All Questions
171 questions
0
votes
0
answers
66
views
Differences on features extraction + logistic regression vs features extraction + dense layer with softmax
I have a small database of images for a classification problem, therefore I opt for the transfer learning methodology. I started with the classic approach in Tensorflow:
A pretrained ResNet50 on the ...
- 155
0
votes
0
answers
54
views
Unable to stack trained models with scickit learn 1.4.1.post1
I have four trained models. 2 VotingClassifier models and 2 StackingClassifier models. They are already trained and saved in joblib format. I can even load them and use them for prediction (in prod, ...
1
vote
1
answer
92
views
String "loss" being passed to metric_name() instead of metric name
I'm following a tutorial, but I'm getting the error:
ValueError: Could not interpret metric identifier: loss
at:
\keras\src\metrics\__init__.py:205, in get(identifier)
It seems like in the wrapper.py ...
- 41
0
votes
0
answers
105
views
Why is my machine learning model always predicting the same wrong answer even when prediction is already in my dataset?
The goal is for my Tensorflow machine learning model to predict the Type of any Beyblade preset combination. A Beyblade preset is made up of 4 parts - Energy Ring, Fusion Wheel, Spin Track, ...
- 11
0
votes
1
answer
91
views
Problem with identical metrics results after machine learning [closed]
When trying machine learning on a dataset, I got the same results for metrics such as accuracy and F-score on different machine learning algorithms.
I have a dataset on which I trained my chosen ...
0
votes
1
answer
45
views
validation extraction using intel image classification
I'm trying to make a machine learning model that operates validation extraction to distinguish whether it's bright or dark.
I used the dataset "intel image classification" from kaggle.
...
0
votes
1
answer
51
views
Stratified grouped 5 fold cross validation issue
I have clinical data, already preprocessed (ECG and PCG) and I want to implement a stratified grouped 5-fold cross-validation, where I have one split/fold for testing, 1 split/fold for validation, and ...
1
vote
1
answer
106
views
Explaining ROC AUC/PR AUC Disparity Between `sklearn` and `tensorflow`
The code below generates a dummy classification problem and reports ROC AUC and PR AUC implemented both in sklearn and tensorflow. For some reason I find that each of the two metrics is quite ...
- 97
0
votes
1
answer
50
views
How can I use the relationship one data series to predict the outcome of a second data series?
I have two data sets, series1 and series2. series1 is complete, and series2 is incomplete. I want to use series1 to predict the missing values in series2.
data= [2, 4, 8, 16, 32, 64, 128, 256, 512, ...
- 336
0
votes
1
answer
84
views
Calculating Metrics (Recall, Precision, F1) in Tensorflow
I use one of the pre-trained sentence-encoder from tensorflow and now the plan is to evaluate the metrics (recall, precision and F1) based on my input and the corresponding embeddings. The input of my ...
- 47
0
votes
1
answer
191
views
Inverse Standardization of predicted values
I've dataset with two columns. At first I've split the data into train, val and test and after that I've standardized all the data (train, val and test).
train_mean = train_data.mean()
train_std = ...
- 303
0
votes
1
answer
661
views
AttributeError: 'NoneType' object has no attribute 'outputs' [duplicate]
How do you save a tensorflow keras model to disk in h5 format when the model is trained in the scikit learn pipeline fashion? I am trying to follow this example but not having any luck.
This works to ...
- 3,530
0
votes
1
answer
104
views
GridSearchCV y should be a 1d array, got an array of shape (54000, 10) instead
I've been trying to do grid search on mnist dataset with an MLP. Since mnist dataset is labeled from 0 to 9 and I have 10 neurons in output, I'm using one-hot encoding. But as soon as I try to run ...
- 858
0
votes
1
answer
70
views
Is there a way to use a data generator in a KerasClassifier function?
According to this doc file of TensorFlow, the input data may accept a data generator. However, the KerasClassifier explicitly accepts x and y attributes. How can I go around this since I would want to ...
- 35
0
votes
1
answer
199
views
How do I perform cross validation before finding optimal parameters for my keras model?
I want to perform cross-validation before picking my sequential model's optimal set of hyperparameters.
I understand that the key-error arises from the difference in columns between x_train_scaled ...
- 249