text stringlengths 83 79.5k |
|---|
H: LSTM returns the same results for different inputs
Hey everyone,
I am working on a LSTM network in TensorFlow that predicts the values of the price-index of different product-categories in a month, based on those same values of the 12 months before. Unfortunately I ran into some issues while trying to predict. My d... |
H: In scikit-learn's LDA implementation, how can I sort the topics by frequency over the entire corpus?
I've used scikit-learn to perform LDA topic modeling, and I'd ultimately like to sort the topics by saliency/frequency over the entire corpus, but I'm unsure how to do as such.
I've used pyldavis though there appear... |
H: How to compute performance of a detection-classification system?
I use a yolo (y) to detect only one object and a multiclassifier (mc) that classifies that object.
Now, the problem is: what I have to do with yolo's false positive and false negative, if I want to compute the whole system accuracy, precision and reca... |
H: 2 Most probable labels with Gaussian Mixture Model Clustering
I want to get the two most probable labels for each sample in my X.
A little context: I am working on a clustering project where I have 1.6M samples that have to be clustered into 12 clusters. First, I did KMeans and it works fine, except that KMeans is ... |
H: In general, what are precision, recall, F1 that are reported in papers?
I used classification_report in sklearn library
And, the picture below shows evaluation on my model (anomaly detector)
In general, what are precision, recall, F1 that are reported in papers ?
I think it's reasonable to use precision and recall... |
H: What is few-shots extrapolation?
I'm reading the paper "Learning how to ask" by Qin & Eisner and in the abstract, they mention that using prompts, language models can perform tasks other than text generation. Examples include fill-in-the-blanks (BERT) and few-shots extrapolation (GPT-3).
I am not sure I understood ... |
H: Image autoencoder w/o thousands of dense neurons? prevent large model
I am trying to get around producing large models. If my desired output is a 120x100 image, then do I need a 120*100=12,000 neuron dense layer preceding it?
...
model.add(Dense(120*100))) #<--- mandatory?
model.add(Reshape((120, 100)))... |
H: Xgboost fit won't recognize my custom eval_metric. Why?
Do you know why my custom_eval_metric doesn't work? I get the error:
XGBoostError: [07:56:32] C:\Users\Administrator\workspace\xgboost-win64_release_1.4.0\src\metric\metric.cc:49: Unknown metric function custom_eval_metric
def custom_eval_metric(preds, dtrain... |
H: How to move column in rows direction in Pandas DataFrame?
For example, I want get right from left DataFrame like on picture (move column B on 2 steps down):
PS. "Na" is not necessary, it can be any controlled value like null, zero or empty string.
AI: You can use the .shift method for this:
df["B"] = df["B"].shift... |
H: What is the logic behind recommended normalization parameters in PyTorch?
On the PyTorch documentation for torchvision.models, it is states that images have to be loaded in a range of [0,1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225]. What is the logic behind these specifi... |
H: Anomaly Detection and Removal/Interpolate
I am performing a machine learning regression task on time series data. I have a data frame filled with the close prices of various assets and economic data. I am looking to perform outlier detection on the entire dataframe. Here is my question:
What are some good methods ... |
H: 'str' object has no attribute 'predict'
I am trying to deploy my ML model using flask. My model contains both categorical and numerical variables. Below is my model.py code:-
#PIPELINE FOR PREPROCESSING
dtr_pipe = Pipeline(steps = [('preproc', preproc), ('model',
dtr_model)])
train_x, test_x, train_y, test_y = t... |
H: How to impute missing text data?
Lets say I have a dataframe consisting of two text columns. By text, I mean the values in those columns are either sentences/paragraphs. In such a case, how do I handle missing 'NaN' values?
If it were a numeric data, I would use frequent/mean/median/knn imputation. But, what to do ... |
H: Train-test split and augmentation strategy for small dataset for video classification problem
I have a small data set of videos of approximately 100 videos for each class for a binary classification problem. This results in a total of 200 videos. I am applying two types of augmentations on the videos.
Transformati... |
H: How Are Kernel Weights Trained in 1-D CNN's with Multi-dimensional Input?
I have far from a perfect understanding of how 1-D convolution neural networks learn, but I think I understand how the kernel operates on 1-D input data. How does 1-D convolution work with multi-dimensional input data? An image from this arti... |
H: How can I convert text data to CoNLL format?
This is the same question that I posted on stackoverflow, but I wondered stackexchange would be appropriate for this question.
I would like to convert text data to CoNLL format.
words.txt
I was born in 1981.
From 12 to 24.
tags.txt
O O O O B-DateTime O
O B-DateTime I-Da... |
H: How can I weight each point in one-class SVM?
I want to give weights to some data points
Specifically, these are points related to anomalies
(I'm implementing one-class SVM for anomaly detection)
Exactly, I want to consider some data points that are likely to be anomalies as more important data points
Is it possibl... |
H: Bert to extract local features
Bert is pre-trained model which can be fine-tuned for the text classification. How to extract local features using BERT
AI: First, it is different to fine-tune BERT than extracting features from it.
In feature extraction, you normally take BERT's output together with the internal rep... |
H: How to interepret BERT Attention
Can we tell BERT extracts local features?
For example consider the sentence "This is my first sentence. This is my second sentence".
Now How Bert extracts the features.
attention is computed for each sentence or as whole?
AI: BERT's self-attention will be computed for each pair of t... |
H: Data type when registering Twitter developer account?
When registering the Twitter developer account, I saw the response as below
"Thank you for your request. In order for us to review, we need a few additional details about your plans for the academic access to our API that you’re requesting. The information we ne... |
H: Stop words list to use for CountVectorization
The sci-kit learn library by defaults provides two options either no stop words
or one can specify stop_words=english to include a list of predefined English words
I am using Naive Bayes for SMS spam detection. Is there any other list of stop words
I can experiment wit... |
H: How do I split a data set into train and test sets using
I have the following matrices for training a model:
INPUT FEATURES MATRIX
$$ X =
\begin{bmatrix}
| & | & & |\\
X_1 & X_2 & ... & X_m\\
| & | & & |
\end{bmatrix};
~ X.shape = (n_x, m)
$$
OUTPUT MATRIX
$$
Y =
\begin{bmatrix}
y_1 & y_2 & y_3 & ... & y_m
\end{bm... |
H: How to program conditional statements for this problem in r
Situation: I'm trying to program the following in r.
Task: I am trying to select for words that appear as nouns in my dataset more than they do as adjectives, verbs, or adverbs etc. I have all these counts and below is an example of one instance of what I ... |
H: MNIST dataset for digit recognition
I noticed that the MNIST dataset for digit recognition is just a CSV file
They don't provide the images.
https://www.kaggle.com/c/digit-recognizer/data
Is it possible to get the corresponding images for the dataset?
AI: Those csv files contain the actual data that would normal... |
H: Does validation data has any effect on training or it acts solely without affecting the training?
When using Keras library of Python, we use validation data with training data while training our model. In every epoch, we get a validation accuracy. Does this validation accuracy have any effect on training on the nex... |
H: Managing Multiple Observation at the same time stamp timeseries forecasting deep learning
I have a dataset timeseries forecasting that includes the categorical columns and numeric as well.
here is a sample of it
Date | categorical _fature_1 |categorical _fature_2| Feature_1_numeric | feature_2_numeric | price
1-1... |
H: Transformers (BERT) vs LSTM on Sentiment Analysis/NER - dataset sizes comparison
I am aware (continuously learning) of the advantages of Transformers over LSTMs.
At the same time, I was wondering from the viewpoint of size of the data needed, contrast of those two techniques, supposing I want to train for a downstr... |
H: When should you use Ensemble?
I have been working on a project as a part of my studies(computer/data science).
I tried to make the best classifier I can with what I learned, and recently I have tried to upgrade this classifier using new things I learned.
I have tried using several ensemble methods such as bagging, ... |
H: When is it meaningful to use ML on data without statistical statistical significance
I have a dataset with three columns, the number of days behind or ahead of the deadline a project was completed, the name of the department whose project it was and a user number for the employee.
