Code and Resources for the Transformer Encoder Reasoning Network (TERN)

Related tags

Deep Learning TERN
Overview

Transformer Encoder Reasoning Network

Code for the cross-modal visual-linguistic retrieval method from "Transformer Reasoning Network for Image-Text Matching and Retrieval", accepted to ICPR 2020 [Pre-print PDF].

This repo is built on top of VSE++.

Setup

  1. Clone the repo and move into it:
git clone https://github.com/mesnico/TERN
cd TERN
  1. Setup python environment using conda:
conda env create --file environment.yml
conda activate tern
export PYTHONPATH=.

Get the data

  1. Download and extract the data folder, containing COCO annotations, the splits by Karpathy et al. and ROUGEL - SPICE precomputed relevances:
wget http://datino.isti.cnr.it/tern/data.tar
tar -xvf data.tar
  1. Download the bottom-up features. We rearranged the ones provided by Anderson et al. in multiple .npy files, one for every image in the COCO dataset. This is beneficial during the dataloading phase. The following command extracts them under data/coco/. If you prefer another location, be sure to adjust the configuration file accordingly.
wget http://datino.isti.cnr.it/tern/features_36.tar
tar -xvf features_36.tar -C data/coco

Evaluate

Download our pre-trained TERN model:

wget http://datino.isti.cnr.it/tern/model_best_ndcg.pth

Then, issue the following commands for evaluating the model on the 1k (5fold cross-validation) or 5k test sets.

python3 test.py model_best_ndcg.pth --config configs/tern.yaml --size 1k
python3 test.py model_best_ndcg.pth --config configs/tern.yaml --size 5k

Train

In order to train the model using the basic TERN configuration, issue the following command:

python3 train.py --config configs/tern.yaml --logger_name runs/tern

runs/tern is where the output files (tensorboard logs, checkpoints) will be stored during this training session.

Reference

If you found this code useful, please cite the following paper:

@article{messina2020transformer,
  title={Transformer Reasoning Network for Image-Text Matching and Retrieval},
  author={Messina, Nicola and Falchi, Fabrizio and Esuli, Andrea and Amato, Giuseppe},
  journal={arXiv preprint arXiv:2004.09144},
  year={2020}
}

License

Apache License 2.0

Comments
  • Where is imdirect library used?

    Where is imdirect library used?

    Hi,

    Looks like imdirect library in your dependencies is archived in github, and is no longer in pip package. The author says we can replace imdirect by PIL function, so could you let me know where of TERN it is used in case I need to modify when running the code? Cannot identify the location with quick exploration.

    Thanks!

    opened by HireTheHero 4
  • about TERN and TERAN results.

    about TERN and TERAN results.

    In your TERAN paper (table1 and table2), you reported [TERN w. Align] results as higher than [original TERN] and lower than [TERAN], but I noticed that TERN and TERAN have very similar architectures but the shared-transformer configuration for the last TE layers. So if you train [TERN w. Align] with both TERN and TERAN objectives, and test it with TERAN inference protocol instead of TERN, shouldn't it be higher than TERAN?
    Thanks again for you work.

    opened by aaaa7788 2
  • Bert Pretrained weights loading

    Bert Pretrained weights loading

    Hello
    I'm trying to run the same experiment mentioned in the paper And I have noticed the following log message :

