exercise-4/Exercise 4 - Consume Data using Data Science.ipynb

{"cells":[{"cell_type":"markdown","source":["# Exercise 4.1 - Consume Data using Fabric Data Science experience"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"1bca4075-28ef-422a-8489-29182dc76e01"},{"cell_type":"markdown","source":["## Exercise outline\n","\n","1) **Read data from silver layer lakehouse tables using Apache Spark**\n","2) **Explore and Visualize Data using Notebooks**\n","3) **Train a machine learning model and register the model artifact using MLflow**\n","4) **Perform Batch Scoring and save trip predictions to gold layer lakehouse**"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"62bf8625-984d-4a8d-bb42-8b1e066ca3b6"},{"cell_type":"markdown","source":["> **NOTE: Please Attach Silver and Gold layer Lakehouses to this notebook before executing below steps** <br>\n","> Steps to attach lakehouses can be found [here](https://github.com/ekote/Build-Your-First-End-to-End-Lakehouse-Solution/blob/fabcon/exercise-2/exercise-2.md#task-25-attach-the-bronze-lakehouse)"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"c70623e0-f2c1-4d56-a10f-7cb3bc61d18f"},{"cell_type":"markdown","source":["#### Step 0:  Define common lakehouse parameters"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"53b214ad-e055-483a-9717-41466d388c39"},{"cell_type":"code","source":["## Lakehouse Names\n","SILVER_LAKEHOUSE = \"silvercleansed\"\n","GOLD_LAKEHOUSE = \"goldcurated\"\n","\n","#Table Names\n","TRAINING_TABLE = \"green201501_cleansed\"\n","EVAL_TABLE = \"green202301_cleansed\"\n","PREDICT_TABLE = \"greentaxi_predicted\"\n","\n","# SET SQL Variables to explore data using Spark SQL\n","spark.sql(f\"SET SILVER_LAKEHOUSE ={SILVER_LAKEHOUSE}\")\n","spark.sql(f\"SET GOLD_LAKEHOUSE ={GOLD_LAKEHOUSE}\")\n","spark.sql(f\"SET TRAINING_TABLE ={TRAINING_TABLE}\")\n","spark.sql(f\"SET EVAL_TABLE ={EVAL_TABLE}\")\n","spark.sql(f\"SET PREDICT_TABLE ={PREDICT_TABLE}\")"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"2f77e7f2-f244-4ad9-882b-463ccc11dd06"},{"cell_type":"markdown","source":["## Step 1: Read data from Fabric lakehouse tables using Apache Spark\n","**Lakehouse**:\n","A lakehouse is a collection of files/folders/tables that represent a database over a data lake used by \n","the Spark engine and SQL engine for big data processing and that includes enhanced capabilities for \n","ACID transactions when using the open-source Delta formatted tables.\n","\n","**Delta Lake**:Delta Lake is an open-source storage layer that brings ACID transactions, scalable metadata management, and batch and streaming data processing to Apache Spark. A Delta Lake table is a data table format that extends Parquet data files with a file-based transaction log for ACID transactions and scalable metadata management."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"1a81da4f-03d9-46ce-9ade-3d5ff54b7796"},{"cell_type":"code","source":["# Read delta table from lakehouse - silver zone\n","data = spark.read.table(f\"{SILVER_LAKEHOUSE}.{TRAINING_TABLE}\")"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"eeb21e6b-35ba-46d1-b96d-f089a1d8caa0"},{"cell_type":"markdown","source":["# Step 2: Explore and Visualize Data\n","\n","In this Step we will use seaborn, a Python data visualization library that provides a high-level interface for building visuals on dataframes and arrays. You can learn more about seaborn [here](https://seaborn.pydata.org/)."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"0a2d44c8-4675-49fa-88e8-f143b73f2ac8"},{"cell_type":"markdown","source":["##### Import visualization libraries and set figure config"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"ee4858cf-d651-4a1b-a6cf-823b1f2a1a81"},{"cell_type":"code","source":["import seaborn as sns\n","import matplotlib.pyplot as plt\n","import matplotlib.ticker as mticker\n","import numpy as np\n","sns.set_theme(style=\"whitegrid\", palette=\"tab10\", rc = {'figure.figsize':(9,6)})"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"e9401902-c185-4592-80bc-13bba22b561f"},{"cell_type":"code","source":["# Note: For the purpose of minimizing runtime in this exercise, \n","# We are using a 1/100 sample to explore and visualize ingested data using pandas\n","SEED = 1234\n","sampled_df = data.sample(True, 0.01, seed=SEED).toPandas()\n","sampled_df"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"46a7c892-4afb-46d9-b566-745856a4cf4a"},{"cell_type":"markdown","source":["##### Visual 1: Distribution of trip duration(minutes) on linear and logarithmic scale"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"1ab43684-ae3c-4aad-9601-8ff13f49f7e6"},{"cell_type":"code","source":["## Compute trip duration(in minutes) on the sample using pandas\n","sampled_df['trip_duration'] = (sampled_df['lpep_dropoff_datetime'] - sampled_df['lpep_pickup_datetime']).astype('timedelta64[s]').dt.seconds/60\n","sampled_df = sampled_df[sampled_df[\"trip_duration\"] > 0]\n","\n","fig, axes = plt.subplots(1, 2, figsize=(18, 6))\n","sns.histplot(ax=axes[0],data=sampled_df,\n","            x=\"trip_duration\",\n","            stat=\"count\",\n","            discrete=True).set(title='Distribution of trip duration(minutes)')\n","sns.histplot(ax=axes[1],data=sampled_df,\n","            x=\"trip_duration\",\n","            stat=\"count\", \n","            log_scale= True).set(title='Distribution of trip duration(log scale)')\n","axes[1].xaxis.set_major_formatter(mticker.ScalarFormatter())\n","plt.show()"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"e83b7d9d-d7d1-496d-b677-e43fa349d15d"},{"cell_type":"markdown","source":["#### Visual 2: Lets create bins to visualize duration of trips better"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"343283b8-dac5-4ec2-a909-f5f242a20e5c"},{"cell_type":"code","source":["## Create bins for trip_duration column\n","sampled_df.loc[sampled_df['trip_duration'].between(0, 10, 'both'), 'durationBin'] = '< 10 Mins'\n","sampled_df.loc[sampled_df['trip_duration'].between(10, 30, 'both'), 'durationBin'] = '10-30 Mins'\n","sampled_df.loc[sampled_df['trip_duration'].between(30, 60, 'both'), 'durationBin'] = '30-60 Mins'\n","sampled_df.loc[sampled_df['trip_duration'].between(60, 120, 'right'), 'durationBin'] = '1-2 Hrs'\n","sampled_df.loc[sampled_df['trip_duration'].between(120, 240, 'right'), 'durationBin'] = '2-4 Hrs'\n","sampled_df.loc[sampled_df['trip_duration'] > 240, 'durationBin'] = '> 4 Hrs'\n","\n","# Plot histogram using the binned column\n","sns.histplot(data=sampled_df, x=\"durationBin\", stat=\"count\", discrete=True, hue = \"durationBin\")\n","plt.title(\"Trip Distribution by Duration Bins\")\n","plt.xlabel('Trip Duration')\n","plt.ylabel('Frequency')"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"5831aa9e-e44d-4767-8ae5-60d4d97a7147"},{"cell_type":"markdown","source":["#### Visual 3: Visualize the distribution of trip_duration and trip_distance and classify by passenger_count"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"705829de-d280-4dca-905c-2a4d4522a064"},{"cell_type":"code","source":["sns.scatterplot(data=sampled_df, x=\"trip_distance\", y=\"trip_duration\", hue=\"passenger_count\")\n","plt.show()"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{},"ms_comments":[{"createdDateUTC":1681409284698,"modifiedDateUTC":1681409284698,"replies":[{"createdDateUTC":1681511071018,"modifiedDateUTC":1681591272371,"replyId":"9d336656-79fb-4bce-b419-e63b4af06c04","text":"+1 here, we can tell a story on how EDA is informing ETL","user":{"idType":"aad","name":"Fatemeh Zamanian"}}],"status":"active","text":"Is it possible to add some markdown comments on any conclusions we can draw from these data exploration steps?","threadId":"2b55020f-e1cf-4f42-92bd-6382604dffd4","user":{"idType":"aad","name":"Nellie Gustafsson"}}],"nteract":{"transient":{"deleting":false}}},"id":"ed73827c-2712-45d3-8a08-f08ba905f47b"},{"cell_type":"markdown","source":["#### Visual 4: Visualize distribution of passenger_count per trip"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"4f714fa0-357a-4cb0-ba81-4692e179f220"},{"cell_type":"code","source":["sns.histplot(data=sampled_df, x=\"passenger_count\", stat=\"count\", discrete=True)\n","plt.title(\"Distribution of passenger count\")\n","plt.xlabel('No. of Passengers')\n","plt.ylabel('Number of trips')"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"c12ad9b5-fbc1-496a-8477-517453c963da"},{"cell_type":"markdown","source":["#### Visual 5:  Create boxplots to visualize the distribution of trip_duration by passenger count\n","A boxplot is a useful tool to understand the variability, symmetry, and outliers of the data.\n","- In first figure lets visualize trip_duration without removing any outliers\n","- In the second figure we are removing trips with duration greater than 3 hours and zero passengers."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"29576e27-622b-46e0-a42b-380e7012d54c"},{"cell_type":"code","source":["fig, axes = plt.subplots(1, 2, figsize=(18, 6))\n","sns.boxplot(ax=axes[0], data=sampled_df, x=\"passenger_count\", y=\"trip_duration\").set(title='Distribution of Trip duration by passenger_count')\n","sampleddf_clean = sampled_df[(sampled_df[\"passenger_count\"] > 0) & (sampled_df[\"trip_duration\"] < 60)]\n","sns.boxplot(ax=axes[1], data=sampleddf_clean, x=\"passenger_count\", y=\"trip_duration\").set(title='Distribution of Trip duration by passenger_count (outliers removed)')\n","plt.show()"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"68a69d9c-9320-4b3b-a5ae-b69cd5959f91"},{"cell_type":"markdown","source":["#### Visual 6: Analyze the relationship of trip_duration and fare_amount classified by payment_type and VendorID using a scatterplot/subplots"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"f8ba94ff-1cbe-4f06-8d03-d07c13d70446"},{"cell_type":"code","source":["f, axes = plt.subplots(1, 2, figsize=(18, 6))\n","sns.scatterplot(ax =axes[0], data=sampled_df, x=\"fare_amount\", y=\"trip_duration\",  hue=\"payment_type\")\n","sns.scatterplot(ax =axes[1],data=sampled_df, x=\"fare_amount\", y=\"trip_duration\",  hue=\"VendorID\")\n","plt.title(\"Distribution of trip_duration by fare_amount\")\n","plt.show()"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"3bac7a48-9f2d-48d3-9f5a-0ec8e9767451"},{"cell_type":"markdown","source":["#### Visual 7: Analyze the frequency of the taxi trips by hour of the day"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"c7468107-4862-4226-bbe6-87dbbe9a83f3"},{"cell_type":"code","source":["sampled_df['hour'] = sampled_df['lpep_pickup_datetime'].dt.hour\n","sampled_df['dayofweek'] = sampled_df['lpep_dropoff_datetime'].dt.dayofweek\n","sampled_df['dayname'] = sampled_df['lpep_dropoff_datetime'].dt.day_name()\n","sns.histplot(data=sampled_df, x=\"hour\", stat=\"count\", discrete=True, kde=True)\n","plt.title(\"Distribution by Hour of the day\")\n","plt.xlabel('Hours')\n","plt.ylabel('Count of trips')\n","plt.show()"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"6548af2e-431f-4f98-b8ae-1dab84f059b7"},{"cell_type":"markdown","source":["#### Visual 8: Analyze average taxi trip duration by hour and day of the week using a heatmap"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"1d0076ee-06b9-4eb4-bce9-c02e8ffcb0ee"},{"cell_type":"code","source":["pv_df = sampled_df[sampled_df[\"trip_duration\"]<180]\\\n","        .groupby([\"hour\",\"dayname\"]).mean(\"trip_duration\")\\\n","        .reset_index().pivot(index = \"hour\", columns= \"dayname\", values = \"trip_duration\")\n","sns.heatmap(pv_df,annot=True,fmt='.2f', cmap=\"Blues\").set(xlabel=None)"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{"5e03e664-d68c-45cf-bebc-2690e7100df5":{"end":{"column":54,"line":4},"start":{"column":47,"line":4},"text":"Blues\")"}},"ms_comments":[{"createdDateUTC":1681421320987,"modifiedDateUTC":1681421663259,"replies":[{"createdDateUTC":1681421504147,"modifiedDateUTC":1681421504147,"replyId":"26a6bfe6-20d2-4bbf-a5e1-b1dc2fabf19e","text":"Done","user":{"idType":"aad","name":"Abid Nazir Guroo"}}],"status":"resolved","text":"AS Nit: i'd probably remove 'dayname', as it's obvious from the labels :)","threadId":"5e03e664-d68c-45cf-bebc-2690e7100df5","user":{"idType":"aad","name":"Abid Nazir Guroo"}}],"nteract":{"transient":{"deleting":false}}},"id":"5ebeda1f-38c9-4e92-bd41-427657c73c88"},{"cell_type":"markdown","source":["#### Visual 9: Create a Correlation plot \n","A correlation plot is a useful tool for exploring the relationships among numerical variables in a dataset. It displays the data points for each pair of variables as a scatterplot, and also calculates the correlation coefficient for each pair. The correlation coefficient indicates how strongly and in what direction the variables are related. A positive correlation means that the variables tend to increase or decrease together, while a negative correlation means that they tend to move in opposite directions."