# %% ../nbs/04_mini_batch_training.ipynb 108
+# %% ../nbs/04_mini_batch_training.ipynb 109
 from torch.utils.data import DataLoader, SequentialSampler, RandomSampler, BatchSampler
 
-# %% ../nbs/04_mini_batch_training.ipynb 119
-def fit(epochs, model, loss_func, opt, train_dl, valid_ld):
+# %% ../nbs/04_mini_batch_training.ipynb 120
+def fit(epochs, model, loss_func, opt, train_dl, valid_dl):
     for epoch in range(epochs):
         model.train()
         for xb, yb in train_dl:
@@ -47,7 +55,7 @@ def fit(epochs, model, loss_func, opt, train_dl, valid_ld):
 
     return total_loss/count, total_acc/count           
 
-# %% ../nbs/04_mini_batch_training.ipynb 120
+# %% ../nbs/04_mini_batch_training.ipynb 121
 def get_dls(train_ds, valid_ds, bs, **kwargs):
     return (
         DataLoader(train_ds, bs, shuffle=True, **kwargs),