# DeepStorm: Deep Learning Framework
## Summary:
Deep Learning Framework from scratch, with an API of a combination of pytorch and keras APIs, only uses numpy for tensor operations.
## Pip install:
```sh
pip install DeepStorm
```
## Layers & DL classes in framework:
- Conv2d
- MaxPool2d
- BatchNorm2d
- Flatten
- Dropout
- Linear
- ReLU
- Softmax
- SgdWithMomentum
- Adam
- CrossEntropyLoss
- Xavier
- He
## Model building:
```py
layers = [
Conv2d(in_channels=1, out_channels=32,
kernel_size=3, stride=1, padding='same'),
BatchNorm2d(32),
Dropout(probability=0.3),
ReLU(),
Conv2d(in_channels=32, out_channels=64,
kernel_size=3, stride=1, padding='same'),
BatchNorm2d(64),
ReLU(),
MaxPool2d(kernel_size=2, stride=2),
Conv2d(in_channels=64, out_channels=64,
kernel_size=3, stride=1, padding='same'),
BatchNorm2d(64),
ReLU(),
MaxPool2d(kernel_size=2, stride=2),
Flatten(),
Linear(in_features=64*7*7, out_features=128),
ReLU(),
Linear(128, 64),
ReLU(),
Linear(64, 10),
SoftMax(),
]
model = Model(layers)
```
Or
```py
model = Model()
model.append_layer(Conv2d(in_channels=1, out_channels=32,
kernel_size=3, stride=1, padding='same'))
model.append_layer(BatchNorm2d(32))
model.append_layer(ReLU())
model.append_layer(Conv2d(in_channels=32, out_channels=64,
kernel_size=3, stride=1, padding='same'))
model.append_layer(BatchNorm2d(64))
model.append_layer(ReLU())
model.append_layer(MaxPool2d(kernel_size=2, stride=2))
model.append_layer(Conv2d(in_channels=64, out_channels=64,
kernel_size=3, stride=1, padding='same'))
model.append_layer(BatchNorm2d(64))
model.append_layer(ReLU())
model.append_layer(MaxPool2d(kernel_size=2, stride=2))
model.append_layer(Flatten())
model.append_layer(Linear(in_features=64*7*7, out_features=128))
model.append_layer(ReLU())
model.append_layer(Linear(in_features=128, out_features=64))
model.append_layer(ReLU())
model.append_layer(Linear(in_features=64, out_features=10))
model.append_layer(SoftMax())
```
## Model compile:
```py
batch_size = 16
model.compile(optimizer=Adam(learning_rate=5e-3, mu=0.98, rho=0.999), loss=CrossEntropyLoss(),
batch_size=batch_size, metrics=['accuracy'])
```
## Model training:
```py
epochs = 25
history = model.fit(x_train=train_images, y_train=train_labels, x_val=val_images, y_val=val_labels, epochs=epochs)
```
## Model performance:
```py
plt.plot(history['accuracy'])
plt.plot(history['val_accuracy'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'val'], loc='upper left')
plt.show()
```
Raw data
{
"_id": null,
"home_page": "https://github.com/HassanRady/DeepStorm",
"name": "DeepStorm",
"maintainer": "",
"docs_url": null,
"requires_python": "==3.9.16",
"maintainer_email": "",
"keywords": "Tweets",
"author": "Hassan Rady",
"author_email": "hassan.khaled.rady@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/b8/a5/4bc2f186d7355fc18c9c51c432bd90378f6ee8c29c9cce8eb626e6a3d797/DeepStorm-1.1.0.tar.gz",
"platform": null,
"description": "# DeepStorm: Deep Learning Framework\n\n## Summary:\nDeep Learning Framework from scratch, with an API of a combination of pytorch and keras APIs, only uses numpy for tensor operations.\n\n## Pip install:\n```sh\npip install DeepStorm\n```\n\n## Layers & DL classes in framework:\n- Conv2d\n- MaxPool2d\n- BatchNorm2d\n- Flatten\n- Dropout\n- Linear\n- ReLU\n- Softmax\n- SgdWithMomentum\n- Adam\n- CrossEntropyLoss\n- Xavier\n- He\n\n## Model building:\n```py\nlayers = [\n Conv2d(in_channels=1, out_channels=32,\n kernel_size=3, stride=1, padding='same'),\n BatchNorm2d(32),\n Dropout(probability=0.3),\n ReLU(),\n\n Conv2d(in_channels=32, out_channels=64,\n kernel_size=3, stride=1, padding='same'),\n BatchNorm2d(64),\n ReLU(),\n MaxPool2d(kernel_size=2, stride=2),\n\n