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"5a0cd118-a464-4b87-a9db-40702de86ebe"},{"cell_type":"code","source":["cols_to_corr = ['trip_duration','fare_amount', 'passenger_count', 'trip_distance', 'extra', 'mta_tax',\n","       'improvement_surcharge', 'tip_amount', 'hour',\"dayofweek\"]\n","sns.heatmap(data = sampled_df[cols_to_corr].corr(),annot=True,fmt='.3f', cmap=\"Blues\")"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"43890ad0-a609-4e01-8f28-8c48292e1430"},{"cell_type":"markdown","source":["# Step 3: Train and register a machine learning model\n","In this Step you will learn to train a machine learning model to predict the total ride duration (trip_duration) of green taxi trips in New York City based on various factors such as pickup and drop-off locations, distance, date, time, number of passengers, and rate code.\n","\n","Once a model is trained, you will learn to register the trained model, and log hyperaparameters used and evaluation metrics using Fabric's native integration with the MLflow framework.\n","\n","[MLflow](https://mlflow.org/docs/latest/index.html) is an open source platform for managing the machine learning lifecycle with features like Tracking, Models, and Model Registry. MLflow is natively integrated with Fabric Data Science Experience."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"b0c00880-1d88-42ca-a17c-fdedceae0e86"},{"cell_type":"markdown","source":["#### Import mlflow and create an experiment to log the run\n","An experiment artifact in Microsoft Fabric is the primary unit of organization and control for all related machine learning runs. You can learn more about ML Experiment and various methods to create experiments in Fabric [here](https://learn.microsoft.com/en-us/fabric/data-science/machine-learning-experiment)."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"4f587955-5d66-415d-a6a8-9942bb1674f5"},{"cell_type":"code","source":["# Create Experiment to Track and register model with mlflow\n","import mlflow\n","print(f\"mlflow lbrary version: {mlflow.__version__}\")\n","EXPERIMENT_NAME = \"nyctaxi_trip_duration\"\n","mlflow.set_experiment(EXPERIMENT_NAME)"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{},"ms_comments":[],"nteract":{"transient":{"deleting":false}}},"id":"a8dce034-000a-42ee-bfc6-93185ba2fee3"},{"cell_type":"markdown","source":["#### Read Cleansed and prepared data from silver zone delta table"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"62b246f3-7e03-48af-9d5b-1f2d947e28b2"},{"cell_type":"code","source":["training_df = spark.read.table(f\"{SILVER_LAKEHOUSE}.{TRAINING_TABLE}\")                       "],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"dda2c478-957d-4f65-8a5c-79741c7e6621"},{"cell_type":"markdown","source":["#### Perform random split to get train and test datasets and define categorical and numeric features"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"692baef1-2e52-4307-a585-37028f71fbeb"},{"cell_type":"code","source":["TRAIN_TEST_SPLIT = [0.75, 0.25]\n","train_df, test_df = training_df.randomSplit(TRAIN_TEST_SPLIT, seed=SEED)\n","\n","# Cache the dataframes to improve the speed of repeatable reads\n","train_df.cache()\n","test_df.cache()\n","\n","print(f\"train set count:{train_df.count()}\")\n","print(f\"test set count:{test_df.count()}\")\n","\n","categorical_features = [\"timeBins\",\"VendorID\",\"weekDayName\",\"pickupHour\",\"RatecodeID\"]\n","numeric_features = ['passenger_count', \"trip_distance\"]"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{"6aed2927-81e3-4dc7-b7bb-30651024b627":{"end":{"column":73,"line":2},"start":{"column":68,"line":2},"text":"SEED)"}},"ms_comments":[{"createdDateUTC":1681362191287,"modifiedDateUTC":1681475134752,"replies":[],"status":"resolved","text":"F.Z. I would define all these hardcoded values as paramsa and log them to mlflow","threadId":"6aed2927-81e3-4dc7-b7bb-30651024b627","user":{"idType":"aad","name":"Abid Nazir Guroo"}}],"nteract":{"transient":{"deleting":false}}},"id":"bc0d2b8f-8b3f-4725-a95f-551fb73878e0"},{"cell_type":"markdown","source":["#### Define the steps to perform additional feature engineering and train the model using Spark ML pipelines and Microsoft SynapseML library\n","You can learn more about Spark ML pipelines [here](https://spark.apache.org/docs/latest/ml-pipeline.html), and SynapseML is documented [here](https://microsoft.github.io/SynapseML/docs/about/)\n","\n","The algorithm used for this exercise, [LightGBM](https://lightgbm.readthedocs.io/en/v3.3.2/) is a fast, distributed, high performance gradient boosting framework based on decision tree algorithms. It is an open source project developed by Microsoft and supports regression, classification and many other machine learning scenarios. Its main advantages are faster training speed, lower memory usage, better accuracy, and support for distributed learning."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"26a343eb-bedd-4cf9-b230-7e440648049f"},{"cell_type":"code","source":["from pyspark.ml.feature import OneHotEncoder, VectorAssembler, StringIndexer\n","from pyspark.ml import Pipeline\n","from synapse.ml.core.platform import *\n","from synapse.ml.lightgbm import LightGBMRegressor\n","\n","# Define a pipeline steps for training a LightGBMRegressor regressor model\n","def lgbm_pipeline(categorical_features,numeric_features, hyperparameters):\n","    # String indexer\n","    stri = StringIndexer(inputCols=categorical_features, \n","                        outputCols=[f\"{feat}Idx\" for feat in categorical_features]).setHandleInvalid(\"keep\")\n","    # encode categorical/indexed columns\n","    ohe = OneHotEncoder(inputCols= stri.getOutputCols(),  \n","                        outputCols=[f\"{feat}Enc\" for feat in categorical_features])\n","    \n","    # convert all feature columns into a vector\n","    featurizer = VectorAssembler(inputCols=ohe.getOutputCols() + numeric_features, outputCol=\"features\")\n","\n","    # Define the LightGBM regressor\n","    lgr = LightGBMRegressor(\n","        objective = hyperparameters[\"objective\"],\n","        alpha = hyperparameters[\"alpha\"],\n","        learningRate = hyperparameters[\"learning_rate\"],\n","        numLeaves = hyperparameters[\"num_leaves\"],\n","        labelCol=\"trip_duration\",\n","        numIterations = hyperparameters[\"iterations\"],\n","    )\n","    # Define the steps and sequence of the SPark ML pipeline\n","    ml_pipeline = Pipeline(stages=[stri, ohe, featurizer, lgr])\n","    return ml_pipeline\n"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"640ccaf5-a7fa-42d1-aa25-6beb10542b8e"},{"cell_type":"markdown","source":["#### Define Training Hyperparameters\n","Hyperparameters are the parameters that you can change to control how a machine learning model is trained. Hyperparameters can affect the speed, quality and accuracy of the model. Some common methods to find the best hyperparameters are by testing different values, using a grid or random search, or using a more advanced optimization technique.