{"cells":[{"cell_type":"markdown","id":"d5a67ee6","metadata":{"id":"d5a67ee6"},"source":["[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](http://colab.research.google.com/github/ai2es/WAF_ML_Tutorial_Part1/blob/main/colab_notebooks/Notebook04_SimpleMLClassification.ipynb)\n","\n","# Notebook 04: Simple ML Classification [Colab Version]\n","\n","### Goal: Basic training a ML using a single feature/predictor/input and a single ML model\n","\n","#### Reminder of Problem Statement\n","\n","Before we jump into the ML, I want to remind you of the ML task we want to accomplish in the paper. \n","\n","1. Does this image contain a thunderstorm? <-- Classification\n","2. How many lightning flashes are in this image? <-- Regression\n","\n","#### Background\n","\n","For the training of machine learning we will use the [scikit-learn](https://scikit-learn.org/stable/) python package. Scikit-learn (also known as sklearn) has a wealth of resources for learning how to apply many ML methods. Furthermore, their documentation on any one method is extensive and very helpful. I encourage you to use their documentation if you want to use more that what we will show you in this tutorial. \n","\n","What is really nice about the sklearn package is that all of its models all work off the same syntax. In general this is how you use any of their models:\n","\n","1. Create or load your input data. It must be of shape ```[n_samples,n_features]```. It is commonly written as ```X``` in all of the sklearn documentation. \n","2. Create or load your output data. It must be of shape ```[n_samples]```. It is commonly written as ```y``` in all of the sklearn documentation \n","3. Initialize your model. To initialize a model in python usually you just need to add ```()``` at the end of the model name. \n","4. Fit your model. To do this all we will need to do is apply the ```.fit(X,y)``` to your initialized model \n","5. Evaluate your trained model. To get the model predictions for evaluation we will use the ```model.predict(X_val)``` or the ```model.predict_proba(X_val)```. \n","\n","Now that we have the basic work flow, lets actually do this with the simple example in the paper, 1 input predictor. We will start with the classification task first\n","\n","#### Step 0: Get the github repo (we need some of the functions there)\n","\n","The first step with all of these Google Colab notebooks will be to grab the github repo and cd into the notebooks directory. \n","\n","To run things from the command line, put a ```!``` before your code"]},{"cell_type":"code","source":["#get the github repo \n","!git clone https://github.com/ai2es/WAF_ML_Tutorial_Part1.git \n","\n","#cd into the repo so the paths work \n","import os \n","os.chdir('/content/WAF_ML_Tutorial_Part1/jupyter_notebooks/')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fC7lgQiPU2YC","executionInfo":{"status":"ok","timestamp":1648652456805,"user_tz":300,"elapsed":21197,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"e1e327b9-9722-4134-dac5-467b781ff021"},"id":"fC7lgQiPU2YC","execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Cloning into 'WAF_ML_Tutorial_Part1'...\n","remote: Enumerating objects: 301, done.\u001b[K\n","remote: Counting objects: 100% (301/301), done.\u001b[K\n","remote: Compressing objects: 100% (197/197), done.\u001b[K\n","remote: Total 301 (delta 139), reused 236 (delta 96), pack-reused 0\u001b[K\n","Receiving objects: 100% (301/301), 195.77 MiB | 24.94 MiB/s, done.\n","Resolving deltas: 100% (139/139), done.\n","Checking out files: 100% (100/100), done.\n"]}]},{"cell_type":"markdown","source":["#### Step 1 & 2: Import packages and load data for Classification  \n","Last notebook, we showed you how to do train/val and test data splits. We will do that all with our pre-made function now. But we only want 1 feature, which is feature 0. So all we need to change is the ```features_to_keep``` keyword to get just the first feature."],"metadata":{"id":"wkPuU9l_U1cr"},"id":"wkPuU9l_U1cr"},{"cell_type":"code","execution_count":2,"id":"a90ffcc8","metadata":{"id":"a90ffcc8","executionInfo":{"status":"ok","timestamp":1648652464446,"user_tz":300,"elapsed":7644,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["#needed packages \n","import xarray as xr\n","import matplotlib.pyplot as plt \n","import numpy as np\n","import pandas as pd\n","\n","#plot parameters that I personally like, feel free to make these your own.