diff --git "a/1\357\274\214kerasmodel_example.ipynb" "b/1\357\274\214kerasmodel_example.ipynb"
index 69693dd..659ea9b 100644
--- "a/1\357\274\214kerasmodel_example.ipynb"
+++ "b/1\357\274\214kerasmodel_example.ipynb"
@@ -21,7 +21,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 1,
"id": "3e915b9b",
"metadata": {},
"outputs": [],
@@ -32,7 +32,7 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 2,
"id": "a1920bc7",
"metadata": {},
"outputs": [],
@@ -47,14 +47,6 @@
"import torchkeras #Attention this line \n"
]
},
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "101cd9c3-5061-450e-a022-df09afed935e",
- "metadata": {},
- "outputs": [],
- "source": []
- },
{
"cell_type": "markdown",
"id": "e73ca619",
@@ -65,113 +57,10 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": null,
"id": "25ab4939",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\n",
- "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to mnist/MNIST/raw/train-images-idx3-ubyte.gz\n"
- ]
- },
- {
- "data": {
- "application/vnd.jupyter.widget-view+json": {
- "model_id": "383bed3915db4058b97803e6ce7dfb56",
- "version_major": 2,
- "version_minor": 0
- },
- "text/plain": [
- " 0%| | 0/9912422 [00:00\n",
- "\n",
- "\n"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"%matplotlib inline\n",
"%config InlineBackend.figure_format = 'svg'\n",
@@ -1047,7 +120,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": null,
"id": "6174ae8a",
"metadata": {},
"outputs": [],
@@ -1095,40 +168,10 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": null,
"id": "ad17716f",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "--------------------------------------------------------------------------\n",
- "Layer (type) Output Shape Param #\n",
- "==========================================================================\n",
- "Conv2d-1 [-1, 32, 26, 26] 320\n",
- "MaxPool2d-2 [-1, 32, 13, 13] 0\n",
- "Conv2d-3 [-1, 64, 9, 9] 51,264\n",
- "MaxPool2d-4 [-1, 64, 4, 4] 0\n",
- "Dropout2d-5 [-1, 64, 4, 4] 0\n",
- "AdaptiveMaxPool2d-6 [-1, 64, 1, 1] 0\n",
- "Flatten-7 [-1, 64] 0\n",
- "Linear-8 [-1, 32] 2,080\n",
- "ReLU-9 [-1, 32] 0\n",
- "Linear-10 [-1, 10] 330\n",
- "==========================================================================\n",
- "Total params: 53,994\n",
- "Trainable params: 53,994\n",
- "Non-trainable params: 0\n",
- "--------------------------------------------------------------------------\n",
- "Input size (MB): 0.000069\n",
- "Forward/backward pass size (MB): 0.263016\n",
- "Params size (MB): 0.205971\n",
- "Estimated Total Size (MB): 0.469055\n",
- "--------------------------------------------------------------------------\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"model = torchkeras.KerasModel(net,\n",
" loss_fn = nn.CrossEntropyLoss(),\n",
@@ -1158,135 +201,10 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": null,
"id": "54bcc553",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\u001b[0;31m<<<<<< 🐌 cpu is used >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:34:30\n",
- "Epoch 1 / 15\n",
- "\n",
- "100%|████████████████| 469/469 [00:18<00:00, 24.99it/s, lr=0.001, train_acc=0.845, train_loss=0.527]\n",
- "100%|████████████████████████████████| 79/79 [00:01<00:00, 46.42it/s, val_acc=0.954, val_loss=0.149]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9544000029563904 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:34:50\n",
- "Epoch 2 / 15\n",
- "\n",
- "100%|█████████████████| 469/469 [00:18<00:00, 25.71it/s, lr=0.001, train_acc=0.96, train_loss=0.134]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 48.45it/s, val_acc=0.973, val_loss=0.0844]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9728000164031982 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:35:10\n",
- "Epoch 3 / 15\n",
- "\n",
- "100%|████████████████| 469/469 [00:18<00:00, 25.63it/s, lr=0.001, train_acc=0.97, train_loss=0.0977]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 47.24it/s, val_acc=0.974, val_loss=0.0777]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.974399983882904 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:35:30\n",
- "Epoch 4 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.35it/s, lr=0.001, train_acc=0.975, train_loss=0.0808]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.35it/s, val_acc=0.977, val_loss=0.0656]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.977400004863739 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:35:51\n",
