82,122 questions
-4
votes
0
answers
72
views
How to Enahnce the predictions of LSTM model [closed]
I'm working on small project using the modern portfolio theory for portfolio optimization, the final required output is a recommendations on which assets to buy and how much to buy.
to achive that ...
Advice
0
votes
2
replies
50
views
TensorFlow models DoS via input-controlled tensor sizes?
Running untrusted TensorFlow models and noticed this:
import tensorflow as tf
import time
class TestLayer(tf.keras.layers.Layer):
def call(self, inputs):
n = tf.cast(inputs[0][0], tf....
Best practices
0
votes
0
replies
39
views
FID score computation while training WGAN-GP
I am training WGAN-GP on Eurosat dataset, splitted into train/val/test sets in counts 18900/4050/4050. Since FID scores are widely used in GANs in image generation, I based my hyperparameter search on ...
Advice
0
votes
0
replies
59
views
Analysis with Modeling on Spotify Tracks Dataset
Context: I’m predicting 8 music genre families using 114,000 Spotify tracks based only on raw audio DNA (danceability, acousticness, energy, etc.).
The Plot Twist: My baseline Neural Network hit 82% ...
Best practices
2
votes
1
replies
95
views
Can i make a Neural network And simply ML Model using a smartphone? (Ain't Have Pc/Laptop Btw)
I am someone who is currently Learn Python, especially libary for make machine learn, but i am not have pc/laptop , can i make my own ML & Neural model?
I haveI've heard that we can use Google ...
- 1
1
vote
1
answer
56
views
Custom PyEnvironment time_step and time_step_spec do not match
I'm creating a custom PyEnvironment in TensorFlow Agents to simulate the track and field decathlon. I've managed to create a functioning environment in the sense that I can use _step and _reset, but ...
- 31
Advice
1
vote
4
replies
108
views
Can I create a layer in Keras that de-scales my output?
I am working on a keras regression network that takes about 60 input variables and outputs 35 variables. For both the input and output, about half of the variables are in the range of ±10, while the ...
- 1
Best practices
0
votes
7
replies
107
views
Which is the best way to detect lines in historically book pages
I am working on an OCR project and need to create a dataset consisting of approximately 1247 pages from 6 books. I need to crop the images line by line and transcribe the text for training a model. ...
Advice
0
votes
0
replies
121
views
Built a Continued Pretraining + Fine-Tuning pipeline for a Veterinary Drug LLM on BioGPT-Large — Looking for feedback on my approach
I've been working on adapting Microsoft's BioGPT-Large for veterinary pharmacology using Plumb's Veterinary Drug Handbook (2023) as my domain corpus. After going through a lot of trial and error, I ...
1
vote
1
answer
156
views
ValueError: shapes mismatch when combining Re-ID Cosine distance and IoU matrices for custom MOT tracking
I am building a custom Multi-Object Tracking (MOT) system using Python, OpenCV, and TensorFlow. My goal is to track people and perform real-time clothing recognition. To prevent ID switches when a ...
- 11
Advice
0
votes
2
replies
71
views
How to correctly extract the CLS token from a Keras Hub ViT backbone, and clarify preprocessor usage and pretraining dataset?
I’m working with a Vision Transformer (ViT) backbone from Keras Hub and building my own classification head. My code looks like this:
python
def get_vit_model(model_variant='vit_base',
...
- 161
1
vote
2
answers
104
views
Input 'y' of 'Sub' Op has type float16 that does not match type float32 of argument 'x' when using mixed precision with custom loss
I am experimenting with mixed precision training in TensorFlow to speed up training on a GPU. I enabled the global mixed precision policy and implemented a custom loss function. However, when training ...
- 25
0
votes
1
answer
67
views
Building tensorflow ambiguity errors
I'm building tensorflow 2.20.0 from source from official repo (from release branch v2.20.0)
with rocm and -march=native
I use:
export HERMETIC_PYTHON_VERSION=3.13
export ROCM_PATH=/opt/rocm/
export ...
- 99
Advice
0
votes
2
replies
117
views
How initializing weights with large value causes vanishing gradient problem in neural network
I was watching this tutorial on weight initialization in neural network, and im not able to understand this statement:
In case of Tanh, Sigmoid activation, If we initialize weights with large values (...
2
votes
1
answer
99
views
Tensorflow Keras CV tutorials incorrectly working [closed]
I'm starting learning Machine Learning by Tensorflow Keras CV tutorials and two of my tutorial projects are not working correctly.
I use these tutorials:
https://www.tensorflow.org/tutorials/load_data/...
- 41