\n","The hyperparameters for the lightgbm model in this exercise have been pre-tuned using a distributed gridsearch run using [hyperopt](https://github.com/hyperopt/hyperopt)"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"8d8f0ac3-4032-4dd6-9d61-d16021f65f26"},{"cell_type":"code","source":["# Tuned hyperparameters for LightGBM Model\n","LGBM_PARAMS = {\"objective\":\"regression\",\n","    \"alpha\":0.08373361416254149,\n","    \"learning_rate\":0.0801709918703746,\n","    \"num_leaves\":92,\n","    \"iterations\":200}"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"edebb973-94d6-4ba1-abd6-9baca4e2e3f2"},{"cell_type":"markdown","source":["#### Fit the lightgbm pipeline with defined hyperparameters on the training dataframe and generate predictions on the test dataset"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"d9dfc966-7653-48d1-a944-cb5fbbafef21"},{"cell_type":"code","source":["lg_pipeline = lgbm_pipeline(categorical_features,numeric_features,LGBM_PARAMS)\n","lg_model = lg_pipeline.fit(train_df)\n","\n","# Get Predictions\n","lg_predictions = lg_model.transform(test_df)\n","## Caching predictions to run model evaluation faster\n","lg_predictions.cache()\n","print(f\"Prediction run for {lg_predictions.count()} samples\")"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{},"ms_comments":[{"createdDateUTC":1681587506535,"modifiedDateUTC":1681587506535,"replies":[],"status":"active","text":"we are using same features as experiment 1, then in Experiments UI we cannot show the diff in input schema, would be good to adjust one of the experiments to have slightly different feature so that we can highlight that feature","threadId":"4cf28ee7-af51-4c3b-9fd4-8963d086d05e","user":{"idType":"aad","name":"Fatemeh Zamanian"}}],"nteract":{"transient":{"deleting":false}}},"id":"9cb121a8-7952-464e-b1c1-c024eadc54e3"},{"cell_type":"markdown","source":["#### Compute Model Statistics for evaluating performance of the trained LightGBMRegressor model"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"b6ef7120-2d3b-4f79-b42d-670fb9a3cbd3"},{"cell_type":"code","source":["from synapse.ml.train import ComputeModelStatistics\n","import json\n","lg_metrics = ComputeModelStatistics(\n","    evaluationMetric=\"regression\", labelCol=\"trip_duration\", scoresCol=\"prediction\"\n",").transform(lg_predictions)\n","lg_metrics_dict = json.loads(lg_metrics.toJSON().first())\n","display(lg_metrics)"],"outputs":[],"execution_count":null,"metadata":{"collapsed":false,"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"1e5c983c-c9da-4d3f-9027-49e366de0123"},{"cell_type":"markdown","source":["#### Register the trained LightGBMRegressor model using MLflow log model utility\n","\n","Fabric Data science experience also supports Autologging which implifies tracking of machine learning model parameters, metrics, and items during training by extending MLflow autologging capabilities. In the below step we will use MLflow to explicitly log parameters, metrics and model artifacts. However, you can refer this doc on how to use Autologging in Fabric."],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"6ce02792-8093-429c-8748-5bd1a0dce424"},{"cell_type":"code","source":["from mlflow.models.signature import ModelSignature \n","from mlflow.types.utils import _infer_schema \n","\n","# Define a function to register a spark model\n","def register_spark_model(run, model, model_name,signature,metrics, hyperparameters):\n","        # log the model, parameters and metrics\n","        mlflow.spark.log_model(model, artifact_path = model_name, signature=signature, registered_model_name = model_name, dfs_tmpdir=\"Files/tmp/mlflow\") \n","        mlflow.log_params(hyperparameters) \n","        mlflow.log_metrics(metrics) \n","        model_uri = f\"runs:/{run.info.run_id}/{model_name}\" \n","        print(f\"Model saved in run{run.info.run_id}\") \n","        print(f\"Model URI: {model_uri}\")\n","        return model_uri\n","\n","# Define Signature object \n","sig = ModelSignature(inputs=_infer_schema(train_df.select(categorical_features + numeric_features)), \n","                     outputs=_infer_schema(train_df.select(\"trip_duration\"))) \n","\n","ALGORITHM = \"lightgbm\" \n","model_name = f\"{EXPERIMENT_NAME}_{ALGORITHM}\"\n","with mlflow.start_run() as run:\n","        print(f\"Active experiment run_id: {run.info.run_id}\")\n","        # Call model register function\n","        model_uri = register_spark_model(run = run,\n","                                        model = lg_model, \n","                                        model_name = model_name, \n","                                        signature = sig, \n","                                        metrics = lg_metrics_dict, \n","                                        hyperparameters = LGBM_PARAMS)\n"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"ms_comment_ranges":{},"ms_comments":[{"createdDateUTC":1681588858925,"modifiedDateUTC":1681588858925,"replies":[],"status":"active","text":"this solution won't work (I tested and we still get 4 runs instead of 2) you need to instantiate run before pipeline definition, ie before where you call lgbm_pipeline function","threadId":"68ad1fef-12c3-4a79-bf5b-200c15b0b075","user":{"idType":"aad","name":"Fatemeh Zamanian"}},{"createdDateUTC":1681587581662,"modifiedDateUTC":1681587581662,"replies":[],"status":"active","text":"if we change the features btw experiment 1 and 2 (above comment) then we need to update signature for model 1 vs model 2","threadId":"53ec17b2-5375-402a-b283-710b546d124f","user":{"idType":"aad","name":"Fatemeh Zamanian"}},{"createdDateUTC":1681591462894,"modifiedDateUTC":1681591462894,"replies":[],"status":"active","text":"nitpicking, the difference btw performance metrics here btw the two run (after bug fix) is negligible. essentially we cannot tell which model is better - I think we need to log performance metrics for training data and test data separately and also making sure the diff btw the two experiments is obvious,  perhaps changing features btw the two experiments can lead to this kind of results","threadId":"0ba78138-7874-483e-b118-a90a2affce56","user":{"idType":"aad","name":"Fatemeh Zamanian"}},{"createdDateUTC":1681587424103,"modifiedDateUTC":1681587424103,"replies":[],"status":"active","text":"here we should log lg_metricstn_dict (for tuned params)","threadId":"f947d25a-415c-40d2-9be7-8675e6e8a9ba","user":{"idType":"aad","name":"Fatemeh Zamanian"}}],"nteract":{"transient":{"deleting":false}}},"id":"66254c43-00d6-4f0d-b234-b798d621306e"},{"attachments":{"f9264106-1696-46c1-bd5c-10ccd129db13.png":{"image/png":"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"}},"cell_type":"markdown","source":["##### Once a Machine Learning model is registered you can view the ML Model artifact in your workspace UI and open the model to view and interact with various ML Model properties like versions, model files, metrics, hyperparameters and more.