Last active
June 12, 2017 11:08
-
-
Save jay-trivedi/bd5f40dc28356c43995a3743e84a73cf to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#Random Forest | |
seed = 1 | |
depth_range = range(1, 30,1) | |
acc_vs_depth_result_rf = {"depth": [],\ | |
"train_acc": [], | |
"valid_acc": [], | |
"top_feature": [], | |
"second_feature": [], | |
"third_feature": []} | |
for depth in depth_range: | |
model = H2ORandomForestEstimator(model_id="model", \ | |
sample_rate=1, \ | |
ntrees=200, \ | |
max_depth=depth, \ | |
seed=seed) | |
model.train(x=x, y=y, training_frame=train) | |
predict_valid = model.predict(valid[x]) | |
predict_train = model.predict(train[x]) | |
t = predict_train["predict"].cbind(train["SalePrice"]).as_data_frame() | |
v = predict_valid["predict"].cbind(valid["SalePrice"]).as_data_frame() | |
acc_vs_depth_result_rf["depth"].append(depth) | |
acc_vs_depth_result_rf["valid_acc"].append(mean_squared_error(y_true = v.SalePrice, y_pred = v.predict)) | |
acc_vs_depth_result_rf["train_acc"].append(mean_squared_error(y_true = t.SalePrice, y_pred = t.predict)) | |
acc_vs_depth_result_rf["top_feature"].append(model.varimp()[0][0]) | |
acc_vs_depth_result_rf["second_feature"].append(model.varimp()[1][0]) | |
acc_vs_depth_result_rf["third_feature"].append(model.varimp()[2][0]) | |
#Converting results to a DataFrame | |
acc_vs_depth_result_df_rf = pd.DataFrame(acc_vs_depth_result_rf) | |
cols = ["depth", "train_acc", "valid_acc", "top_feature", "second_feature", "third_feature"] | |
acc_vs_depth_result_df_rf = acc_vs_depth_result_df_rf[cols] | |
acc_vs_depth_result_df_rf | |
#Plotting results | |
fig = plt.figure(figsize=(10, 7)) | |
plt.plot(acc_vs_depth_result_df_rf.depth, acc_vs_depth_result_df_rf.train_acc, label="train MSE (RF)") | |
plt.plot(acc_vs_depth_result_df_rf.depth, acc_vs_depth_result_df_rf.valid_acc, label="validation MSE (RF)") | |
plt.plot(acc_vs_depth_result_df.depth, acc_vs_depth_result_df.train_acc, label="train MSE (DT)") | |
plt.plot(acc_vs_depth_result_df.depth, acc_vs_depth_result_df.valid_acc, label="validation MSE (DT)") | |
plt.legend(loc='upper left', frameon=False) | |
plt.xlabel('Tree Depth') | |
plt.ylabel('MSE') | |
plt.savefig("figures/House_pricing_RF.png") | |
plt.show() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment