2051 lines
256 KiB
Plaintext
2051 lines
256 KiB
Plaintext
{
|
||
"nbformat": 4,
|
||
"nbformat_minor": 0,
|
||
"metadata": {
|
||
"accelerator": "GPU",
|
||
"colab": {
|
||
"name": "Varying_learning_rate_on_non_scaled_data.ipynb",
|
||
"provenance": [],
|
||
"include_colab_link": true
|
||
},
|
||
"hide_input": false,
|
||
"kernelspec": {
|
||
"display_name": "Python 3",
|
||
"language": "python",
|
||
"name": "python3"
|
||
},
|
||
"language_info": {
|
||
"codemirror_mode": {
|
||
"name": "ipython",
|
||
"version": 3
|
||
},
|
||
"file_extension": ".py",
|
||
"mimetype": "text/x-python",
|
||
"name": "python",
|
||
"nbconvert_exporter": "python",
|
||
"pygments_lexer": "ipython3",
|
||
"version": "3.7.4"
|
||
},
|
||
"widgets": {
|
||
"application/vnd.jupyter.widget-state+json": {
|
||
"2d72b65a1f4c465cb8915d21a4ded81a": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_db11204073db42099a81f19a9ef35a57",
|
||
"IPY_MODEL_676ea1ac89f14b8eae9fb1b4110999eb"
|
||
],
|
||
"layout": "IPY_MODEL_4258a98ff8d8436f813aa29764089d7e"
|
||
}
|
||
},
|
||
"4258a98ff8d8436f813aa29764089d7e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"db11204073db42099a81f19a9ef35a57": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "info",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_36e6a6c0c8fd4a1284f31cb11fe5be07",
|
||
"max": 1,
|
||
"min": 0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_cf8d7d6369bb4f7a92a02f57336ec3eb",
|
||
"value": 1
|
||
}
|
||
},
|
||
"676ea1ac89f14b8eae9fb1b4110999eb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_d8e28abad66f4ece93da1f30c682e289",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_e89d5fb85edb41968d3265c027423dd1",
|
||
"value": " 26427392/? [00:20<00:00, 3846154.66it/s]"
|
||
}
|
||
},
|
||
"cf8d7d6369bb4f7a92a02f57336ec3eb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": "initial"
|
||
}
|
||
},
|
||
"36e6a6c0c8fd4a1284f31cb11fe5be07": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"e89d5fb85edb41968d3265c027423dd1": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "DescriptionStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "DescriptionStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"d8e28abad66f4ece93da1f30c682e289": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"fa55a956e0ec426684ffb0bdd2aaf65d": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_8d2e6d4ec31543e7964121dfa4fe1f9f",
|
||
"IPY_MODEL_301fccdfd2254c658701abd2d5b9368e"
|
||
],
|
||
"layout": "IPY_MODEL_51267828ab3a4313b0e6012f6bf0b328"
|
||
}
|
||
},
|
||
"51267828ab3a4313b0e6012f6bf0b328": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"8d2e6d4ec31543e7964121dfa4fe1f9f": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_88443dd117e5459fbff8729f7e0f7ac7",
|
||
"max": 1,
|
||
"min": 0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_b97cec467a56407483e60df732c5b038",
|
||
"value": 1
|
||
}
|
||
},
|
||
"301fccdfd2254c658701abd2d5b9368e": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_d3687ee113de4e6aa32b51dbae4ae439",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_b37c0ec1fefa4a628c2b2aaa0eafdc37",
|
||
"value": " 32768/? [00:03<00:00, 8840.11it/s]"
|
||
}
|
||
},
|
||
"b97cec467a56407483e60df732c5b038": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": "initial"
|
||
}
|
||
},
|
||
"88443dd117e5459fbff8729f7e0f7ac7": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"b37c0ec1fefa4a628c2b2aaa0eafdc37": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "DescriptionStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "DescriptionStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"d3687ee113de4e6aa32b51dbae4ae439": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"f0216a2a6381403aa3e4100f4ae05a0b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_f134ca2f225841f58f880fe94cda0682",
|
||
"IPY_MODEL_2dcb661d6ebc4cfb9217ee14f4dcc6fe"
|
||
],
|
||
"layout": "IPY_MODEL_111c4426dc2a451383d4235ed11de4af"
|
||
}
|
||
},
|
||
"111c4426dc2a451383d4235ed11de4af": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"f134ca2f225841f58f880fe94cda0682": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_ae440217fb4f48269a215aa1548c7bd1",
|
||
"max": 1,
|
||
"min": 0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_212e04d78edb4e3f98d52e43961118c2",
|
||
"value": 1
|
||
}
|
||
},
|
||
"2dcb661d6ebc4cfb9217ee14f4dcc6fe": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_dab4a4ef69b0414392602812bd054d4f",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_3775293154cd4d5b9433d9ddf3cc7ecb",
|
||
"value": " 4423680/? [00:02<00:00, 1569289.74it/s]"
|
||
}
|
||
},
|
||
"212e04d78edb4e3f98d52e43961118c2": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": "initial"
|
||
}
|
||
},
|
||
"ae440217fb4f48269a215aa1548c7bd1": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"3775293154cd4d5b9433d9ddf3cc7ecb": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "DescriptionStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "DescriptionStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"dab4a4ef69b0414392602812bd054d4f": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"0e18c3f0a671480aae5b419040f68b69": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HBoxModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HBoxModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HBoxView",
|
||
"box_style": "",
|
||
"children": [
|
||
"IPY_MODEL_ad36ad050558442998f0e00eb89d262b",
|
||
"IPY_MODEL_58cc93d98ec546f1ba7b7414fef8e1cd"
|
||
],
|
||
"layout": "IPY_MODEL_a5d70c71fbb6452ab21e283210129f6e"
|
||
}
|
||
},
|
||
"a5d70c71fbb6452ab21e283210129f6e": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"ad36ad050558442998f0e00eb89d262b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "FloatProgressModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "FloatProgressModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "ProgressView",
|
||
"bar_style": "success",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_91b0bf8c64a14494b8e23049fa4a9ee7",
|
||
"max": 1,
|
||
"min": 0,
|
||
"orientation": "horizontal",
|
||
"style": "IPY_MODEL_d8ac9b97251c418397239b49db8a2c9b",
|
||
"value": 1
|
||
}
|
||
},
|
||
"58cc93d98ec546f1ba7b7414fef8e1cd": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "HTMLModel",
|
||
"state": {
|
||
"_dom_classes": [],
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "HTMLModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/controls",
|
||
"_view_module_version": "1.5.0",
|
||
"_view_name": "HTMLView",
|
||
"description": "",
|
||
"description_tooltip": null,
|
||
"layout": "IPY_MODEL_fc18640e80a34c54a00e9fb8feba2b21",
|
||
"placeholder": "",
|
||
"style": "IPY_MODEL_941b82bd3e7e46ab958b66a6ab8e431f",
|
||
"value": " 8192/? [00:00<00:00, 9504.00it/s]"
|
||
}
|
||
},
|
||
"d8ac9b97251c418397239b49db8a2c9b": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "ProgressStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "ProgressStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"bar_color": null,
|
||
"description_width": "initial"
|
||
}
|
||
},
|
||
"91b0bf8c64a14494b8e23049fa4a9ee7": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
},
|
||
"941b82bd3e7e46ab958b66a6ab8e431f": {
|
||
"model_module": "@jupyter-widgets/controls",
|
||