\n","\n","![MLModelArtifact.png](attachment:f9264106-1696-46c1-bd5c-10ccd129db13.png)"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"8c20c2a6-67f4-4753-92b9-f7584a2bba50"},{"cell_type":"markdown","source":["# Step 5: Perform batch scoring and save predictions to lakehouse"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"e1aef894-3cf6-4443-a771-237232a7459a"},{"cell_type":"markdown","source":["#### Read cleansed data from lakehouse for the year 2023 to generate predictions"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"67ab614e-3d8d-446a-8253-e23e96f80c2c"},{"cell_type":"code","source":["input_df = spark.read.table(f\"{SILVER_LAKEHOUSE}.{EVAL_TABLE}\")"],"outputs":[],"execution_count":null,"metadata":{"collapsed":false,"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"54af4251-1fe8-4aa1-b49b-2d8787861072"},{"cell_type":"markdown","source":["#### Get the trained and registered model to generate predictions\n","In the below cell we will be demonstrating how to load a Fabric model with just a model name, version and the model artifact will be loaded into the notebook session and is ready for generating prediction.\n"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"2e10558b-0f14-4311-b2bf-eb91bf801279"},{"cell_type":"code","source":["import mlflow\n","## Define model_uri to fetch the model\n","version = \"latest\"\n","model_uri = f\"models:/{model_name}/{version}\"\n","print(f\"Loading {version} version of the registered model - {model_name}\")\n","loaded_model = mlflow.spark.load_model(model_uri)"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"6251be60-e619-404a-8873-428e2c7d7d03"},{"cell_type":"markdown","source":["#### Run model transform on the input dataframe to generate predictions and remove unnecessary vector features created for model training\n","In the below cell we using the registered and loaded model's transform function to generate predictions on data. Other methodologies of invoking BATCH predict for a ML Model are documented [here](https://learn.microsoft.com/en-us/fabric/data-science/model-scoring-predict)"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"710d5695-cf34-4a0c-8db6-a376184790fc"},{"cell_type":"code","source":["# Generate predictions by applying model transform on the input dataframe\n","predictions = loaded_model.transform(input_df)\n","cols_toremove = ['storeAndFwdFlagIdx', 'timeBinsIdx', 'VendorIDIdx', 'payment_typeIdx', 'VendorIDEnc',\n"," 'rateCodeIDEnc', 'payment_typeEnc', 'weekDayEnc', 'pickupHourEnc', 'storeAndFwdFlagEnc', 'timeBinsEnc', 'features','weekDayNameIdx',\n"," 'pickupHourIdx', 'rateCodeIDIdx', 'weekDayNameEnc']\n","output_df = predictions.withColumnRenamed(\"prediction\", \"predictedtrip_duration\").drop(*cols_toremove)"],"outputs":[],"execution_count":null,"metadata":{"collapsed":false,"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"ca7b1e9f-1d8f-4cd0-a3c2-6e3e4a2b8e1d"},{"cell_type":"markdown","source":["#### Save predictions to gold layer lakehouse delta table"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"2a27ef65-6a17-42f5-a257-c28cc6b85209"},{"cell_type":"code","source":["table_name = \"nyctaxi_green_pred\"\n","output_df.write.mode(\"overwrite\").format(\"delta\").saveAsTable(f\"{GOLD_LAKEHOUSE}.{PREDICT_TABLE}\")\n","print(f\"Output Predictions saved to delta table: {table_name}\")"],"outputs":[],"execution_count":null,"metadata":{"jupyter":{"outputs_hidden":false,"source_hidden":false},"nteract":{"transient":{"deleting":false}}},"id":"dab8dacb-5f93-48b1-a1f9-a3fdff2d0b5d"},{"cell_type":"markdown","source":["#### Preview and explore predicted dataframe using SparkSQL"],"metadata":{"nteract":{"transient":{"deleting":false}}},"id":"792afbeb-484c-4a5d-9b2a-5b6a06e102e6"},{"cell_type":"code","source":["%%sql\n","SELECT * FROM ${GOLD_LAKEHOUSE}.${PREDICT_TABLE} LIMIT 20"],"outputs":[],"execution_count":null,"metadata":{"collapsed":false,"jupyter":{"outputs_hidden":false,"source_hidden":false},"microsoft":{"language":"sparksql"},"nteract":{"transient":{"deleting":false}}},"id":"6728fdf2-b12b-492b-a9d8-2ab98ae27fd5"}],"metadata":{"kernel_info":{"name":"synapse_pyspark"},"kernelspec":{"name":"synapse_pyspark","language":"Python","display_name":"Synapse PySpark"},"language_info":{"name":"python"},"microsoft":{"language":"python","ms_spell_check":{"ms_spell_check_language":"en"}},"notebook_environment":{},"nteract":{"version":"nteract-front-end@1.0.0"},"widgets":{"application/vnd.jupyter.widget-state+json":{"version_major":2,"version_minor":0,"state":{"1a343654d6b34a1cbf421298c798c6cf":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"59a69694d2b5476dacd107e9394afa05":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"781e22439f9b49a1aae4870ad668873c":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"4eee0bbf4ccd4a34b29667d7d3de358a":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"4535e469548c43989a880f9b07d95842":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_5905670cf62242f6ab49060f17d9aee4","style":"IPY_MODEL_bf6feecc7e074e69bcbdfd9032af6eaf"}},"c0b39ba3da274e34afddcca9cd3e1754":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_ad7b8726accd447e977bb6b96378d89c","layout":"IPY_MODEL_be250468d29d4ca38650845e797cba59"}},"d9bfb89da0334c58b5b498da70a1a16c":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_52e7f2eb78614879b1183f4d48fc4280","style":"IPY_MODEL_2c31824268364eaea21b31ff6d873a3d"}},"1b202cc53ec344b08ea7104fdf763b40":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"5002da99e375492db73645644d2dca7e":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 48.29it/s]","layout":"IPY_MODEL_4c90473e1cf1480a9853879ffacbf983","style":"IPY_MODEL_ded08fe5d75b4c1ead63ff85b3a317f7"}},"299dd5b591ad4cfead5e10598f867349":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"fe6252d752f34129a1fb0a88be5fa999":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 