\n","import matplotlib\n","matplotlib.rcParams['axes.facecolor'] = [0.9,0.9,0.9] #makes a grey background to the axis face\n","matplotlib.rcParams['axes.labelsize'] = 14 #fontsize in pts\n","matplotlib.rcParams['axes.titlesize'] = 14 \n","matplotlib.rcParams['xtick.labelsize'] = 12 \n","matplotlib.rcParams['ytick.labelsize'] = 12 \n","matplotlib.rcParams['legend.fontsize'] = 12 \n","matplotlib.rcParams['legend.facecolor'] = 'w' \n","matplotlib.rcParams['savefig.transparent'] = False\n","\n","#make default resolution of figures much higher (i.e., High definition)\n","%config InlineBackend.figure_format = 'retina'\n","\n","#import some helper functions for our other directory.\n","import sys\n","sys.path.insert(1, '../scripts/')\n","from aux_functions import load_n_combine_df\n","(X_train,y_train),(X_validate,y_validate),(X_test,y_test) = load_n_combine_df(path_to_data='../datasets/sevir/',features_to_keep=np.arange(0,1,1),class_labels=True)"]},{"cell_type":"markdown","id":"dd293191","metadata":{"id":"dd293191"},"source":["Let's print out the shapes of things to make sure they match what we expected, ```[n_samples,n_features]``` for ```X``` and ```[n_samples]``` for ```y```"]},{"cell_type":"code","execution_count":3,"id":"3d32e5df","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3d32e5df","executionInfo":{"status":"ok","timestamp":1648652464447,"user_tz":300,"elapsed":6,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"e736adaa-5647-4a17-d6b0-fc7012572ebd"},"outputs":[{"output_type":"stream","name":"stdout","text":["X_train, y_train shapes: (446320, 1),(446320,)\n","X_val, y_val shapes: (86292, 1),(86292,)\n","X_test, y_test shapes: (86297, 1),(86297,)\n"]}],"source":["print('X_train, y_train shapes: {},{}'.format(X_train.shape,y_train.shape))\n","print('X_val, y_val shapes: {},{}'.format(X_validate.shape,y_validate.shape))\n","print('X_test, y_test shapes: {},{}'.format(X_test.shape,y_test.shape))"]},{"cell_type":"markdown","id":"ab0fa5d2","metadata":{"id":"ab0fa5d2"},"source":["Okay, things look correct. Our ```n_samples``` match of each dataset, and ```X``` is indeed ```[n_samples,n_features]```. Let us reproduce Figure 8 to *gauge* the predictive power of the minimum brightness temperature from the clean infrared channel. To do this, we need to find which samples have lightning flashes and which do not. We will use the ```np.where``` to find the indicies of ```X``` and ```y``` where the label (i.e., ```y```) is equal to 1. When you set ```class_labels=True``` in the ```load_n_combine_df``` it will already label ```y``` as either 0 where there are no flashes, and 1 where there is greater than or equal to 1 flash in the image.  "]},{"cell_type":"code","execution_count":4,"id":"732c01b5","metadata":{"id":"732c01b5","executionInfo":{"status":"ok","timestamp":1648652464447,"user_tz":300,"elapsed":4,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["#find samples with more at least 1 flash \n","idx_flash = np.where(y_train == 1)[0]\n","#find samples with no flashes \n","idx_noflash = np.where(y_train == 0)[0]"]},{"cell_type":"markdown","id":"59d24361","metadata":{"id":"59d24361"},"source":["For those who have never used the ```np.where``` function before, we can double check it worked by looking at a few histograms."]},{"cell_type":"code","execution_count":5,"id":"60466166","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":820},"id":"60466166","executionInfo":{"status":"ok","timestamp":1648652465164,"user_tz":300,"elapsed":720,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"7b9440e2-59e3-4ba0-c837-b95117dfb518"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["(array([224262.,      0.,      0.]),\n"," array([-0.01,  0.01,  0.99,  1.01]),\n"," <a list of 3 Patch objects>)"]},"metadata":{},"execution_count":5},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":402,"height":250},"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":402,"height":250},"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":402,"height":250},"needs_background":"light"}}],"source":["#show us the whole dataset\n","plt.hist(y_train,bins=[-0.01,0.01,0.99,1.01],edgecolor='k')\n","\n","#show us the flash data\n","plt.figure()\n","plt.hist(y_train[idx_flash],bins=[-0.01,0.01,0.99,1.01],edgecolor='k')\n","\n","#show us the no flash data\n","plt.figure()\n","plt.hist(y_train[idx_noflash],bins=[-0.01,0.01,0.99,1.01],edgecolor='k')"]},{"cell_type":"markdown","id":"2f5cef69","metadata":{"id":"2f5cef69"},"source":["Now that things look to be working properly, lets make the histogram of brightness temperatures. The expectation is that images with flashes in them will have generally colder minimum brightness temperatures since they will likely have stronger storms (on average). "]},{"cell_type":"code","execution_count":6,"id":"0e3bd13c","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":367},"id":"0e3bd13c","executionInfo":{"status":"ok","timestamp":1648652466339,"user_tz":300,"elapsed":1178,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"c823393c-b0c4-422a-8c9f-766dbb43c7fb"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 1 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":353,"height":350}}}],"source":["#this is something to help make the ticks show up where I want them \n","from matplotlib.ticker import (MultipleLocator, FormatStrFormatter,\n","                               AutoMinorLocator)\n","\n","#this is an array of bins I wish to do the histogram with, in degC\n","xbins = np.arange(-100,50,2)\n","#some colors I like. The order of ratios is [Red,Green,Blue]\n","r = [255/255,127/255,127/255]\n","b = [126/255,131/255,248/255]\n","\n","#make figure  \n","fig = plt.figure(figsize=(5,5))\n","#set background color to white so we can copy paste out of the notebook if we want \n","fig.set_facecolor('w')\n","\n","#get axis for drawing\n","ax = plt.gca()\n","\n","#use matplotlib histogram function \n","ax.hist(X_train[idx_flash,0],density=True,bins=xbins,color=b,alpha=0.5,zorder=0,label='T-storm',edgecolor=b)\n","ax.hist(X_train[idx_noflash,0],density=True,bins=xbins,color=r,alpha=0.5,zorder=0,label='No T-Storm',edgecolor=r)\n","\n","#add title and axis labels \n","ax.set_title('Minimum IR')\n","ax.set_ylabel(r'Normalized count')\n","ax.set_xlabel(\"$T_{b}$, [$\\degree$C]\")\n","#set tick locations \n","ax.xaxis.set_minor_locator(MultipleLocator(5))\n","ax.yaxis.set_minor_locator(MultipleLocator(0.001))\n","#force xlimit\n","ax.set_xlim([-100,50])\n","#turn grid on \n","ax.grid('on')\n","#draw legend\n","ax.legend()\n","\n","plt.tight_layout()"]},{"cell_type":"markdown","id":"334acc45","metadata":{"id":"334acc45"},"source":["Great! It follows our expectation. While there is considerable overlap between the two classes, there does seem to be a good decision boundary to separate the two. Let's start out by training a logistic regression first. \n","\n","#### Step 3: Initialize model\n","\n","Like I mentioned in the background, to initialize we can use the ```()``` after the model name. This is also where you can choose various hyperparameters. For now, just leave it empty, it will use the default parameters, which is a good place to start"]},{"cell_type":"code","execution_count":7,"id":"7e27560b","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7e27560b","executionInfo":{"status":"ok","timestamp":1648652467533,"user_tz":300,"elapsed":1196,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"2db31ae8-cc28-40c8-a9b4-8345e886f766"},"outputs":[{"output_type":"stream","name":"stdout","text":["LogisticRegression()\n"]}],"source":["#load model from sklearn\n","from sklearn.linear_model import LogisticRegression\n","\n","#initialize\n","model = LogisticRegression()\n","\n","print(model)"]},{"cell_type":"markdown","id":"e43577bf","metadata":{"id":"e43577bf"},"source":["#### Step 4: Train your ML model! \n","\n","This is it, we have finally made it to actually doing some machine learning, are you ready? I am! So let's go!"]