"model_name": "DescriptionStyleModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/controls",
|
||
"_model_module_version": "1.5.0",
|
||
"_model_name": "DescriptionStyleModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "StyleView",
|
||
"description_width": ""
|
||
}
|
||
},
|
||
"fc18640e80a34c54a00e9fb8feba2b21": {
|
||
"model_module": "@jupyter-widgets/base",
|
||
"model_name": "LayoutModel",
|
||
"state": {
|
||
"_model_module": "@jupyter-widgets/base",
|
||
"_model_module_version": "1.2.0",
|
||
"_model_name": "LayoutModel",
|
||
"_view_count": null,
|
||
"_view_module": "@jupyter-widgets/base",
|
||
"_view_module_version": "1.2.0",
|
||
"_view_name": "LayoutView",
|
||
"align_content": null,
|
||
"align_items": null,
|
||
"align_self": null,
|
||
"border": null,
|
||
"bottom": null,
|
||
"display": null,
|
||
"flex": null,
|
||
"flex_flow": null,
|
||
"grid_area": null,
|
||
"grid_auto_columns": null,
|
||
"grid_auto_flow": null,
|
||
"grid_auto_rows": null,
|
||
"grid_column": null,
|
||
"grid_gap": null,
|
||
"grid_row": null,
|
||
"grid_template_areas": null,
|
||
"grid_template_columns": null,
|
||
"grid_template_rows": null,
|
||
"height": null,
|
||
"justify_content": null,
|
||
"justify_items": null,
|
||
"left": null,
|
||
"margin": null,
|
||
"max_height": null,
|
||
"max_width": null,
|
||
"min_height": null,
|
||
"min_width": null,
|
||
"object_fit": null,
|
||
"object_position": null,
|
||
"order": null,
|
||
"overflow": null,
|
||
"overflow_x": null,
|
||
"overflow_y": null,
|
||
"padding": null,
|
||
"right": null,
|
||
"top": null,
|
||
"visibility": null,
|
||
"width": null
|
||
}
|
||
}
|
||
}
|
||
}
|
||
},
|
||
"cells": [
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {
|
||
"id": "view-in-github",
|
||
"colab_type": "text"
|
||
},
|
||
"source": [
|
||
"<a href=\"https://colab.research.google.com/github/PacktPublishing/Hands-On-Computer-Vision-with-PyTorch/blob/master/Chapter03/Varying_learning_rate_on_non_scaled_data.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "-BdJ4Kt46Qsq",
|
||
"outputId": "00ce7b7f-06a9-4d4e-e00c-15b35851d56e",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 437,
|
||
"referenced_widgets": [
|
||
"2d72b65a1f4c465cb8915d21a4ded81a",
|
||
"4258a98ff8d8436f813aa29764089d7e",
|
||
"db11204073db42099a81f19a9ef35a57",
|
||
"676ea1ac89f14b8eae9fb1b4110999eb",
|
||
"cf8d7d6369bb4f7a92a02f57336ec3eb",
|
||
"36e6a6c0c8fd4a1284f31cb11fe5be07",
|
||
"e89d5fb85edb41968d3265c027423dd1",
|
||
"d8e28abad66f4ece93da1f30c682e289",
|
||
"fa55a956e0ec426684ffb0bdd2aaf65d",
|
||
"51267828ab3a4313b0e6012f6bf0b328",
|
||
"8d2e6d4ec31543e7964121dfa4fe1f9f",
|
||
"301fccdfd2254c658701abd2d5b9368e",
|
||
"b97cec467a56407483e60df732c5b038",
|
||
"88443dd117e5459fbff8729f7e0f7ac7",
|
||
"b37c0ec1fefa4a628c2b2aaa0eafdc37",
|
||
"d3687ee113de4e6aa32b51dbae4ae439",
|
||
"f0216a2a6381403aa3e4100f4ae05a0b",
|
||
"111c4426dc2a451383d4235ed11de4af",
|
||
"f134ca2f225841f58f880fe94cda0682",
|
||
"2dcb661d6ebc4cfb9217ee14f4dcc6fe",
|
||
"212e04d78edb4e3f98d52e43961118c2",
|
||
"ae440217fb4f48269a215aa1548c7bd1",
|
||
"3775293154cd4d5b9433d9ddf3cc7ecb",
|
||
"dab4a4ef69b0414392602812bd054d4f",
|
||
"0e18c3f0a671480aae5b419040f68b69",
|
||
"a5d70c71fbb6452ab21e283210129f6e",
|
||
"ad36ad050558442998f0e00eb89d262b",
|
||
"58cc93d98ec546f1ba7b7414fef8e1cd",
|
||
"d8ac9b97251c418397239b49db8a2c9b",
|
||
"91b0bf8c64a14494b8e23049fa4a9ee7",
|
||
"941b82bd3e7e46ab958b66a6ab8e431f",
|
||
"fc18640e80a34c54a00e9fb8feba2b21"
|
||
]
|
||
}
|
||
},
|
||
"source": [
|
||
"from torchvision import datasets\n",
|
||
"import torch\n",
|
||
"data_folder = '~/data/FMNIST' # This can be any directory you want to \n",