Conv2d(in_channels=64, out_channels=64,\n kernel_size=3, stride=1, padding='same'),\n BatchNorm2d(64),\n ReLU(),\n MaxPool2d(kernel_size=2, stride=2),\n\n Flatten(),\n\n Linear(in_features=64*7*7, out_features=128),\n ReLU(),\n Linear(128, 64),\n ReLU(),\n Linear(64, 10),\n SoftMax(),\n]\n\nmodel = Model(layers)\n```\n\nOr\n\n```py\nmodel = Model()\n\nmodel.append_layer(Conv2d(in_channels=1, out_channels=32,\n kernel_size=3, stride=1, padding='same'))\nmodel.append_layer(BatchNorm2d(32))\nmodel.append_layer(ReLU())\nmodel.append_layer(Conv2d(in_channels=32, out_channels=64,\n kernel_size=3, stride=1, padding='same'))\nmodel.append_layer(BatchNorm2d(64))\nmodel.append_layer(ReLU())\nmodel.append_layer(MaxPool2d(kernel_size=2, stride=2))\n\nmodel.append_layer(Conv2d(in_channels=64, out_channels=64,\n kernel_size=3, stride=1, padding='same'))\nmodel.append_layer(BatchNorm2d(64))\nmodel.append_layer(ReLU())\nmodel.append_layer(MaxPool2d(kernel_size=2, stride=2))\nmodel.append_layer(Flatten())\nmodel.append_layer(Linear(in_features=64*7*7, out_features=128))\nmodel.append_layer(ReLU())\nmodel.append_layer(Linear(in_features=128, out_features=64))\nmodel.append_layer(ReLU())\nmodel.append_layer(Linear(in_features=64, out_features=10))\nmodel.append_layer(SoftMax())\n```\n\n## Model compile:\n\n\n```py\nbatch_size = 16\nmodel.compile(optimizer=Adam(learning_rate=5e-3, mu=0.98, rho=0.999), loss=CrossEntropyLoss(),\n batch_size=batch_size, metrics=['accuracy'])\n```\n\n## Model training:\n```py\nepochs = 25\nhistory = model.fit(x_train=train_images, y_train=train_labels, x_val=val_images, y_val=val_labels, epochs=epochs)\n```\n\n## Model performance:\n```py\nplt.plot(history['accuracy'])\nplt.plot(history['val_accuracy'])\nplt.title('model accuracy')\nplt.ylabel('accuracy')\nplt.xlabel('epoch')\nplt.legend(['train', 'val'], loc='upper left')\nplt.show()\n```\n",
"bugtrack_url": null,
"license": "MIT license",
"summary": "Deep Learning framework from scratch",
"version": "1.1.0",
"project_urls": {
"Homepage": "https://github.com/HassanRady/DeepStorm"
},
"split_keywords": [
"tweets"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "0398a614ce9c12d16945c0bbeb02fa1b77a297d2cbb4a5b48f6a4f05d0ef81d7",
"md5": "94be8371332b217007b66f28394ec327",
"sha256": "8d57c48775e9cac298632ca1c301e8a047f322beaef1855f93d621e6c64bf228"
},
"downloads": -1,
"filename": "DeepStorm-1.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "94be8371332b217007b66f28394ec327",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "==3.9.16",
"size": 16352,
"upload_time": "2023-08-17T07:29:56",
"upload_time_iso_8601": "2023-08-17T07:29:56.879813Z",
"url": "https://files.pythonhosted.org/packages/03/98/a614ce9c12d16945c0bbeb02fa1b77a297d2cbb4a5b48f6a4f05d0ef81d7/DeepStorm-1.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "b8a54bc2f186d7355fc18c9c51c432bd90378f6ee8c29c9cce8eb626e6a3d797",
"md5": "a3ac23acd1e2515094c006ab756186f9",
"sha256": "deef0d645c11e89b6cb14e6c3840d87c91e5ca2c229400322a448d1932455243"
},
"downloads": -1,
"filename": "DeepStorm-1.1.0.tar.gz",
"has_sig": false,
"md5_digest": "a3ac23acd1e2515094c006ab756186f9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "==3.9.16",
"size": 26527,
"upload_time": "2023-08-17T07:29:58",
"upload_time_iso_8601": "2023-08-17T07:29:58.522053Z",
"url": "https://files.pythonhosted.org/packages/b8/a5/4bc2f186d7355fc18c9c51c432bd90378f6ee8c29c9cce8eb626e6a3d797/DeepStorm-1.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-08-17 07:29:58",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "HassanRady",
"github_project": "DeepStorm",
"github_not_found": true,
"lcname": "deepstorm"
}
JSON
