Questions tagged [nlp]
Natural language processing (NLP) is a subfield of artificial intelligence that involves transforming or extracting useful information from natural language data. Methods include machine-learning and rule-based approaches.
20,628
questions
0
votes
0
answers
19
views
BERT embedding cosine similarities look very random and useless
I am new to this field, so maybe I am misunderstanding something. However, I thought you can use BERT embeddings to determine semantic similarity. I was trying to group some words in categories using ...
- 93
0
votes
0
answers
16
views
How do I install language model for spacy on Kaggle?
Aloha! Everybody knows how to install model at home:
python -m spacy download ru_core_news_md
But since python notebook on Kaggle is isolated of the global web, it does not seem possible to do so.
...
0
votes
0
answers
12
views
module 'keras_nlp' has no attribute 'models
HAS ANYONE ELSE EXPERIENCED THE SAME ERROR WHEN RUNNING IT LOCALLY? IT RUNS CORRECTLY ON COLAB.
module 'keras_nlp' has no attribute 'models
i
Tried to install the updated version of
pip install -U ...
- 1
0
votes
0
answers
13
views
“Bus Error and Resource Tracker Warning When Training PyTorch Model on GPU with MPS”
I’ve built a vanilla Transformer using PyTorch for machine translation and am encountering issues while trying to train it on an Apple Mac M3 with a 12-core CPU and an 18-core GPU (18GB RAM) ...
-2
votes
0
answers
16
views
Enhancing Document Layout Analysis by Adding Positional and Character Information to CNN Inputs
I am working on document layout analysis and have been exploring CNNs and transformer-based networks for this task. Typically, images are passed as 3-channel RGB inputs to these networks. However, my ...
-2
votes
0
answers
13
views
Using RAG to identify the source of error in a code file [closed]
I am trying to implement a small tool that automatically identifies which part of the code from a set of code files (given as input) is responsible for the error text displayed during execution.
The ...
- 7
1
vote
0
answers
16
views
Text summarizations of comments and replace the duplicates with the first occurrence if the meaning is comment is same
Context - Doing an NLP project to analyze comments column in a data frame. I want to replace the duplicates with the first occurrence if the meaning of the comments are same.
I wants to compare all ...
-1
votes
0
answers
14
views
Models for getting similarity scores between categories and keywords
I want to get a similarity score between a category like vehicles and a list of words like headphone, water, truck, and green. The goal would be for each score to be low on words outside the category ...
0
votes
0
answers
12
views
Feature Importance Chart in Keras API
I am implementing a series of three deep learning models (RNN, 1D-CNN, and custom transformer), all utilizing the Keras API, for an NLP binary classification problem. I would like to generate a ...
- 113
-2
votes
0
answers
14
views
What's the difference between contextual encoders like USE and OpenAI's text-embedding-ada-002? [closed]
Just came across Universal Sentence Encoder which can also keep up the context during the semantic search and other operations. Is OpenAI's text-embedding-ada-002 much more advance or more or less ...
0
votes
0
answers
29
views
The Impact of Pretraining on Fine-tuning and Inference
I am working on a binary prediction classification task, primarily focusing on fine-tuning a BERT model to learn the association between CVEs and CWEs. I've structured my task into three phases: first,...
- 19
-1
votes
0
answers
21
views
Live transcription with word-level timestamps
I'm new in Machine learning. I have task to live transcribe audio from microphone and also make word-level transcripts of this live transcription instead of utterance level.
I tried these projects:
...
- 1
0
votes
0
answers
18
views
How to Combine Semantic Search with SQL Analytical Queries?
I'm creating an LLM-agent that can provide insights from a complex database. The database includes several columns of different types (datetime, numeric, and text).
For simplicity, let's assume I have ...
- 43
0
votes
0
answers
20
views
Memory usage when using spaCy Doc extensions
Issue
Before preprocessing my data with spaCy, I typically have my data stored in a Pandas Series. Since I'd like to preserve the index for each document before serializing my Docs, I decided to use ...
- 359
0
votes
0
answers
26
views
Combine strings with redundant information without losing business logic in Python
I have a business entity called "Groups". This groups are linked to a PQL-like string called "rule". This rules are evaluated to check which products from an ecommerce match with ...
- 23
0
votes
0
answers
11
views
Which RAG methods/concepts can I use for a benchmark?
I am writing a practical assignment for my uni. There I have to analyse different RAG methods and compare them. Since I am in my 2nd semester of information systems and I lack of experience within the ...
0
votes
0
answers
16
views
spacy-llm & SpanCat for address parsing
I'm currently developing a project to standardize and correct a dataset of inconsistently formatted addresses using spaCy-LLM and spaCy.SpanCat.v3. The goal is to train a model on examples of ...