|
||
"# download FMNIST to\n",
|
||
"fmnist = datasets.FashionMNIST(data_folder, download=True, train=True)\n",
|
||
"tr_images = fmnist.data\n",
|
||
"tr_targets = fmnist.targets"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw/train-images-idx3-ubyte.gz\n"
|
||
],
|
||
"name": "stdout"
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "2d72b65a1f4c465cb8915d21a4ded81a",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": []
|
||
}
|
||
},
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Extracting /root/data/FMNIST/FashionMNIST/raw/train-images-idx3-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw\n",
|
||
"Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw/train-labels-idx1-ubyte.gz\n"
|
||
],
|
||
"name": "stdout"
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "fa55a956e0ec426684ffb0bdd2aaf65d",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": []
|
||
}
|
||
},
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Extracting /root/data/FMNIST/FashionMNIST/raw/train-labels-idx1-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw\n",
|
||
"Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw/t10k-images-idx3-ubyte.gz\n"
|
||
],
|
||
"name": "stdout"
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "f0216a2a6381403aa3e4100f4ae05a0b",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": []
|
||
}
|
||
},
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Extracting /root/data/FMNIST/FashionMNIST/raw/t10k-images-idx3-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw\n",
|
||
"Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz\n",
|
||
"\n",
|
||
"\n"
|
||
],
|
||
"name": "stdout"
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"application/vnd.jupyter.widget-view+json": {
|
||
"model_id": "0e18c3f0a671480aae5b419040f68b69",
|
||
"version_major": 2,
|
||
"version_minor": 0
|
||
},
|
||
"text/plain": [
|
||
"HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": []
|
||
}
|
||
},
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Extracting /root/data/FMNIST/FashionMNIST/raw/t10k-labels-idx1-ubyte.gz to /root/data/FMNIST/FashionMNIST/raw\n",
|
||
"Processing...\n",
|
||
"Done!\n",
|
||
"\n"
|
||
],
|
||
"name": "stdout"
|
||
},
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/usr/local/lib/python3.6/dist-packages/torchvision/datasets/mnist.py:469: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:141.)\n",
|
||
" return torch.from_numpy(parsed.astype(m[2], copy=False)).view(*s)\n"
|
||
],
|
||
"name": "stderr"
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "S4Ss3qAj6cCN"
|
||
},
|
||
"source": [
|
||
"val_fmnist = datasets.FashionMNIST(data_folder, download=True, train=False)\n",
|
||
"val_images = val_fmnist.data\n",
|
||
"val_targets = val_fmnist.targets"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "MhUgyxQv6dWF"
|
||
},
|
||
"source": [
|
||
"import matplotlib.pyplot as plt\n",
|
||
"%matplotlib inline\n",
|
||
"import numpy as np\n",
|
||
"from torch.utils.data import Dataset, DataLoader\n",
|
||
"import torch\n",
|
||
"import torch.nn as nn\n",
|
||
"device = 'cuda' if torch.cuda.is_available() else 'cpu'"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {
|
||
"id": "XxBVyrwixgx-"
|
||
},
|
||
"source": [
|
||
"### High Learning Rate"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "wHgNxifc6edk"
|
||
},
|
||
"source": [
|
||
"class FMNISTDataset(Dataset):\n",
|
||
" def __init__(self, x, y):\n",
|
||
" x = x.float()\n",
|
||
" x = x.view(-1,28*28)\n",
|
||
" self.x, self.y = x, y \n",
|
||
" def __getitem__(self, ix):\n",
|
||
" x, y = self.x[ix], self.y[ix] \n",
|
||