The company are interested in pred... |
H: When should I use 'rbf' and 'polynomial' kernel trick in machine learning algo?
I have a problem about hate-speech classification using support-vector machine algorithm.
The task is to identify the sentence that contains 'positive' or 'negative' sentiment.
Which is the best Kernel Trick? ('rbf' or 'polynomial')
AI:... |
H: An issue for sub-word tokenization preprocessing transformer
I'm stacked with executing the sub-word tokenization preprocessing to use transformer.
According to the tutorial on the article, I have executed the sample code.
However, one function was not defined properly and no hint to fix it on the article.
If you h... |
H: Sub-word tokenization preprocessing to use transformer
I'm stacked with executing the sub-word tokenization preprocessing to use transformer.
According to the tutorial on the article, I have executed the sample code.
However, one function was not defined properly and no hint to fix it on the article.
If you have an... |
H: Model performance worsens after Cross Validation
I am training a logistic regression model on a dataset with only numerical features. I performed the following steps:-
1.) heatmap to remove collinearity between variables
2.) scaling using StandarScaler
3.) cross validation after splitting, for my baseline model
4.)... |
H: Select current column when iterating dataframe
I have a dataframe with 1k+ columns like this(i cut it).
I want to iterate the dataframe and check row by row if value is True then assign it the Incident Type.
How do i get the current column it is iterating to? I thought of something like that:
for index,rows in df3... |
H: Labeling and aggregating features issue
I am trying build a simple binary classifier (some tree based algorithm for now) and my training data will have features aggregated at the user level. So I'll have a unique records of each user. These aggregated features are like "number of logged in sessions", "number of tim... |
H: Transforming time series into static features?
I'm working on a side project where I have a mixture of static data and time series, and the goal would be to perform clustering on the data. There's a bunch of data sources, but basically the main thing would be some static information about users (like age, sex, loca... |
H: An example of explainable, but not interpretable ML model
This post attempts to explain the difference between explainability and interpretability of ML models. However, the explanation is somewhat unclear. Can somebody provide specific examples of models that are explainable but not interpretable (or the over way ... |
H: Whats the minimum size sample required for a LSTM RNN model?
I have a data set of 100 rows x 100 to 300 columns. Will an LSTM RNN model work for my data or do I need more data?
If my sample size is a problem are there other RNN architectures capable of modeling using small dataset sizes?
AI: This totally depends on... |
H: Perfect scores for multiclass classification
I am working on a multiclass classification problem with 3 (1, 2, 3) classes being perfectly distributed. (70 instances of each class resulting in (210, 8) dataframe). Now my data has all the 3 classes distributed in order i.e first 70 instances are class1, next 70 insta... |
H: Which is meant by +/-9.2e18 years in timespan?
I was able to convert the 9.2e18 AD to a date, but I am confused about the exact date. Which date is 9.2e18 AD and and 9.2e18 BC?
Time span (absolute) - [9.2e18 BC, 9.2e18 AD] i.e +/- 9.2e18 years
NumPy documentation, section "Datetime Units" under "Datetimes and Tim... |
H: In keras seq2seq model, what is the difference between `model.predict()` and the inference model?
I am looking into seq2seq model in keras, for example, this blog post from keras or this. All the examples I have seen have some inference model, that depicts the original model. That inference model is then used to ma... |
H: Problem with CNN
I am using the BreakHis database. More specifically, I am trying to classify the 400X images. The sizes of the images are $700x460x3$.
Here are the details of the dataset. Also, here is the code for the classification:
from keras.preprocessing.image import ImageDataGenerator
datagen = ImageDataGene... |
H: Invalid shape (4, 460, 513) for image data
I am using read_image to read the image.
from torchvision.io import read_image
image = read_image("/content/train/000001-11.jpg")
Now, when I try to find the shape of the image, I get $(4, 460, 513)$ as the image shape.
But, when I use opencv to read the image, I get $(46... |
H: Why does the test set class imbalance influences my model's performance?