600.48it/s]","layout":"IPY_MODEL_010b51d2a25d49058231b0bcfc2906d8","style":"IPY_MODEL_30f3e9c527f5480885c71ca6929df3d5"}},"47f1a87c4257436a9802e0eb1c8d2486":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_1e9053bd7ead46198f752729e8d209f4","layout":"IPY_MODEL_d843bf8d7d2f4f219dadbf925ed66ce8"}},"c11dd3c924b34acfbd1eaf701e48be8b":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_924a01d75ed44aa8b01261d67bdb217c","layout":"IPY_MODEL_3645cbe95c7647089c971bb090d1ac2a"}},"6b877960a42448aa938dfa07f63b8f4a":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"d843bf8d7d2f4f219dadbf925ed66ce8":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"74f745f324fd420a87e39cc0f931e84c":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"1daa2ca88db34a848759b488fbf96103":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"f3c9aba544da4475bda090a228ff7a18":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_a05e61e9c07a426682956053b5eaae2a","IPY_MODEL_62f8cd218a6547f888ee5e3343deb947","IPY_MODEL_3f40333ee90046d2ac6f237fe865159b"],"layout":"IPY_MODEL_84349fc5f531400ca2331fa6392b7438"}},"e096073daa2749ebb28d6cd414d8da3b":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"92943c23efba45f787e27a7c4672139d":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"1e46f6a7287e4df98d7c4595aa636147":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"365142e0bb2f42cba1d54cfd6097a75d":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"bf99735e079f457e90f2fbaf49464fc7":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_b2701ea3ab1d48028e88da41ce2cac3e","style":"IPY_MODEL_de12313ce25a4ecbbf59f3d73dd1248a"}},"01922d7dcead4145ba69efac79cb1531":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"06a55972cbef4e84afb521d0fb92f9d7":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_c7eab6bb86744471aee353013e687b6d","layout":"IPY_MODEL_78e0611e45724c0a9cfad87ce0770104"}},"f8dcc84ebff14ef78f87d17dd2aca4a3":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"81867727d3f444d0ad46bc5ecac8c18a":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"450a682b8071428aba9e6542f834c671":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"ded08fe5d75b4c1ead63ff85b3a317f7":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"3f40333ee90046d2ac6f237fe865159b":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 19.16it/s]","layout":"IPY_MODEL_51b5befb4864415e9f311a665fdaec0d","style":"IPY_MODEL_f6ecea0fd3fe49fcac187b90351242c9"}},"c91ad78b86e64ed6ab474312e22aaa2d":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_10f08b4cdac74cdf8cb7739616abc7bd","layout":"IPY_MODEL_7a8c79ea5fb24d769b3a77e7ff2f32f4"}},"e532ab34a06f4adb9a52d8d0de38f76f":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"d1dbd80052fb491982ea5bbb36cd5e9e":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"e8848e8e57f6405ba4717382eb3d6daf":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"bbab4c196f164d3e8310daa4f5fda6e2":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_b4da54878f1642198fedda50c5314e39","IPY_MODEL_36ef87e55e0a44ce95d6932017bbc412","IPY_MODEL_85d469d87e844487a6f6c979992fe959"],"layout":"IPY_MODEL_63414588541c44d2b5f28bb20f572375"}},"5e84b3b47e0f4eb9a87f96d40f2cc8db":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_a2441f76710f4e7aa2b881c992666fac","layout":"IPY_MODEL_cb7ee6ddaf734a3db0ba10c9b9fa722e"}},"ec0698d524284e5ab8d3dbcc7c7f2f42":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"f68b201a011b4a2a9f129b8a407cb173":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"bbef7c29300e4150bd44728d3c2ac6f6":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"02501f4cd8fb43ca80df516233274ab9":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"c656ab13aa8e49e18a6935153fe77169":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"52e7f2eb78614879b1183f4d48fc4280":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"203def9624184367a59f144013e854ff":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"5771b53413a3460495b9effb532954e1":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_bb2509052f154c0db814731a6dfc5f3c","style":"IPY_MODEL_4439cfcf4e35493392e94d64f3509e35"}},"30f3e9c527f5480885c71ca6929df3d5":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"63414588541c44d2b5f28bb20f572375":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"de12313ce25a4ecbbf59f3d73dd1248a":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"84349fc5f531400ca2331fa6392b7438":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"63b56f091f2341dca98e22ddbcdba393":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"5905670cf62242f6ab49060f17d9aee4":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"3645cbe95c7647089c971bb090d1ac2a":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"c7a7afebc69c4bd08ad9701522b0ab75":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_4535e469548c43989a880f9b07d95842","IPY_MODEL_fd9dd023984d4c8a8348e3ad6921669b","IPY_MODEL_be6e12a0801043dea672f94f6eb104e0"],"layout":"IPY_MODEL_8303afdeb96b4602aa8d33284e221bd3"}},"aabe7cfc6a904033af298239a0351017":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"010b51d2a25d49058231b0bcfc2906d8":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"ec1ba08282144547b27fcb8d7efb8f0a":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_d9bfb89da0334c58b5b498da70a1a16c","IPY_MODEL_18508e973b954634b65d40fb587f6d64","IPY_MODEL_12b983464db94ab7b7ac70e9814146a8"],"layout":"IPY_MODEL_1daa2ca88db34a848759b488fbf96103"}},"44d0cb7709da4595891b07bb2facaa6e":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"78e0611e45724c0a9cfad87ce0770104":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"6ed157651e3c4fbd89b0a12df2de5284":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_509374bea38e44dd9c8ca60788571b88","IPY_MODEL_06a55972cbef4e84afb521d0fb92f9d7","IPY_MODEL_2a93b40c219a4256b29694610ec6ed81"],"layout":"IPY_MODEL_92943c23efba45f787e27a7c4672139d"}},"1e0a08a0bc2f43a4a78b96f74ccb24d0":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"7a8c79ea5fb24d769b3a77e7ff2f32f4":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"2c31824268364eaea21b31ff6d873a3d":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"bf6feecc7e074e69bcbdfd9032af6eaf":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"03749e74a1f148d9a03d9708692c4f59":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"dd58176281ff4304945de2abb84db671":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"539a836fec994bd98b7a6164e49b6c09":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 