" return x.to(device), y.to(device)\n",
|
||
" def __len__(self): \n",
|
||
" return len(self.x)\n",
|
||
"\n",
|
||
"from torch.optim import SGD, Adam\n",
|
||
"def get_model():\n",
|
||
" model = nn.Sequential(\n",
|
||
" nn.Linear(28 * 28, 1000),\n",
|
||
" nn.ReLU(),\n",
|
||
" nn.Linear(1000, 10)\n",
|
||
" ).to(device)\n",
|
||
"\n",
|
||
" loss_fn = nn.CrossEntropyLoss()\n",
|
||
" optimizer = Adam(model.parameters(), lr=1e-1)\n",
|
||
" return model, loss_fn, optimizer\n",
|
||
"\n",
|
||
"def train_batch(x, y, model, opt, loss_fn):\n",
|
||
" model.train()\n",
|
||
" prediction = model(x)\n",
|
||
" batch_loss = loss_fn(prediction, y)\n",
|
||
" batch_loss.backward()\n",
|
||
" optimizer.step()\n",
|
||
" optimizer.zero_grad()\n",
|
||
" return batch_loss.item()\n",
|
||
"\n",
|
||
"def accuracy(x, y, model):\n",
|
||
" model.eval()\n",
|
||
" # this is the same as @torch.no_grad \n",
|
||
" # at the top of function, only difference\n",
|
||
" # being, grad is not computed in the with scope\n",
|
||
" with torch.no_grad():\n",
|
||
" prediction = model(x)\n",
|
||
" max_values, argmaxes = prediction.max(-1)\n",
|
||
" is_correct = argmaxes == y\n",
|
||
" return is_correct.cpu().numpy().tolist()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "CfnVtUMO6nhR"
|
||
},
|
||
"source": [
|
||
"def get_data(): \n",
|
||
" train = FMNISTDataset(tr_images, tr_targets) \n",
|
||
" trn_dl = DataLoader(train, batch_size=32, shuffle=True)\n",
|
||
" val = FMNISTDataset(val_images, val_targets) \n",
|
||
" val_dl = DataLoader(val, batch_size=len(val_images), shuffle=False)\n",
|
||
" return trn_dl, val_dl"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "wAN-GtKb6o83"
|
||
},
|
||
"source": [
|
||
"@torch.no_grad()\n",
|
||
"def val_loss(x, y, model):\n",
|
||
" prediction = model(x)\n",
|
||
" val_loss = loss_fn(prediction, y)\n",
|
||
" return val_loss.item()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "7EhlA61S6qM3"
|
||
},
|
||
"source": [
|
||
"trn_dl, val_dl = get_data()\n",
|
||
"model, loss_fn, optimizer = get_model()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "h-yph5GO6rQ6",
|
||
"outputId": "18df4a22-014b-4925-f827-be715d7c8762",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 108
|
||
}
|
||
},
|
||
"source": [
|
||
"train_losses, train_accuracies = [], []\n",
|
||
"val_losses, val_accuracies = [], []\n",
|
||
"for epoch in range(5):\n",
|
||
" print(epoch)\n",
|
||
" train_epoch_losses, train_epoch_accuracies = [], []\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" batch_loss = train_batch(x, y, model, optimizer, loss_fn)\n",
|
||
" train_epoch_losses.append(batch_loss) \n",
|
||
" train_epoch_loss = np.array(train_epoch_losses).mean()\n",
|
||
"\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" is_correct = accuracy(x, y, model)\n",
|
||
" train_epoch_accuracies.extend(is_correct)\n",
|
||
" train_epoch_accuracy = np.mean(train_epoch_accuracies)\n",
|
||
" for ix, batch in enumerate(iter(val_dl)):\n",
|
||
" x, y = batch\n",
|
||
" val_is_correct = accuracy(x, y, model)\n",
|
||
" validation_loss = val_loss(x, y, model)\n",
|
||
" val_epoch_accuracy = np.mean(val_is_correct)\n",
|
||
" train_losses.append(train_epoch_loss)\n",
|
||
" train_accuracies.append(train_epoch_accuracy)\n",
|
||
" val_losses.append(validation_loss)\n",
|
||
" val_accuracies.append(val_epoch_accuracy)"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"0\n",
|
||
"1\n",
|
||
"2\n",
|
||
"3\n",
|
||
"4\n"
|
||
],
|
||
"name": "stdout"
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "HtZsoP8w6sNY",
|
||
"outputId": "62f7dfff-924c-4259-e2fd-c4114d0ddb60",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 337