Considering a balanced training set, I noticed that the results of a classification primarily depend on the class imbalance of the test set.
As shown in this article, unless the classes are perfectly separable, the performance (precision & rec... |
H: How to properly present data set with categorical, numeric and time series data?
I have a following toy dataset example with target variable repair_type
id | car | mileage | repair_type | sex | age
1 | Honda | 12000 | engine | 1 | 50
1 | Honda | 12000 | suspension | 1 | 50
1 | Honda | 15000 ... |
H: 'Collision' resolution for precision in object detection
For object detection we often use metrics based on precision/recall. My question is what is generally the process of matching the prediction and ground truth bound boxes, when there are multiple intersecting boxes.
I.e. consider the image bellow for single-cl... |
H: Shouldn't a test be repeated X times and average the results to determine the best machine learning model?
I have searched in several web pages how to choose the best machine learning model for a dataset and they all seem to agree that they should be compared using the same seed. However, they only run the test onc... |
H: Activation Function
I am very new to machine learning and made an experiment myself. I have a few questions:
Can I use $Y = sin(x)$ or $Y = 2x$ as an activation function for a neural network?
Is it necessary to increase the size of a convolutional kernel to improve the effect of a convolutional neural network (CNN... |
H: Row embedding as output of a transformer - how are they defined?
I am reading the paper Tabular transformers for modeling multivariate time series and am having issues understanding the structure in Fig. 2. In Sec. 2.2, the authors say that the field transformer processes rows individually, creating row embeddings.... |
H: How to convert post-padded sequence to pre-padded sequence?
I have numpy array as follows:
train_x = [[1,2,3,0,0], [2,5,0,0,0], [2,3,0,0,0], [0,0,0,0,0], [0,0,0,0,0,0]]
Now, I would like to transform it to as shown below:
new_train_x = [[0,0,0,0,0],[0,0,0,0,0,0],[0,0,1,2,3],[0,0,0,2,5],[0,0,0,2,3]]
I tried writin... |
H: How to predict only those values that our model is 95% sure of?
I have 5 classes. I made a XGBoost Classification model and used model.predict(test) to predict the classes of test dataset. Out of all those values predicted by my model, I would like to know only those values that my model is more than 95% sure that ... |
H: Explain TFX Metadata Store data model definitions
GOAL
Explain the following definitions in plane simple english?
Many practical examples of what they can have?
What each of them do?
ORIGINAL
This is the original https://www.tensorflow.org/tfx/guide/mlmd#data_model that I have issues with. In my description I wil... |
H: How can I extract an optimized matrix of correlations from a larger data set?
Consider an Excel sheet containing a matrix of correlations between individual stocks and the combined portfolio as a whole:
How can I extract an optimized matrix such that most stocks have a low correlation to each other? If I simply go... |
H: Tensorflow for Deeplearning and Machine learrning
We can use TensorFlow for both machine learning and deep learning. So why do we use scikit-learn more in machine learning and not TensorFlow?
Are they both alternatives of each other?
AI: Tensorflow is usually used for deep learning (neural networks) and scikit is u... |
H: Can't use The SGD optimizer
I am using the following code:
from tensorflow.keras.regularizers import l2
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Add, Conv2D, MaxPooling2D, Dropout, Flatten, Dense, BatchNormalization, Activation
from tensorflow.keras import activations
CNN_... |
H: What are some application of Google's PageRank Algorithm in Data Science
I came across a topic on computational linear algebra that talks about iterative algorithms to compute eigenvalues. I've worked with power method which is an iterative algorithm that converges a sequence of vectors to the largest eigenvalue.
O... |
H: Remedy for small batch size?
I am trying to reproduce results of other people's research, but we cannot afford to do it with the same batch size as theirs, due to limited computing resources.
The method they use is a simple sequence classification using BERT.
They do it with batch size 48, learning rate 4e-5, optim... |
H: What is the point of generating new features (linear or non linear) out of existing features in a dataset?
During feature engineering, we can create new features out of existing ones by using arithmetic operations albeit linear or not.