-1
votes
0
answers
16
views
AssertionError: Unexpected kwargs: {'use_flash_attention_2': False}
I'm using EvolvingLMMs-Lab/lmms-eval to evaluate LLaVa model
after running accelerate launch --num_processes=8 -m lmms_eval --model llava --model_args pretrained="liuhaotian/llava-v1.5-7b" ...
- 41
-1
votes
0
answers
20
views
BERT: how to get a quoted string as token
I eventually managed to train a model, based on BERT (bert-base-uncased) and TensorFlow, to extract intents and slots for texts like this:
create a doc document named doc1
For this text, my model ...
- 1,526
-2
votes
0
answers
27
views
Free speech-to-text APIs, libraries, or open-source solutions for Python? [closed]
I'm developing a Python application that requires speech-to-text functionality, preferably for multiple languages including Turkish. I'm looking for free or open-source solutions that can be easily ...
- 1
0
votes
1
answer
19
views
How is coherence score calculated in Mallet?
I do understand how the diagnostics output shows the coherence values for each topic but my values range between -150 and -600 and other posts that I have seen where Mallet was used show coherence ...
- 31
-1
votes
0
answers
28
views
Word Prediction to complete sentence [closed]
The visuals in this movie are _________, but the story is paper-thin.
It's all flash and no substance.
How can I better recommend word to complete the sentence
Tried a code and got output as The ...
- 7
-3
votes
0
answers
17
views
Aı deep learning code generator for code vectorization [closed]
Python Hello, I have language and code headings consisting of json structure and there are codes under the code headings. Can you help me clean and vectorize the codes? Thank you. Note: I want my ...
0
votes
0
answers
18
views
C# english word stemming and lemmatizing using Catalyst - how to do that
I added Catalyst nuget package to my C# project, to help me lemmatize english words. However its documentation is not clear, and lacks of examples, I tried to lemmatize/stem only one word:
Catalyst....
- 1,117
-1
votes
0
answers
15
views
Checking semantic meaning of 2 texts while considering the order of the texts
I am doing a task related to checking the semantic meaning similarity between 2 texts. There I used BERT sentence-transformers/all-MiniLM-L6-v2 model.
Input 1 - "Object moves in uniform ...
- 194
1
vote
1
answer
43
views
not able END function using "add_conditional_edges" in lang graph
this is my code:
import os
from dotenv import load_dotenv
load_
dotenv()
from langchain_openai import ChatOpenAI
from langgraph.graph import StateGraph, END
from langgraph.graph import Graph, ...
- 13
-5
votes
0
answers
34
views
Is there any way to apply NLP on my survey responses of stress [closed]
Is there any way to apply NLP technique on my survey responses of stress. I have survey data for stress. My columns data are like:
Stress_Frequency Low_Energy_Frequency \
0 1 Arts and ...
-2
votes
0
answers
22
views
HuggingFaceInstructor Embedding giving error? [closed]
I am working on a chatbot to handle FAQs. I have tried Huggin FaceInstructor embedding but it has a lot of depedencies and i don't know why my anaconda prompt is not able to install all the required ...
1
vote
1
answer
28
views
How to make dynamic API calls based on user input in a Gemini application python nlp?
I'm working on a Gemini application where I need to make dynamic API calls based on user input. Specifically, I want to perform different API requests depending on the user's query. For example, if ...
- 23
0
votes
0
answers
46
views
Natural Language Question Answering: How do you train and evaluate using ML.Net
I want to know how to train and evaluate Natural Language Question Answering model using ML.Net. I already have the training working but I got stuck at the Evaluation part where you ask a question and ...
- 71
0
votes
0
answers
17
views
Why charater based LSTM are taking more time than word based LSTM while next word prediction
I wanted to train LSTM Model for Next Word Prediction using word-based and character based. I used similar data processing technique for both character-based and word-based.
For Character,
class ...
- 373
0
votes
0
answers
10
views
Calling Asr model in BHasini throws - {"code":"something went wrong","message":null,"timestamp":"2024-07-03T06:08:52.407+00:00"}
I am trying to call the asr model of bhasini api to get transcription of my audio.
I tried the following code.
import requests
import json
url = "https://meity-auth.ulcacontrib.org/ulca/apis/v0/...
0
votes
0
answers
26
views
Understanding and improving coherence values using Mallet
I am attempting to run an LDA topic model using Mallet. My corpus consists of user comments from news websites. It's a relatively small corpus with approx. 614k words.
The first approach I took was to ...
- 31
0
votes
0
answers
17
views
Named Entity Recognition using TFLite on Android
I have a named entity recognition TensorFlow Lite model that I trained in Python and that I would like to run on an Android. To make predictions, the model takes in a dictionary of two tensors, namely ...
- 1
0
votes
0
answers
37
views
GliNER finetuning - no validation loss is logging
I am trying to fine-tune using this notebook: GLiNER/examples/finetune.ipynb at main · urchade/GLiNER (github.com)
However, the logs only show 'loss' , which I assume is the training data set loss, ...
- 33
0
votes
0
answers
67
views
UndefinedFile: could not open extension control file "/usr/share/postgresql/14/extension/vector.control": No such file or directory
I'm trying to run postgres along with pgvector but getting this error: UndefinedFile: could not open extension control file "/usr/share/postgresql/14/extension/vector.control": No such file ...
- 1
0
votes
1
answer
34
views
How can I make my Hugging Face fine-tuned model's config.json file reference a specific revision/commit from the original pretrained model?
I uploaded this model: https://huggingface.co/pamessina/CXRFE, which is a fine-tuned version of this model: https://huggingface.co/microsoft/BiomedVLP-CXR-BERT-specialized
Unfortunately, CXR-BERT-...
- 441
-2
votes
1
answer
23
views
Hugging Face model with large context window [closed]
Im looking for a model that accepts input like 50k characters and a prompt answering a question based on that text. Is there something available like that? Not sure how to find it, new to AI.
- 764
-1
votes
1
answer
49
views
English text tokenization in C# not python is possible? [closed]
In our software we have to analyze a plain text file. First we should break the text into paragraph, then into sentences, then into tokens. Final steps (as far as I understand) is the stemming and ...
- 1,117
0
votes
0
answers
15
views
Spark code for NLP processing for china data
I am using pySpark and used jieba for china data. I have have UDF in code and did broadcast but column enriching is taking hours and hours for 5 million record. I am clueless what we need to here to ...
- 1,018
1
vote
1
answer
42
views
Alternative to Receptive field in Transformers and what factors impact it
I have two transformer networks. One with 3 heads per attention and 15 layers in total and second one with 5 heads per layer and 30 layers in total. Given an arbitrary set of documents (2048 tokens ...
- 421
0
votes
0
answers
32
views
"KeyError: 0" when calling model.fit() in Keras
I am training a simple sequential model for NLP in keras. Here is the model architecture:
from keras.models import Sequential
from keras import layers
from keras.layers import Embedding, Flatten, ...
- 113
1
vote
0
answers
19
views
Spacy detect correctly GPE
I've a set of string where I shall detetect the country its belongs to, referring to detected GPE.
sentences = [
"I watched TV in germany",
"Mediaset ITA canale 5",
&...
- 677
0
votes
0
answers
23
views
why isn't tf.keras.layers.TextVectorization accepting standardization=None?
I'm still trying to get this work (and to learn!) so I am using a tiny corpus.
I do some preprocessing on the text in order to get specific bi-gram collocations using nltk (not relevant here but I ...
- 131
-2
votes
0
answers
23
views
is there any best pretrained model of any spacy,ner or any for resume parsing else is there any other way to create a resume parser application
pls im working on this rom past 25 days is there any way or solution for this if there is any pretrained model i can use its also good for getting atleast 80% accuracy for entity extraction from ...
0
votes
0
answers
41
views
How to generate output of HuggingFace PEFT model with previous message history as context?
I am trying to generate text from my fine-tuned Llama3 model which uses the PEFT AutoPeftModelForCausalLM library while also passing in previous message history.
This is how I am currently generating ...
- 27
0
votes
0
answers
18
views
A prepand *Paraphrase* is showig all the time after runnig the code instead of actual paraphrased Sentence
I am creating an URDU TEXT PARAPHRASING tool for my semester.
I have used T5 Model and fine tuned it.
Now when im running this code:
**import torch
from transformers import T5ForConditionalGeneration, ...
0
votes
0
answers
10
views
Understanding time complexity for dynamic indexing
While reading the text, Introduction to Information Retrieval by Manning et al., I came across dynamic indexing (Sec 4.5, Page 79). It consists of two indices, the main index (stored on disk) and an ...
- 1,249
0
votes
0
answers
43
views
Python Rasa train not compatible with M3 Apple
Unable to run the "rasa train" in mac m3 terminal..
giving below error.
2024-06-26 17:27:12 INFO rasa.cli.train - Started validating domain and training data...
zsh: illegal hardware ...
- 727
-1
votes
0
answers
51
views
How to open `gguf` models in Windows?
How to run a file with .gguf extension on a Windows machine? I tried to open the file with PyCharm or Visual, but they did not recognize this file. And, when I tried to run the command to use the ...
