search

py-why / EconML

ALICE (Automated Learning and Intelligence for Causation and Economics) is a Microsoft Research project aimed at applying Artificial Intelligence concepts to economic decision making. One of its goals is to build a toolkit that combines state-of-the-art machine learning techniques with econometrics in order to bring automation to complex causal inference problems. To date, the ALICE Python SDK (econml) implements orthogonal machine learning algorithms such as the double machine learning work of Chernozhukov et al. This toolkit is designed to measure the causal effect of some treatment variable(s) t on an outcome variable y, controlling for a set of features x.
https://www.microsoft.com/en-us/research/project/alice/
Other
3.74k stars 708 forks source link

Problem fitting DeepIV #572

Open juandavidgutier opened 2 years ago

juandavidgutier commented 2 years ago

Hello,

I am new with DeepIV, and I am trying to fit a model of DeepIV, however, I get this error message: TypeError: 'NoneType' object is not subscriptable, and I don't know how to interpretate it.

Here is my code and dataset Data_DeepIV.csv

Thanks for your cooperation

`import pandas as pd import scipy.stats as stats from econml.iv.nnet import DeepIV import keras

import data

data_col = pd.read_csv("D:/clases/UDES/articulo leishmaniasis/causal_inference/dowhy/Data_DeepIV.csv", encoding='latin-1') data_col = data_col.dropna()

subset

Colombia_temp = data_col[['incidence100k', 'Temperature', 'soi', 'Code_DANE']]

transform variables to sd units

Colombia_temp.Temperature = stats.zscore(Colombia_temp.Temperature)
Colombia_temp.soi = stats.zscore(Colombia_temp.soi)

y = Colombia_temp.incidence100k y.to_numpy() x = Colombia_temp.Code_DANE x.to_numpy() z = Colombia_temp.soi z.to_numpy() t = Colombia_temp.Temperature t.to_numpy()

Defining the neural network models

treatment_model = keras.Sequential([keras.layers.Dense(128, activation='relu', input_shape=(2,)), keras.layers.Dropout(0.17), keras.layers.Dense(64, activation='relu'), keras.layers.Dropout(0.17), keras.layers.Dense(32, activation='relu'), keras.layers.Dropout(0.17)])

response_model = keras.Sequential([keras.layers.Dense(128, activation='relu', input_shape=(2,)), keras.layers.Dropout(0.17), keras.layers.Dense(64, activation='relu'), keras.layers.Dropout(0.17), keras.layers.Dense(32, activation='relu'), keras.layers.Dropout(0.17), keras.layers.Dense(1)])

Instantiate the DeepIV class using these models.

keras_fit_options = { "epochs": 30, "validation_split": 0.1, "callbacks": [keras.callbacks.EarlyStopping(patience=2, restore_best_weights=True)]}

deepIvEst = DeepIV(n_components = 10, m = lambda z, x : treatment_model(keras.layers.concatenate([z,x])), h = lambda t, x : response_model(keras.layers.concatenate([t,x])),
n_samples = 1, use_upper_bound_loss = False, n_gradient_samples = 1, optimizer='adam', first_stage_options = keras_fit_options, second_stage_options = keras_fit_options)

Fitting and predicting using the model

deepIvEst.fit(Y=y,T=t,X=x,Z=z, inference="bootstrap")

ate

deepIvEst.ate_inference()

ci ate

deepIvEst.ate_interval()

dowhy wrapper

deepIvEst = deepIvEst.dowhy

placebo

placebo_temp = deepIvEst.refute_estimate(method_name="placebo_treatment_refuter", placebo_type="permute", num_simulations=5) print(placebo_temp)

`

kbattocchi commented 2 years ago

Could you provide the full stack trace, as well as the output of pip list, please?