},{"cell_type":"code","execution_count":8,"id":"b0e12ba5","metadata":{"id":"b0e12ba5","executionInfo":{"status":"ok","timestamp":1648652468575,"user_tz":300,"elapsed":1045,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["model = model.fit(X_train,y_train)"]},{"cell_type":"markdown","id":"fe2bd057","metadata":{"id":"fe2bd057"},"source":["<img src=\"https://drive.google.com/uc?id=1NX3-59H6rxo6TXAe53YAoDUdsrJ_j04v\" width=\"400\" height=\"200\" class=\"center\" />\n","\n","Thats it! Congrats, you have officially trained your first ML model. I know, it seems a bit underwhelming because there is no fanfare from python, but to make you feel better enjoy the gif. After you are done celebrating, time to move on to evaluating how well it did!\n","\n","#### Step 5: Evaluate your ML model\n","\n","In order to evaluate your model, we will need the predictions of the model on the validation dataset. We will use the ```model.predict()``` method to get a prediction value for each sample"]},{"cell_type":"code","execution_count":9,"id":"83f7b0a8","metadata":{"id":"83f7b0a8","executionInfo":{"status":"ok","timestamp":1648652477313,"user_tz":300,"elapsed":192,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[],"source":["yhat = model.predict(X_validate)"]},{"cell_type":"markdown","id":"708e3dbc","metadata":{"id":"708e3dbc"},"source":["I name the prediction vector yhat (i.e., $\\hat{y}$) to match usual math notation. But you can name it ypred if you want. This vector is a label for each sample, so we could look at a histogram of the results, or we can calculate some more informative metrics. \n","\n","##### a). Contingency Table: Accuracy \n","\n","To get those more informative metrics, lets get the contingency table which is:\n","\n","\n","|             | Yes | No|\n","| ----------- | :----: | :----: |\n","|    **Yes**      | Hits       | False Alarms|\n","|    **No**      | Misses        | Correct Nulls |\n","\n","The left column is the model predictions, while the top row is the observations (true labels). Each spot in the table is filled out based on:\n","- if both the model and observation predict **Yes** (both predict a label of 1; which is a hit or *True Positive*)\n","- if the model outputs a **Yes** but the observation is **No** (model=1,obs=0; False Alarm or *False Positive*)\n","- if the model says **No** but the observation says **Yes** (model=0,obs=1; Miss or *False Positive*)\n","- if the model and the observation say **No** (model=0,obs=0; Correct Null or *True Negative*)\n","\n","We have coded this up for you, so all we need to use is the function called ```get_contingency_table``` from ```gewitter_functions.py```. These functions were original written by Dr. Lagerquist during his PhD studies, but have been adapted by me to be able to run out of the box with just the ```gewitter_functions.py``` script."]},{"cell_type":"code","execution_count":10,"id":"daf40aa7","metadata":{"id":"daf40aa7","outputId":"8172a105-1042-4e39-b325-77605d7cf571","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1648652480011,"user_tz":300,"elapsed":204,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["{'num_true_positives': 37545, 'num_false_positives': 9775, 'num_false_negatives': 6378, 'num_true_negatives': 32594}\n"]}],"source":["from gewitter_functions import get_contingency_table\n","\n","#the contingency table calculator expects y_true,y_pred\n","cont_table = get_contingency_table(y_validate,yhat)\n","\n","\n","#the function returns a dictionary (thats what the curly brackets mean) of each entry in the table. \n","print(cont_table)"]},{"cell_type":"markdown","id":"e8aa8bde","metadata":{"id":"e8aa8bde"},"source":["Now that we have the contingency table, we can calculate a simple metric, accuracy. Which is:\n","\n","$$ Accuracy = \\frac{TruePositives + TrueNegatives}{SumOfAllCategories} \\times{100}$$\n","\n","for conciseness, we have made a function to do this named ```get_accuracy```"]},{"cell_type":"code","execution_count":11,"id":"2358bc77","metadata":{"id":"2358bc77","outputId":"1f8fe948-984d-476a-901b-80e9be16eade","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1648652532850,"user_tz":300,"elapsed":193,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["Accuracy = 81.0%\n"]}],"source":["from gewitter_functions import