Let's say we have two features x and z. We can then create (engineer) a new feat... |
H: Preprocessing , EDA , and Feature Engineering
What is the difference between EDA, Feature Engineering, and Preprocessing?
The main purpose is to make the raw data suitable for modeling. In EDA, we are cleaning the data and so does the preprocessing. Wheras in FE, we are scaling and imputing.
AI: EDA(Exploratory Dat... |
H: Encoding when different number of records for each month-One hot or different type?
I am working on a dataset which is pretty small: 1169 records.
There is a column called month and it takes the values 'Jan','Feb', or 'March'.
The number of records for each month is different. I have 542 for 'Jan', 443 for 'Feb', a... |
H: New to keras neural network and k-fold cross validation
I'm new to learning neural networks and I found an example online to test accuracy with k-fold cross validation.
The example is for binary data but I want to test MAE or RMSE (I guess?) for my regression prediction. I'm a bit lost now on how I can test it beca... |
H: How to convert mixed datetime formats into single one in pandas?
I am working with DataFrame which contains multiple datetime formats in one column. For example:
2020-11-09 00:00:48
2020-11-09 00:00:48
2020-11-09 00:00:48
2020-11-09 00:00:48
2020-11-09 00:00:48
2020-08-25
2020-08-25
2017-08-25
2018-08-25
2020-08-25... |
H: Ongoing feature selection
If you have a set of n features you have 2^n-1 non-empty feature subsets. As a result, if you pick one of them you are unlikely to have found the best one.
To me, it seems intuitive that as you build your model, you would want to look at the things it does badly and try to find features th... |
H: why i got error when using SimpleImputer for impute Nan values?
I have the following code, where sp_col is a sliced column of my dataframe df_1:
from sklearn.impute import SimpleImputer
import numpy as np
imputer = SimpleImputer(missing_values=np.NaN,strategy='mean')
sp_col =df_1.iloc[:-1,9:28].values.reshape(-1,1... |
H: Error when using simple imputer to impute nan values?
df_2 is data frame
My code:
from sklearn.impute import SimpleImputer
impute = SimpleImputer(missing_values=np.NaN,strategy='mean')
df_2.iloc[:,2:9] = impute.fit(df_2.iloc[:,2:9])
I got error:
TypeError: float() argument must be a string or a number, n... |
H: Looping when predicting with keras
I created a prediction with keras and I want to see what are results if I repeat the prediction 10 times for example. If I loop I get a warning
WARNING:tensorflow:6 out of the last 6 calls to <function Model.make_predict_function.<locals>.predict_function at 0x1507503a0> triggered... |
H: Splitting before tfidf or after?
When should I perform preprocessing and matrix creation of text data in NLP, before or after train_test_split? Below is my sample code where I have done preprocessing and matrix creation (tfidf) before train_test_split. I want to know will there be data leakage?
corpus = []
for i i... |
H: How to load any particular folder files from a zip dataset
Twitter is a great source of information. Using The Health-News-Tweets.zip dataset contains tweets by different agencies like BBC Health, CBC Health, etc. I will perform a clustering on these tweets and will use the elbow method to identify the ideal numbe... |
H: Fastest way to replace a value in a pandas DataFrame?
I am loading in 1.5m images with 80,000 classes (or I will have to when I eventually train) into a Keras generator and am using a pandas dataframe to do so. The problem is, with so many images, my code takes a long time to run. I have an issue with the specific ... |
H: Why are words represented by frequency counts before embedding?
Before getting vector representations of words by embedding, the words are mapped to numbers. These numbers are chosen to be the frequency of that word in the dataset. Why does this convention exist? Does it have any effects, or is it arbitrary?
AI: Th... |
H: Failed to convert a NumPy array to a Tensor
i use this model
model = Sequential([
Dense(units=10, input_shape=(1,), activation='relu'),
Dense(units=32, activation='relu'),
Dense(units=10, activation='softmax')
])
model.compile(optimizer=Adam(learning_rate=0.0001), loss='sparse_c... |
H: Train set and Test set For result and conclusion
I have performed a Logistic regression on a binary classification dataset.