623.27it/s]","layout":"IPY_MODEL_1e0a08a0bc2f43a4a78b96f74ccb24d0","style":"IPY_MODEL_a5a7e4f48e1b41fab2ff0610b92d33ee"}},"924a01d75ed44aa8b01261d67bdb217c":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"8303afdeb96b4602aa8d33284e221bd3":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"18a7249e05e4406da782adaf9a351a2e":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_c16255d1ed594965913a9106b261605b","IPY_MODEL_c0b39ba3da274e34afddcca9cd3e1754","IPY_MODEL_e8b442dd2dc94e84a3a578088018f4ec"],"layout":"IPY_MODEL_1a343654d6b34a1cbf421298c798c6cf"}},"78d92e7f4815412abe1db781cf3cde8c":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"e3f09501e37e4545b7dc5296b23efa3d":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_e48a371c1353408f9a545994471b9a40","IPY_MODEL_9cc9976105c14234b91d64a512f48566","IPY_MODEL_5002da99e375492db73645644d2dca7e"],"layout":"IPY_MODEL_e0911aae9ec6498585d368530bff9916"}},"f7b4ff4910584b19937b9788a26cdda0":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"242948a464e641a49de605ae7887a481":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_432e3f4defc1414dbd8b3f96612b68a9","layout":"IPY_MODEL_a436d3952cd24929a1a7aea3580e182f"}},"e5c628067891436a8af2123652f00a5b":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"0a644df5fb894a18825eb8d58b435a7b":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_01c0005d12ba49b9b42bed3cb7e1ad59","IPY_MODEL_c91ad78b86e64ed6ab474312e22aaa2d","IPY_MODEL_539a836fec994bd98b7a6164e49b6c09"],"layout":"IPY_MODEL_83cb8f61d3d14faa8ad2cdb7e563ce0a"}},"b33c305409ca475e8f7da6b993073fca":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_6bf0de6396de4ba1a96e274794f31828","style":"IPY_MODEL_3fd347d07b8f4588859d81eec8568443"}},"7d07f76b4932475987d295b70092cb64":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"a5a1f985cb314f9da7dd46dba0dc73cf":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"c1c42d75bd1d44e69f9aa7deb2a21a87":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"12b983464db94ab7b7ac70e9814146a8":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 560.98it/s]","layout":"IPY_MODEL_95e0f03d08074a56a57dae335c9ce14f","style":"IPY_MODEL_f8dcc84ebff14ef78f87d17dd2aca4a3"}},"67aef0c3371347f7ab3d6d1872a0fcf0":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_ec0698d524284e5ab8d3dbcc7c7f2f42","style":"IPY_MODEL_e5c628067891436a8af2123652f00a5b"}},"a2441f76710f4e7aa2b881c992666fac":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"5ab228333c094bab942e29f7ee8f385c":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 118.16it/s]","layout":"IPY_MODEL_59a69694d2b5476dacd107e9394afa05","style":"IPY_MODEL_bbef7c29300e4150bd44728d3c2ac6f6"}},"9da132b5f541405da6411a464010a780":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"01c0005d12ba49b9b42bed3cb7e1ad59":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_203def9624184367a59f144013e854ff","style":"IPY_MODEL_79939577c16248e0a7a0ef281c6d7304"}},"e8b442dd2dc94e84a3a578088018f4ec":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 686.62it/s]","layout":"IPY_MODEL_74f745f324fd420a87e39cc0f931e84c","style":"IPY_MODEL_b6b1f6da91824064a5ebb25a199ed2f6"}},"f6ecea0fd3fe49fcac187b90351242c9":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"432e3f4defc1414dbd8b3f96612b68a9":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"b4da54878f1642198fedda50c5314e39":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_63b56f091f2341dca98e22ddbcdba393","style":"IPY_MODEL_450a682b8071428aba9e6542f834c671"}},"ad7b8726accd447e977bb6b96378d89c":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"9cc9976105c14234b91d64a512f48566":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_03749e74a1f148d9a03d9708692c4f59","layout":"IPY_MODEL_aabe7cfc6a904033af298239a0351017"}},"a05e61e9c07a426682956053b5eaae2a":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_efab58bfe70442dab167a43372c5a9a3","style":"IPY_MODEL_4c7de1c889494f1da8bd3e16b0d87ac4"}},"b6310f692e5f497b9c7da89c75137421":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"fbb52ba102cb42389367aa92e1435b2f":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 23.45it/s]","layout":"IPY_MODEL_e457c146cac84d99be5b622ed05e4394","style":"IPY_MODEL_19fb1b3b58ca4331aa2f0c7d1c72067b"}},"efab58bfe70442dab167a43372c5a9a3":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"c16255d1ed594965913a9106b261605b":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_44d0cb7709da4595891b07bb2facaa6e","style":"IPY_MODEL_299dd5b591ad4cfead5e10598f867349"}},"19fb1b3b58ca4331aa2f0c7d1c72067b":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"c5036306f1ba457880bec252af0a553f":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"6bf0de6396de4ba1a96e274794f31828":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"36ef87e55e0a44ce95d6932017bbc412":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_7d07f76b4932475987d295b70092cb64","layout":"IPY_MODEL_a5a1f985cb314f9da7dd46dba0dc73cf"}},"18508e973b954634b65d40fb587f6d64":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_d1dbd80052fb491982ea5bbb36cd5e9e","layout":"IPY_MODEL_81867727d3f444d0ad46bc5ecac8c18a"}},"34315ce07865471c9728e0aaadd75778":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_1b202cc53ec344b08ea7104fdf763b40","style":"IPY_MODEL_4eee0bbf4ccd4a34b29667d7d3de358a"}},"4439cfcf4e35493392e94d64f3509e35":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"0443d40f962147669ac1170b58b82778":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"4c7de1c889494f1da8bd3e16b0d87ac4":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"62f8cd218a6547f888ee5e3343deb947":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_e532ab34a06f4adb9a52d8d0de38f76f","layout":"IPY_MODEL_365142e0bb2f42cba1d54cfd6097a75d"}},"84199d8e58084929af9869c403bd938d":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"41a7d6063756403fad55e5ac976f088d":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"55e13aeddca442fea115944425d6967d":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"e48a371c1353408f9a545994471b9a40":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_dd58176281ff4304945de2abb84db671","style":"IPY_MODEL_02501f4cd8fb43ca80df516233274ab9"}},"1e9053bd7ead46198f752729e8d209f4":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"85d469d87e844487a6f6c979992fe959":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 