|
||
}
|
||
},
|
||
"source": [
|
||
"epochs = np.arange(5)+1\n",
|
||
"import matplotlib.ticker as mtick\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"import matplotlib.ticker as mticker\n",
|
||
"%matplotlib inline\n",
|
||
"plt.subplot(211)\n",
|
||
"plt.plot(epochs, train_losses, 'bo', label='Training loss')\n",
|
||
"plt.plot(epochs, val_losses, 'r', label='Validation loss')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation loss with 0.1 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Loss')\n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()\n",
|
||
"plt.subplot(212)\n",
|
||
"plt.plot(epochs, train_accuracies, 'bo', label='Training accuracy')\n",
|
||
"plt.plot(epochs, val_accuracies, 'r', label='Validation accuracy')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation accuracy with 0.1 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Accuracy')\n",
|
||
"plt.gca().set_yticklabels(['{:.0f}%'.format(x*100) for x in plt.gca().get_yticks()]) \n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "Oe7U2lsn8Cub",
|
||
"outputId": "55ac716f-5d75-4afd-af9f-ba9194caaf56",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 1000
|
||
}
|
||
},
|
||
"source": [
|
||
"for ix, par in enumerate(model.parameters()):\n",
|
||
" if(ix==0):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting input to hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==1):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix==2):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting hidden to output layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==3):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of output layer')\n",
|
||
" plt.show() "
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "EFAUGptW7By6"
|
||
},
|
||
"source": [
|
||
""
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {
|
||
"id": "qFosnD377CVI"
|
||
},
|
||
"source": [
|
||
"### Medium learning rate"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "qixgqA-x7Dzv"
|
||
},
|
||
"source": [
|
||
"def get_model():\n",
|
||
" model = nn.Sequential(\n",
|
||
" nn.Linear(28 * 28, 1000),\n",
|
||
" nn.ReLU(),\n",
|
||
" nn.Linear(1000, 10)\n",
|
||
" ).to(device)\n",
|
||
"\n",
|
||
" loss_fn = nn.CrossEntropyLoss()\n",
|
||
" optimizer = Adam(model.parameters(), lr=1e-3)\n",
|
||
" return model, loss_fn, optimizer"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "7RJpjJAc7G1Y"
|
||
},
|
||
"source": [
|
||
"trn_dl, val_dl = get_data()\n",
|
||
"model, loss_fn, optimizer = get_model()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "AXEL4biQ7KGE",
|
||
"outputId": "2adc56a3-c749-4de2-f8d5-16b96bc08a50",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 108
|
||
}
|
||
},
|
||
"source": [
|
||
"train_losses, train_accuracies = [], []\n",
|
||
"val_losses, val_accuracies = [], []\n",
|
||
"for epoch in range(5):\n",
|
||
" print(epoch)\n",
|
||
" train_epoch_losses, train_epoch_accuracies = [], []\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" batch_loss = train_batch(x, y, model, optimizer, loss_fn)\n",
|
||
" train_epoch_losses.append(batch_loss) \n",
|
||
" train_epoch_loss = np.array(train_epoch_losses).mean()\n",
|
||
"\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" is_correct = accuracy(x, y, model)\n",
|
||
" train_epoch_accuracies.extend(is_correct)\n",
|
||
" train_epoch_accuracy = np.mean(train_epoch_accuracies)\n",
|
||
" for ix, batch in enumerate(iter(val_dl)):\n",
|
||
" x, y = batch\n",
|
||
" val_is_correct = accuracy(x, y, model)\n",
|
||
" validation_loss = val_loss(x, y, model)\n",
|
||