The result are as follow :
The training-set accuracy score is 0.8523 while the test-set accuracy to be 0.8442.
For Model evaluation and improvement using Kfold and GridSearch cv :
kfold valida... |
H: Compare 2 Pandas Column, over different rows and Output the resultant in a 3rd column
I have a data frame like this.
EMPLOYEE_ID
NAME
MANAGER_EMPLOYEE_ID
0
42
S
40
1
40
G
NaN
2
T
M
40
3
0c
H
42
I want a separate column that outputs the Name of the Manager of every person.
So for S manager would b... |
H: NLP techniques for converting from a direct speech to a reported speech
Any idea of some NLP techniques to transform a direct speech to a reported speech ?
Example converting : "I'm learning NLP" said a user to : a user said he's learning NLP.
I thought about paraphrasing but not sure..
Thank you!
AI: What do you m... |
H: Speculation regarding rising trend in softwares with a potential in diminishing the role of a data scientist
I have recently came across a demo of for a software that's built on natural language processing and its job is to convert a text the user types on a Python interpreter such as Jupyter and the software autom... |
H: I am getting (loss: nan - accuracy: 0.0000e+00) for all epochs after training the model
I made a simple model to train my data set which consists of (210 samples and each sample consists of a numpy array of 22 values)
and x_trian and y_trian look like:
and this is my simple code:
import tensorflow as tf
from tenso... |
H: How to Inference With Keras Sequential Models (Text Classification)
I have the following LSTM model and I can't make inference with it:
print("Define LSTM model")
rnnmodel=Sequential()
rnnmodel.add(embedding_layer)
rnnmodel.add(LSTM(128, dropout=0.2, recurrent_dropout=0.2))
rnnmodel.add(Dense(2, activation="sigmo... |
H: LSTM model, poor performance
I have been working on a project on the demand for a product.
I am using data from 2016 to train the LSTM model.
The architecture is as follows:
model_2016 = Sequential()
model_2016.add(LSTM(units = 256, input_shape=(1, look_back), return_sequences = True))
model_2016.add(Dropout(0.2))... |
H: Text mining match in Python
I have one column called A and one column called B where column B is a more detailed description of column A, for example:
A
B
want
red apple
I love RED apple
100%
organic orange
I love citric fruit
0%
BANANA lover
I have 2 banana
50%
Basically the algorithm is if we have... |
H: how to choose between data normalization or standadization?
I have been studying about data scaling. Two common methods for it are the StandardScaler and MinMaxScaler. As I understood, StandardScaler expects the data to be normally distributed, but I have seem some examples online where this method is used for skew... |
H: Plot two categorical variables against two numeric variable in ggplot
In my dataset, I have two numeric revenue features, one for each month, and two categorical features one for region and other for value segment. what I want to do is compare these two revenues col by col for each region and facet wrap by value se... |
H: How to get column names after One Hot Encoding when using Pipelines?
I am using Pipeline and ColumnTransformer to preprocess the data. Basically I am using them to impute null values, scale the numerical data and finally perform OneHotEncoding. When I fit the ColumnTransformer object to my train and test data the r... |
H: Why is a MinMax Scaler scaling every coloumn independently?
Why is a MinMax Scaler scaling each column independently? Isn't it losing information if the values are somehow connected?
If the value in column B is always an upper limit for the value in column C, after scaling the value in column B will be smaller than... |
H: 1D target tensor expected, multi-target not supported
I am trying to train my model.
My model outputs a [4,2] tensor where 4 is the batch size and 2 because of binary classification.
After receiving the outputs I found the index of the maximum element for each row.so , now the shape is [4,1] and the shape of my lab... |
H: How to expand lists?
In lists the main noun is often only mentioned at the end. However, for e.g. NER-tasks, I would like to "expand" them:
Outward and return flight -> outward flight and return flight
Proboscis, Vervet, and golden snub-nosed monkey -> Proboscis monkey, Vervet monkey, and golden snub-nosed monkey
... |
H: When calculating lexical richness (e.g. TTR) do you lemmatize first?