get_acc\n","accuracy = get_acc(cont_table)\n","print('Accuracy = {}%'.format(np.round(accuracy,0)))"]},{"cell_type":"markdown","id":"09c1c27d","metadata":{"id":"09c1c27d"},"source":["81% accuracy is good! Given we only use 1 input! But as pointed out in the paper, it is good to always use more than 1 evaluation metric. \n","\n","##### b). Contingency Table: Performance Diagram\n","\n","<img src=\"https://drive.google.com/uc?id=17xbuOMIaBbRf_gaUYSIbfFMBO3i_3vx9\" width=\"400\" height=\"400\" class=\"center\" />\n","\n","One of my favorite diagrams to gauge model skill is a performance diagram (shown above). I have added some hopefully helpful annotations that show you where better performing models (top-right) and worse performing models (bottom left, top left or bottom right) show up on the diagram. \n","\n","To make this diagram we need to calculate 2 variables, the Probability of Detection (POD; y-axis) and Success Ratio (SR; x-axis), both that are calculated from the contingency table. \n","\n","$$ \\mathrm{POD} = \\frac{TruePositives}{TruePositives + FalseNegatives} $$\n","\n","$$ \\mathrm{SR} = \\frac{TruePositives}{TruePositives + FalsePositives} $$\n","\n","One perk of this diagram is that we can glean the *Critical Success Index* (CSI) from the same diagram. CSI is formulated as: \n","\n","$$ \\mathrm{CSI} = \\frac{TruePositives}{TruePositives + FalsePositives + FalseNegatives} $$ \n","\n","What is great about the performance diagram is that it helps us evaluate ML tasks that might have unbalanced data. By unbalanced data, I mean if the scenario we are trying to predict is rare (e.g., hail, tornadoes). What happens in rare scenarios is that the data naturally have a ton of Null cases, so some metrics that consider TrueNegatives in the numerator (e.g., Accuracy) might be inflated. \n","\n","We have included the calculation of these parameters as functions as well ```get_pod```, ```get_sr```, ```csi_from_sr_and_pod```."]},{"cell_type":"code","execution_count":12,"id":"b73677cb","metadata":{"id":"b73677cb","outputId":"52b6bee6-da9e-4244-eb00-750eaf874934","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1648652948011,"user_tz":300,"elapsed":214,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["POD:0.85, SR:0.79, CSI:0.7\n"]}],"source":["from gewitter_functions import get_pod,get_sr,csi_from_sr_and_pod\n","pod = get_pod(cont_table)\n","sr = get_sr(cont_table)\n","csi = csi_from_sr_and_pod(sr,pod)\n","\n","print('POD:{}, SR:{}, CSI:{}'.format(np.round(pod,2),np.round(sr,2),np.round(csi,2)))"]},{"cell_type":"markdown","id":"9487fb90","metadata":{"id":"9487fb90"},"source":["Now that we have our calculated values, you could put these in a table, or you could put it on the performance diagram. We made a function to make the performance diagram axis for us, named ```make_performance_diagram```"]},{"cell_type":"code","execution_count":13,"id":"5a1fbd98","metadata":{"id":"5a1fbd98","outputId":"bceec954-48b3-43fd-bb4b-734e3b2c927a","colab":{"base_uri":"https://localhost:8080/","height":344},"executionInfo":{"status":"ok","timestamp":1648652950666,"user_tz":300,"elapsed":780,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 2 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":353,"height":327}}}],"source":["from gewitter_functions import make_performance_diagram_axis\n","ax = make_performance_diagram_axis()"]},{"cell_type":"markdown","id":"5c07c80a","metadata":{"id":"5c07c80a"},"source":["the function returns the axis (```ax```) so we can plot on it. "]},{"cell_type":"code","execution_count":14,"id":"844c1e1b","metadata":{"id":"844c1e1b","outputId":"9ef1a19d-3f6d-49e0-f42c-f20eb08af2e8","colab":{"base_uri":"https://localhost:8080/","height":344},"executionInfo":{"status":"ok","timestamp":1648652953753,"user_tz":300,"elapsed":697,"user":{"displayName":"Randy C","userId":"12301342565919601334"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 360x360 with 2 Axes>"],"image/png":"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\n"},"metadata":{"image/png":{"width":353,"height":327}}}],"source":["ax = make_performance_diagram_axis()\n","ax.plot(sr,pod,'o',color='dodgerblue',markerfacecolor='w');"]},{"cell_type":"markdown","id":"7a739b4d","metadata":{"id":"7a739b4d"},"source":["and