- "Epoch 5 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.23it/s, lr=0.001, train_acc=0.978, train_loss=0.0702]\n",
- "100%|████████████████████████████████| 79/79 [00:01<00:00, 46.67it/s, val_acc=0.98, val_loss=0.0643]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9797000288963318 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:36:11\n",
- "Epoch 6 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.24it/s, lr=0.001, train_acc=0.981, train_loss=0.0611]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 47.87it/s, val_acc=0.982, val_loss=0.0554]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9815000295639038 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:36:31\n",
- "Epoch 7 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.14it/s, lr=0.001, train_acc=0.983, train_loss=0.0528]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 48.14it/s, val_acc=0.982, val_loss=0.0566]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9815999865531921 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:36:51\n",
- "Epoch 8 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.29it/s, lr=0.001, train_acc=0.984, train_loss=0.0482]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.61it/s, val_acc=0.983, val_loss=0.0568]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9825000166893005 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:37:12\n",
- "Epoch 9 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 24.91it/s, lr=0.001, train_acc=0.986, train_loss=0.0432]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.04it/s, val_acc=0.985, val_loss=0.0444]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9850999712944031 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:37:32\n",
- "Epoch 10 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.09it/s, lr=0.001, train_acc=0.987, train_loss=0.0409]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 47.21it/s, val_acc=0.984, val_loss=0.0473]\n",
- "\n",
- "================================================================================2023-02-09 19:37:52\n",
- "Epoch 11 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.14it/s, lr=0.001, train_acc=0.988, train_loss=0.0372]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 47.82it/s, val_acc=0.987, val_loss=0.0445]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9868000149726868 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:38:13\n",
- "Epoch 12 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 24.91it/s, lr=0.001, train_acc=0.988, train_loss=0.0343]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.46it/s, val_acc=0.985, val_loss=0.0479]\n",
- "\n",
- "================================================================================2023-02-09 19:38:33\n",
- "Epoch 13 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.03it/s, lr=0.001, train_acc=0.989, train_loss=0.0335]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 47.64it/s, val_acc=0.988, val_loss=0.0418]\n",
- "\u001b[0;31m<<<<<< reach best val_acc : 0.9876000285148621 >>>>>>\u001b[0m\n",
- "\n",
- "================================================================================2023-02-09 19:38:54\n",
- "Epoch 14 / 15\n",
- "\n",
- "100%|████████████████| 469/469 [00:18<00:00, 24.98it/s, lr=0.001, train_acc=0.99, train_loss=0.0287]\n",
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.02it/s, val_acc=0.985, val_loss=0.0487]\n",
- "\n",
- "================================================================================2023-02-09 19:39:14\n",
- "Epoch 15 / 15\n",
- "\n",
- "100%|███████████████| 469/469 [00:18<00:00, 25.03it/s, lr=0.001, train_acc=0.991, train_loss=0.0273]\n",
- "100%|████████████████████████████████| 79/79 [00:01<00:00, 46.63it/s, val_acc=0.987, val_loss=0.041]\n",
- " epoch train_loss train_acc lr val_loss val_acc\n",
- "0 1 0.526868 0.845133 0.001 0.148538 0.9544\n",
- "1 2 0.134293 0.959750 0.001 0.084383 0.9728\n",
- "2 3 0.097699 0.970250 0.001 0.077715 0.9744\n",
- "3 4 0.080804 0.974733 0.001 0.065645 0.9774\n",
- "4 5 0.070243 0.977833 0.001 0.064307 0.9797\n",
- "5 6 0.061092 0.981017 0.001 0.055448 0.9815\n",
- "6 7 0.052803 0.982733 0.001 0.056596 0.9816\n",
- "7 8 0.048157 0.984267 0.001 0.056754 0.9825\n",
- "8 9 0.043176 0.986050 0.001 0.044400 0.9851\n",
- "9 10 0.040888 0.986500 0.001 0.047307 0.9843\n",
- "10 11 0.037161 0.987867 0.001 0.044459 0.9868\n",
- "11 12 0.034350 0.988250 0.001 0.047916 0.9849\n",