When calculating Type-Token Ration (TTR) and Hapax richness (along with similar measures), do you lemmatize the corpus first?
AI: To my knowledge there's no standard way to use lexical density measures such as Type-Token Ratio (TTR).
It's common t... |
H: Why cant we further tune/change the model after evaluating on the test set?
Every thread on stackexchange that I've found says that you can only use the test set once and thats it. So for instance, if you used a linear regression model and got poor results on the test set, you cannot change the the model to say a r... |
H: How to set the priority to Machine leaning algorithms for Binary classification among Four based on accuracy and fitting
Rain Classification in Australia
Under this context, sklearn classification algorithms will be used, namely:
Logistic Regression Classification (Parametric)
Decision Tree Classification (Non para... |
H: Convolutional neural network low performance
Problem description
I'm doing Signal Modulation Classification using a Convolutional Neural Network, but performances are very low (around 15% accuracy) and I can't find out why.
Data
Dataset is composed by 220.000 rows like these. Data is perfectly balanced: I have 20.0... |
H: Reinforcement Learning - after 800k timesteps agent wont learng
I'm study and try a reinforcement learning.
now im using gym and stable-baseline.
My project have a step where I calculate a reward with a function.
-> step() -> calculate_reward() -> return calculated reward -> in step():
I set total_reward = -1000 ... |
H: Get the percentage of each class in classification
Let me explain my question with an example.
Imagine our machine learning model is trained using many different classes such as rectangle, circle, and triangular, etc.
Now I want to give it a new shape, for instance, a rounded corner rectangle.
What I need is not on... |
H: Why is sequence prediction always the objective in RNN and LSTM like algorithms
The title is pretty much my question. I haven't seen any literature yet that uses a different training objective. The goal is to find the hidden states eventually, then why is it that only 1 method is so popular, and there are no others... |
H: Annotating NER dataset
I am working on annotating a dataset for the purpose of named entity recognition.
In principle, I have seen that for multi-phrase (not single word) elements, annotations work like this (see this example below):
Romania (B-CNT)
United States of America (B-CNT C-CNT C-CNT C-CNT)
where B-CNT s... |
H: Taking the squared Euclidean distance for kNN classification of images
A problem I'm working on states:
Computes the squared Euclidean distance between each element of the training
set and each element of the test set. Images should be flattened and treated
as vectors.
The training set is a tensor of dimensions: ... |
H: How to measure the accuracy of an NLP paraphrasing model?
I using the HuggingFace library to do sentence paraphrasing (given an input sentence, the model outputs a paraphrase). How am I supposed to compare the results of two separate models (one trained with t5-base, the other with t5-small) for this task? Can I ju... |
H: Calculating Dissimilarity between attributes
I would like to figure out how to calculate the dissimilarity between attributes Jack and Jim. Given the attributes table shown below
Given the Relational table.
and the example calculations
I would like to understand how the dissimilarity is calculated for Jack and J... |
H: "Up or down but not sideways" bimodal time series prediction - what is the best way to model it?
Say I have a time series (e.g. bitcoin price). I want to predict tomorrow's price, specifically tomorrow's % change in price from today. Let's say this is gaussian distributed, with the mean at 0%.
If the market is tren... |
H: Tensorflow parameters for CNN
I created the below simple model (taken from a Coursera course). It has a total of five convolutions.
model = tf.keras.models.Sequential([
# Note the input shape is the desired size of the image 300x300 with 3 bytes color
# This is the first convolution
tf.keras.layers.Conv... |
H: How are the values for the `sex` feature in sklearn Diabetes dataset obtained?
I'm just starting out with using sklearn for my own Machine Learning project and I'm using sklearn's built-in "Diabetes" dataset.
While performing data exploration on the features, I noticed something a bit confusing to me about the sex ... |
H: Significance of Convex Loss Function with Nonlinear Models
When used in a linear model, a convex loss function guarantees a unique global minimum for the parameters, which can be found by local optimization methods.
However, when the model is nonlinear (e.g. MLPs), local minima are possible for a convex loss.
Are t... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.