BAM. You have plotted your model's performance on the performance diagram. Turns out this model is doing very well, with a large POD and SR and is near the center line. \n","\n","#### Step 6: Save your trained model\n","\n","Sometimes loading the training dataset and re-training the model each time can be cumbersome. There is a way to save the trained models. We will use the python ```joblib``` package to do this. \n","\n","**Notes:** \n","\n","1. Storage efficiency: While most of the models in sklearn can be stored in a decent file size, your random forest model could become large (on the order of 1 GB). I know a GB is not alot of space nowadays, but it is important to note. This is also why the trained random forest from the paper is not found in the ```datasets``` folder.\n","2. Keep note of your ```sklearn``` version: ```joblib``` will load the model to the syntax of the ```sklearn``` that it is saved with. So if you are getting some odd error with ```sklearn``` while loading the saved model, check to make sure the version hasn’t changed."]},{"cell_type":"code","execution_count":15,"id":"868e1bd9","metadata":{"id":"868e1bd9","colab":{"base_uri":"https://localhost:8080/","height":222},"executionInfo":{"status":"error","timestamp":1648652962920,"user_tz":300,"elapsed":180,"user":{"displayName":"Randy C","userId":"12301342565919601334"}},"outputId":"17ad10dc-e017-4ee2-9e54-c87964591f83"},"outputs":[{"output_type":"error","ename":"FileNotFoundError","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)","\u001b[0;32m<ipython-input-15-3667fb553b5a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mname\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'LgR.pkl'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mstart_path\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'../datasets/sklearnmodels/classification/onefeature/'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m \u001b[0msavefile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstart_path\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'wb'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m \u001b[0mjoblib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdump\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msavefile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m9\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m#the 9 here is the highest compression. Slows the time of saving/loading but saves disk space\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '../datasets/sklearnmodels/classification/onefeature/LgR.pkl'"]}],"source":["import joblib\n","name = 'LogisticRegression.pkl'\n","start_path = '../datasets/sklearnmodels/classification/onefeature/'\n","savefile = open(start_path + name,'wb')\n","joblib.dump(model, savefile, 9) #the 9 here is the highest compression. Slows the time of saving/loading but saves disk space"]},{"cell_type":"markdown","id":"61254a51","metadata":{"id":"61254a51"},"source":["#### Step 7: Load a saved model\n","\n","Now that you have it saved, if you need to load it do the following:"]},{"cell_type":"code","execution_count":null,"id":"298d94ba","metadata":{"id":"298d94ba","outputId":"808bebd0-d3ab-4fc1-b0de-614ea2c4a023"},"outputs":[{"name":"stdout","output_type":"stream","text":["LogisticRegression()\n"]}],"source":["import joblib\n","name = 'LogisticRegression.pkl'\n","start_path = '../datasets/sklearnmodels/classification/onefeature/'\n","#notice the change from wb to rb \n","savefile = open(start_path + name,'rb')\n","#notice the change from dump to load \n","model = joblib.load(savefile)\n","print(model)"]},{"cell_type":"markdown","id":"f3d0f087","metadata":{"id":"f3d0f087"},"source":["In the next notebook we will look at the Regression problem. If you want to continue on with the classification, check out Notebook 6. "]}],"metadata":{"kernelspec":{"display_name":"Python 3 (ipykernel)","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":"3.9.7"},"colab":{"name":"Notebook04_SimpleMLClassification.ipynb","provenance":[],"collapsed_sections":[]}},"nbformat":4,"nbformat_minor":5}