201 lines
67 KiB
Plaintext
201 lines
67 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import os\n",
|
|
"import sys\n",
|
|
"import matplotlib\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"import numpy as np\n",
|
|
"%matplotlib inline\n",
|
|
"plt.style.use('ggplot')\n",
|
|
"experiment_dir = 'path/to/mlpractical_directory' #Replace this with your path to the mlpractical directory"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 15,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def collect_experiment_dicts(target_dir, test_flag=False):\n",
|
|
" experiment_dicts = dict()\n",
|
|
" for subdir, dir, files in os.walk(target_dir):\n",
|
|
" for file in files:\n",
|
|
" filepath = None\n",
|
|
" if not test_flag:\n",
|
|
" if file == 'summary.csv':\n",
|
|
" filepath = os.path.join(subdir, file)\n",
|
|
" \n",
|
|
" elif test_flag:\n",
|
|
" if file == 'test_summary.csv':\n",
|
|
" filepath = os.path.join(subdir, file)\n",
|
|
" \n",
|
|
" if filepath is not None:\n",
|
|
" \n",
|
|
" with open(filepath, 'r') as read_file:\n",
|
|
" lines = read_file.readlines()\n",
|
|
" \n",
|
|
" current_experiment_dict = {key: [] for key in lines[0].replace('\\n', '').split(',')}\n",
|
|
" idx_to_key = {idx: key for idx, key in enumerate(lines[0].replace('\\n', '').split(','))}\n",
|
|
" \n",
|
|
" for line in lines[1:]:\n",
|
|
" for idx, value in enumerate(line.replace('\\n', '').split(',')):\n",
|
|
" current_experiment_dict[idx_to_key[idx]].append(float(value))\n",
|
|
" \n",
|
|
" experiment_dicts[subdir.split('/')[-2]] = current_experiment_dict\n",
|
|
" \n",
|
|
" return experiment_dicts\n",
|
|
" \n",
|
|
" "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 22,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"VGG_08 ['train_acc', 'train_loss', 'val_acc', 'val_loss']\n",
|
|
"VGG_38 ['train_acc', 'train_loss', 'val_acc', 'val_loss']\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"result_dict = collect_experiment_dicts(target_dir=experiment_dir)\n",
|
|
"for key, value in result_dict.items():\n",
|
|
" print(key, list(value.keys()))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 23,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import matplotlib.pyplot as plt\n",
|
|
"%matplotlib inline\n",
|
|
"plt.style.use('ggplot')\n",
|
|
"\n",
|
|
"def plot_result_graphs(plot_name, stats, keys_to_plot, notebook=True):\n",
|
|
" \n",
|
|
" fig_1 = plt.figure(figsize=(8, 4))\n",
|
|
" ax_1 = fig_1.add_subplot(111)\n",
|
|
" for name in keys_to_plot:\n",
|
|
" for k in ['train_loss', 'val_loss']:\n",
|
|
" item = stats[name][k]\n",
|
|
" ax_1.plot(np.arange(0, len(item)), \n",
|
|
" item, label='{}_{}'.format(name, k))\n",
|
|
" \n",
|
|
" ax_1.legend(loc=0)\n",
|
|
" ax_1.set_ylabel('Loss')\n",
|
|
" ax_1.set_xlabel('Epoch number')\n",
|
|
"\n",
|
|
" # Plot the change in the validation and training set accuracy over training.\n",
|
|
" fig_2 = plt.figure(figsize=(8, 4))\n",
|
|
" ax_2 = fig_2.add_subplot(111)\n",
|
|
" for name in keys_to_plot:\n",
|
|
" for k in ['train_acc', 'val_acc']:\n",
|
|
" item = stats[name][k]\n",
|
|
" ax_2.plot(np.arange(0, len(item)), \n",
|
|
" item, label='{}_{}'.format(name, k))\n",
|
|
" \n",
|
|
" ax_2.legend(loc=0)\n",
|
|
" ax_2.set_ylabel('Accuracy')\n",
|
|
" ax_2.set_xlabel('Epoch number')\n",
|
|
" \n",
|
|
" fig_1.savefig('../data/{}_loss_performance.pdf'.format(plot_name), dpi=None, facecolor='w', edgecolor='w',\n",
|
|
" orientation='portrait', papertype=None, format='pdf',\n",
|
|
" transparent=False, bbox_inches=None, pad_inches=0.1,\n",
|
|
" frameon=None, metadata=None)\n",
|
|
" \n",
|
|
" fig_2.savefig('../data/{}_accuracy_performance.pdf'.format(plot_name), dpi=None, facecolor='w', edgecolor='w',\n",
|
|
" orientation='portrait', papertype=None, format='pdf',\n",
|
|
" transparent=False, bbox_inches=None, pad_inches=0.1,\n",
|
|
" frameon=None, metadata=None)\n",
|
|
" \n",
|
|
" "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 24,
|
|
"metadata": {
|
|
"scrolled": true
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stderr",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"<ipython-input-23-0810a62e0421>:32: MatplotlibDeprecationWarning: \n",
|
|
"The frameon kwarg was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use facecolor instead.\n",
|
|
" fig_1.savefig('../data/{}_loss_performance.pdf'.format(plot_name), dpi=None, facecolor='w', edgecolor='w',\n",
|
|
"<ipython-input-23-0810a62e0421>:37: MatplotlibDeprecationWarning: \n",
|
|
"The frameon kwarg was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use facecolor instead.\n",
|
|
" fig_2.savefig('../data/{}_accuracy_performance.pdf'.format(plot_name), dpi=None, facecolor='w', edgecolor='w',\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 576x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 576x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plot_result_graphs('problem_model', result_dict, keys_to_plot=['VGG_38', 'VGG_08'])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "mlp",
|
|
"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": "3.12.5"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|