" val_epoch_accuracy = np.mean(val_is_correct)\n",
|
||
" train_losses.append(train_epoch_loss)\n",
|
||
" train_accuracies.append(train_epoch_accuracy)\n",
|
||
" val_losses.append(validation_loss)\n",
|
||
" val_accuracies.append(val_epoch_accuracy)"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"0\n",
|
||
"1\n",
|
||
"2\n",
|
||
"3\n",
|
||
"4\n"
|
||
],
|
||
"name": "stdout"
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "lg-Po1mj7MD6",
|
||
"outputId": "b8cbcd71-344a-4571-948c-110823586c47",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 337
|
||
}
|
||
},
|
||
"source": [
|
||
"epochs = np.arange(5)+1\n",
|
||
"import matplotlib.ticker as mtick\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"import matplotlib.ticker as mticker\n",
|
||
"%matplotlib inline\n",
|
||
"plt.subplot(211)\n",
|
||
"plt.plot(epochs, train_losses, 'bo', label='Training loss')\n",
|
||
"plt.plot(epochs, val_losses, 'r', label='Validation loss')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation loss with 0.001 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Loss')\n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()\n",
|
||
"plt.subplot(212)\n",
|
||
"plt.plot(epochs, train_accuracies, 'bo', label='Training accuracy')\n",
|
||
"plt.plot(epochs, val_accuracies, 'r', label='Validation accuracy')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation accuracy with 0.001 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Accuracy')\n",
|
||
"plt.gca().set_yticklabels(['{:.0f}%'.format(x*100) for x in plt.gca().get_yticks()]) \n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "CgmXa6F67d-C",
|
||
"outputId": "2a263e17-bb10-403c-accd-f0b85c265272",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 1000
|
||
}
|
||
},
|
||
"source": [
|
||
"for ix, par in enumerate(model.parameters()):\n",
|
||
" if(ix==0):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting input to hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==1):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix==2):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting hidden to output layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==3):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of output layer')\n",
|
||
" plt.show() "
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {
|
||
"id": "L7emwfEi7eo_"
|
||
},
|
||
"source": [
|
||
"### Low learning rate"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "XDRXpRl87f2p"
|
||
},
|
||
"source": [
|
||
"def get_model():\n",
|
||
" model = nn.Sequential(\n",
|
||
" nn.Linear(28 * 28, 1000),\n",
|
||
" nn.ReLU(),\n",
|
||
" nn.Linear(1000, 10)\n",
|
||
" ).to(device)\n",
|
||
"\n",
|
||
" loss_fn = nn.CrossEntropyLoss()\n",
|
||
" optimizer = Adam(model.parameters(), lr=1e-5)\n",
|
||
" return model, loss_fn, optimizer "
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "qZojB4U47iuN"
|
||
},
|
||
"source": [
|
||
"trn_dl, val_dl = get_data()\n",
|
||
"model, loss_fn, optimizer = get_model()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "uJ12zMlZ7kt7",
|
||
"outputId": "5623307d-6d2e-436d-96eb-b20d3556004c",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 108
|
||
}
|
||
},
|
||
"source": [
|
||
"train_losses, train_accuracies = [], []\n",
|
||
"val_losses, val_accuracies = [], []\n",
|
||
"for epoch in range(5):\n",
|
||
" print(epoch)\n",
|
||
" train_epoch_losses, train_epoch_accuracies = [], []\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" batch_loss = train_batch(x, y, model, optimizer, loss_fn)\n",
|
||
" train_epoch_losses.append(batch_loss) \n",
|
||
" train_epoch_loss = np.array(train_epoch_losses).mean()\n",
|