164.06it/s]","layout":"IPY_MODEL_4620e61610c34b98848433f6e33bb3a4","style":"IPY_MODEL_c656ab13aa8e49e18a6935153fe77169"}},"c7eab6bb86744471aee353013e687b6d":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"c70775cd03d1402a9f77dff90ee9c29a":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_b6310f692e5f497b9c7da89c75137421","layout":"IPY_MODEL_e8848e8e57f6405ba4717382eb3d6daf"}},"b2701ea3ab1d48028e88da41ce2cac3e":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"2a93b40c219a4256b29694610ec6ed81":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 607.77it/s]","layout":"IPY_MODEL_f7b4ff4910584b19937b9788a26cdda0","style":"IPY_MODEL_7efb86a050eb4191b24fca56edd06cee"}},"458d434a9bcd4bbda27c195bfea65135":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00,  8.27it/s]","layout":"IPY_MODEL_781e22439f9b49a1aae4870ad668873c","style":"IPY_MODEL_cfa589d8ec8843ea97486fc0f0462c91"}},"51b5befb4864415e9f311a665fdaec0d":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"509374bea38e44dd9c8ca60788571b88":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_1e46f6a7287e4df98d7c4595aa636147","style":"IPY_MODEL_a74f8651d78e4670945362bd69185eb1"}},"fd9dd023984d4c8a8348e3ad6921669b":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":27,"max":27,"bar_style":"success","style":"IPY_MODEL_8edc568500264d00bfde5d41f1eebf64","layout":"IPY_MODEL_9da132b5f541405da6411a464010a780"}},"7973234ed29a4bfcb64cf2d53c4fd52f":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_5771b53413a3460495b9effb532954e1","IPY_MODEL_47f1a87c4257436a9802e0eb1c8d2486","IPY_MODEL_5ab228333c094bab942e29f7ee8f385c"],"layout":"IPY_MODEL_6b877960a42448aa938dfa07f63b8f4a"}},"10f08b4cdac74cdf8cb7739616abc7bd":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"e2929fc9cab04f2895ef07c35bb7b987":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"cfa589d8ec8843ea97486fc0f0462c91":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"b1a812d49ee84db3a1528b7a93cb2a70":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"e457c146cac84d99be5b622ed05e4394":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"816929939e3b4355836902ff97f1b9cf":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"95e0f03d08074a56a57dae335c9ce14f":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"da5312f975f04350a152f5fce2eed9b1":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 610.14it/s]","layout":"IPY_MODEL_c5036306f1ba457880bec252af0a553f","style":"IPY_MODEL_e2929fc9cab04f2895ef07c35bb7b987"}},"4620e61610c34b98848433f6e33bb3a4":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"e0911aae9ec6498585d368530bff9916":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"a436d3952cd24929a1a7aea3580e182f":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"be6e12a0801043dea672f94f6eb104e0":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 27/27 [00:00&lt;00:00, 557.19it/s]","layout":"IPY_MODEL_e096073daa2749ebb28d6cd414d8da3b","style":"IPY_MODEL_01922d7dcead4145ba69efac79cb1531"}},"a5a7e4f48e1b41fab2ff0610b92d33ee":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"532c346b654d4e14a0ad776b465bce7d":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_2e71422d38d04a448564af7ad39df5e6","IPY_MODEL_c11dd3c924b34acfbd1eaf701e48be8b","IPY_MODEL_af80b6c71cf842fdbd6a48d3db1a886a"],"layout":"IPY_MODEL_816929939e3b4355836902ff97f1b9cf"}},"b6b1f6da91824064a5ebb25a199ed2f6":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"577199ae0cd449819b7685bee39144f3":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"8edc568500264d00bfde5d41f1eebf64":{"model_name":"ProgressStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":""}},"be250468d29d4ca38650845e797cba59":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"a74f8651d78e4670945362bd69185eb1":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"3fd347d07b8f4588859d81eec8568443":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"cc5b262969214722bf979dbdbd93bdb5":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_b33c305409ca475e8f7da6b993073fca","IPY_MODEL_5e84b3b47e0f4eb9a87f96d40f2cc8db","IPY_MODEL_da5312f975f04350a152f5fce2eed9b1"],"layout":"IPY_MODEL_577199ae0cd449819b7685bee39144f3"}},"79939577c16248e0a7a0ef281c6d7304":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"7efb86a050eb4191b24fca56edd06cee":{"model_name":"HTMLStyleModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"description_width":"","font_size":null,"text_color":null}},"bb2509052f154c0db814731a6dfc5f3c":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"83cb8f61d3d14faa8ad2cdb7e563ce0a":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"12b0e31d204848949d2ce925e6645a76":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"4c90473e1cf1480a9853879ffacbf983":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"2592fde218fb45e48900822f8dc36f09":{"model_name":"FloatProgressModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":1,"max":1,"bar_style":"success","style":"IPY_MODEL_84199d8e58084929af9869c403bd938d","layout":"IPY_MODEL_f68b201a011b4a2a9f129b8a407cb173"}},"56b05b9bb47f41af841d9c9369c1265c":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_34315ce07865471c9728e0aaadd75778","IPY_MODEL_2592fde218fb45e48900822f8dc36f09","IPY_MODEL_458d434a9bcd4bbda27c195bfea65135"],"layout":"IPY_MODEL_55e13aeddca442fea115944425d6967d"}},"af80b6c71cf842fdbd6a48d3db1a886a":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":" 1/1 [00:00&lt;00:00, 162.99it/s]","layout":"IPY_MODEL_12b0e31d204848949d2ce925e6645a76","style":"IPY_MODEL_c1c42d75bd1d44e69f9aa7deb2a21a87"}},"dcb6623f40944473b9531dcd97154399":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_67aef0c3371347f7ab3d6d1872a0fcf0","IPY_MODEL_242948a464e641a49de605ae7887a481","IPY_MODEL_fe6252d752f34129a1fb0a88be5fa999"],"layout":"IPY_MODEL_41a7d6063756403fad55e5ac976f088d"}},"2e71422d38d04a448564af7ad39df5e6":{"model_name":"HTMLModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"value":"Downloading artifacts: 100%","layout":"IPY_MODEL_b1a812d49ee84db3a1528b7a93cb2a70","style":"IPY_MODEL_0443d40f962147669ac1170b58b82778"}},"cb7ee6ddaf734a3db0ba10c9b9fa722e":{"model_name":"LayoutModel","model_module":"@jupyter-widgets/base","model_module_version":"2.0.0","state":{}},"1f7e773b494d4ebc888b9bb21fdcd7fb":{"model_name":"HBoxModel","model_module":"@jupyter-widgets/controls","model_module_version":"2.0.0","state":{"children":["IPY_MODEL_bf99735e079f457e90f2fbaf49464fc7","IPY_MODEL_c70775cd03d1402a9f77dff90ee9c29a","IPY_MODEL_fbb52ba102cb42389367aa92e1435b2f"],"layout":"IPY_MODEL_78d92e7f4815412abe1db781cf3cde8c"}}}}},"description":"","synapse_widget":{"state":{},"version":"0.1"},"save_output":true,"spark_compute":{"compute_id":"/trident/default","session_options":{"conf":{},"enableDebugMode":false}},"dependencies":{"lakehouse":{}}},"nbformat":4,"nbformat_minor":5}