- "12 13 0.033464 0.989000 0.001 0.041772 0.9876\n",
- "13 14 0.028671 0.990467 0.001 0.048730 0.9850\n",
- "14 15 0.027258 0.991050 0.001 0.041001 0.9870\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"# if gpu/mps is available, will auto use it, otherwise cpu will be used.\n",
"dfhistory=model.fit(train_data=dl_train, \n",
@@ -1300,853 +218,10 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": null,
"id": "1ddc3317",
"metadata": {},
- "outputs": [
- {
- "data": {
- "image/svg+xml": [
- "\n",
- "\n",
- "\n"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"# visual the prediction\n",
"device = None\n",
@@ -2178,7 +253,7 @@
},
{
"cell_type": "code",
- "execution_count": 50,
+ "execution_count": null,
"id": "520c6a1b",
"metadata": {},
"outputs": [],
@@ -2204,2146 +279,30 @@
},
{
"cell_type": "code",
- "execution_count": 53,
+ "execution_count": null,
"id": "a179179f-65ff-434e-b317-17ca56355e7e",
"metadata": {},
- "outputs": [
- {
- "data": {
- "image/svg+xml": [
- "\n",
- "\n",
- "\n"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"plot_metric(dfhistory,'loss')"
]
},
{
"cell_type": "code",
- "execution_count": 51,
+ "execution_count": null,
"id": "960119f4",
"metadata": {},
- "outputs": [
- {
- "data": {
- "image/svg+xml": [
- "\n",
- "\n",
- "\n"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"plot_metric(dfhistory,\"acc\")"
]
},
{
"cell_type": "code",
- "execution_count": 52,
+ "execution_count": null,
"id": "62f8cf7b",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 46.50it/s, val_acc=0.988, val_loss=0.0418]\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "{'val_loss': 0.041772333724491295, 'val_acc': 0.9876000285148621}"
- ]
- },
- "execution_count": 52,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"model.evaluate(dl_val)"
]
@@ -4366,371 +325,10 @@
},
{
"cell_type": "code",
- "execution_count": 54,
+ "execution_count": null,
"id": "91b21a4e-be1c-4fba-ae99-7fe4962c67b6",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "y_pred = 0\n",
- "y_prob = 0.9997945427894592\n"
- ]
- },
- {
- "data": {
- "image/svg+xml": [
- "\n",
- "\n",
- "\n"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"net = model.net\n",
"net.eval();\n",
@@ -4758,28 +356,10 @@
},
{
"cell_type": "code",
- "execution_count": 55,
+ "execution_count": null,
"id": "45089d19",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "100%|███████████████████████████████| 79/79 [00:01<00:00, 48.51it/s, val_acc=0.988, val_loss=0.0418]\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "{'val_loss': 0.041772333724491295, 'val_acc': 0.9876000285148621}"
- ]
- },
- "execution_count": 55,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"# used the saved model parameters \n",
"net_clone = create_net() \n",
diff --git a/push2github.md b/push2github.md
index 64957d6..5bbaa8b 100644
--- a/push2github.md
+++ b/push2github.md
@@ -2,6 +2,7 @@
```python
%run setup.py sdist bdist_wheel
+
```
```python
@@ -58,7 +59,7 @@
```
```python
-!git push -f origin master
+!git push origin master
```
```python
@@ -68,3 +69,18 @@
```python
```
+
+## gitignore
+
+```python
+# %load .gitignore
+.DS_store
+.ipynb_checkpoints
+.ipynb_checkpoints/*
+torchkeras.egg-info/*
+dist/*
+build/*
+torchkeras/__pycache__/
+.idea
+
+```
diff --git a/setup.py b/setup.py
index 6f51607..95bcd08 100644
--- a/setup.py
+++ b/setup.py
@@ -1,7 +1,7 @@
# -*- coding:utf-8 -*-
import re
from pathlib import Path
-from setuptools import find_packages, setup
+from setuptools import find_packages, setup, find_namespace_packages
# Settings
FILE = Path(__file__).resolve()
@@ -32,7 +32,7 @@ def get_version():
],
long_description_content_type="text/markdown",
url="https://github.com/lyhue1991/torchkeras",
- packages=find_packages(),
+ packages=find_namespace_packages(exclude=['torchkeras.assets','data']),
include_package_data=True,
classifiers=[
"Programming Language :: Python :: 3",
diff --git a/torchkeras/data.py b/torchkeras/data.py
new file mode 100644
index 0000000..7c0628d
--- /dev/null
+++ b/torchkeras/data.py
@@ -0,0 +1,14 @@
+from pathlib import Path
+from PIL import Image
+import os
+
+path = Path(__file__)
+
+def get_example_image(img_name='park.jpg'):
+ 'name can be bus.jpg / park.jpg / zidane.jpg'
+ img_path = str(path.parent/f"assets/{img_name}")
+ assert os.path.exists(img_path), 'img_name can only be bus.jpg / park.jpg / zidane.jpg'
+ return Image.open(img_path)
+
+
+
\ No newline at end of file