||
"\n",
|
||
" for ix, batch in enumerate(iter(trn_dl)):\n",
|
||
" x, y = batch\n",
|
||
" is_correct = accuracy(x, y, model)\n",
|
||
" train_epoch_accuracies.extend(is_correct)\n",
|
||
" train_epoch_accuracy = np.mean(train_epoch_accuracies)\n",
|
||
" for ix, batch in enumerate(iter(val_dl)):\n",
|
||
" x, y = batch\n",
|
||
" val_is_correct = accuracy(x, y, model)\n",
|
||
" validation_loss = val_loss(x, y, model)\n",
|
||
" val_epoch_accuracy = np.mean(val_is_correct)\n",
|
||
" train_losses.append(train_epoch_loss)\n",
|
||
" train_accuracies.append(train_epoch_accuracy)\n",
|
||
" val_losses.append(validation_loss)\n",
|
||
" val_accuracies.append(val_epoch_accuracy)"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "stream",
|
||
"text": [
|
||
"0\n",
|
||
"1\n",
|
||
"2\n",
|
||
"3\n",
|
||
"4\n"
|
||
],
|
||
"name": "stdout"
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "-3bQsBA37mn_",
|
||
"outputId": "8f24486d-2467-4edd-cae8-41aed81e5c82",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 337
|
||
}
|
||
},
|
||
"source": [
|
||
"epochs = np.arange(5)+1\n",
|
||
"import matplotlib.ticker as mtick\n",
|
||
"import matplotlib.pyplot as plt\n",
|
||
"import matplotlib.ticker as mticker\n",
|
||
"%matplotlib inline\n",
|
||
"plt.subplot(211)\n",
|
||
"plt.plot(epochs, train_losses, 'bo', label='Training loss')\n",
|
||
"plt.plot(epochs, val_losses, 'r', label='Validation loss')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation loss with 0.00001 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Loss')\n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()\n",
|
||
"plt.subplot(212)\n",
|
||
"plt.plot(epochs, train_accuracies, 'bo', label='Training accuracy')\n",
|
||
"plt.plot(epochs, val_accuracies, 'r', label='Validation accuracy')\n",
|
||
"plt.gca().xaxis.set_major_locator(mticker.MultipleLocator(1))\n",
|
||
"plt.title('Training and validation accuracy with 0.00001 learning rate')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Accuracy')\n",
|
||
"plt.gca().set_yticklabels(['{:.0f}%'.format(x*100) for x in plt.gca().get_yticks()]) \n",
|
||
"plt.legend()\n",
|
||
"plt.grid('off')\n",
|
||
"plt.show()"
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "r2Gnn6Gq70BN",
|
||
"outputId": "1792e99d-f09f-44a2-89ea-e7228a3c9d52",
|
||
"colab": {
|
||
"base_uri": "https://localhost:8080/",
|
||
"height": 1000
|
||
}
|
||
},
|
||
"source": [
|
||
"for ix, par in enumerate(model.parameters()):\n",
|
||
" if(ix==0):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting input to hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==1):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of hidden layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix==2):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of weights conencting hidden to output layer')\n",
|
||
" plt.show()\n",
|
||
" elif(ix ==3):\n",
|
||
" plt.hist(par.cpu().detach().numpy().flatten())\n",
|
||
" plt.title('Distribution of biases of output layer')\n",
|
||
" plt.show() "
|
||
],
|
||
"execution_count": null,
|
||
"outputs": [
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
},
|
||
{
|
||
"output_type": "display_data",
|
||
"data": {
|
||
"image/png": "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\n",
|
||
"text/plain": [
|
||
"<Figure size 432x288 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {
|
||
"tags": [],
|
||
"needs_background": "light"
|
||
}
|
||
}
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"metadata": {
|
||
"id": "1COLvsAL77V9"
|
||
},
|
||
"source": [
|
||
""
|
||
],
|
||
"execution_count": null,
|
||
"outputs": []
|
||
}
|
||
]
|
||
} |