    2020-05-17 17:26:54,976 loading weights file https://cdn.huggingface.co/bert-base-uncased-pytorch_model.bin from cache at /home/ameenali/.cache/torch/transformers/f2ee78bdd635b758cc0a12352586868bef80e47401abe4c4fcc3832421e7338b.36ca03ab34a1a5d5fa7bc3d03d55c4fa650fed07220e2eeebc06ce58d0e9a157 2020-05-17 17:26:56,475 Weights from pretrained model not used in BertModel: ['bert.encoder.layer.6.attention.self.query.weight', 'bert.encoder.layer.6.attention.self.query.bias', 'bert.encoder.layer.6.attention.self.key.weight', 'bert.encoder.layer.6.attention.self.key.bias', 'bert.encoder.layer.6.attention.self.value.weight', 'bert.encoder.layer.6.attention.self.value.bias', 'bert.encoder.layer.6.attention.output.dense.weight', 'bert.encoder.layer.6.attention.output.dense.bias', 'bert.encoder.layer.6.intermediate.dense.weight', 'bert.encoder.layer.6.intermediate.dense.bias', 'bert.encoder.layer.6.output.dense.weight', 'bert.encoder.layer.6.output.dense.bias', 'bert.encoder.layer.7.attention.self.query.weight', 'bert.encoder.layer.7.attention.self.query.bias', 'bert.encoder.layer.7.attention.self.key.weight', 'bert.encoder.layer.7.attention.self.key.bias', 'bert.encoder.layer.7.attention.self.value.weight', 'bert.encoder.layer.7.attention.self.value.bias', 'bert.encoder.layer.7.attention.output.dense.weight', 'bert.encoder.layer.7.attention.output.dense.bias', 'bert.encoder.layer.7.intermediate.dense.weight', 'bert.encoder.layer.7.intermediate.dense.bias', 'bert.encoder.layer.7.output.dense.weight', 'bert.encoder.layer.7.output.dense.bias', 'bert.encoder.layer.8.attention.self.query.weight', 'bert.encoder.layer.8.attention.self.query.bias', 'bert.encoder.layer.8.attention.self.key.weight', 'bert.encoder.layer.8.attention.self.key.bias', 'bert.encoder.layer.8.attention.self.value.weight', 'bert.encoder.layer.8.attention.self.value.bias', 'bert.encoder.layer.8.attention.output.dense.weight', 'bert.encoder.layer.8.attention.output.dense.bias', 'bert.encoder.layer.8.intermediate.dense.weight', 'bert.encoder.layer.8.intermediate.dense.bias', 'bert.encoder.layer.8.output.dense.weight', 'bert.encoder.layer.8.output.dense.bias', 'bert.encoder.layer.9.attention.self.query.weight', 'bert.encoder.layer.9.attention.self.query.bias', 'bert.encoder.layer.9.attention.self.key.weight', 'bert.encoder.layer.9.attention.self.key.bias', 'bert.encoder.layer.9.attention.self.value.weight', 'bert.encoder.layer.9.attention.self.value.bias', 'bert.encoder.layer.9.attention.output.dense.weight', 'bert.encoder.layer.9.attention.output.dense.bias', 'bert.encoder.layer.9.intermediate.dense.weight', 'bert.encoder.layer.9.intermediate.dense.bias', 'bert.encoder.layer.9.output.dense.weight', 'bert.encoder.layer.9.output.dense.bias', 'bert.encoder.layer.10.attention.self.query.weight', 'bert.encoder.layer.10.attention.self.query.bias', 'bert.encoder.layer.10.attention.self.key.weight', 'bert.encoder.layer.10.attention.self.key.bias', 'bert.encoder.layer.10.attention.self.value.weight', 'bert.encoder.layer.10.attention.self.value.bias', 'bert.encoder.layer.10.attention.output.dense.weight', 'bert.encoder.layer.10.attention.output.dense.bias', 'bert.encoder.layer.10.intermediate.dense.weight', 'bert.encoder.layer.10.intermediate.dense.bias', 'bert.encoder.layer.10.output.dense.weight', 'bert.encoder.layer.10.output.dense.bias', 'bert.encoder.layer.11.attention.self.query.weight', 'bert.encoder.layer.11.attention.self.query.bias', 'bert.encoder.layer.11.attention.self.key.weight', 'bert.encoder.layer.11.attention.self.key.bias', 'bert.encoder.layer.11.attention.self.value.weight', 'bert.encoder.layer.11.attention.self.value.bias', 'bert.encoder.layer.11.attention.output.dense.weight', 'bert.encoder.layer.11.attention.output.dense.bias', 'bert.encoder.layer.11.intermediate.dense.weight', 'bert.encoder.layer.11.intermediate.dense.bias', 'bert.encoder.layer.11.output.dense.weight', 'bert.encoder.layer.11.output.dense.bias', 'bert.encoder.layer.6.attention.output.LayerNorm.weight', 'bert.encoder.layer.6.attention.output.LayerNorm.bias', 'bert.encoder.layer.6.output.LayerNorm.weight', 'bert.encoder.layer.6.output.LayerNorm.bias', 'bert.encoder.layer.7.attention.output.LayerNorm.weight', 'bert.encoder.layer.7.attention.output.LayerNorm.bias', 'bert.encoder.layer.7.output.LayerNorm.weight', 'bert.encoder.layer.7.output.LayerNorm.bias', 'bert.encoder.layer.8.attention.output.LayerNorm.weight', 'bert.encoder.layer.8.attention.output.LayerNorm.bias', 'bert.encoder.layer.8.output.LayerNorm.weight', 'bert.encoder.layer.8.output.LayerNorm.bias', 'bert.encoder.layer.9.attention.output.LayerNorm.weight', 'bert.encoder.layer.9.attention.output.LayerNorm.bias', 'bert.encoder.layer.9.output.LayerNorm.weight', 'bert.encoder.layer.9.output.LayerNorm.bias', 'bert.encoder.layer.10.attention.output.LayerNorm.weight', 'bert.encoder.layer.10.attention.output.LayerNorm.bias', 'bert.encoder.layer.10.output.LayerNorm.weight', 'bert.encoder.layer.10.output.LayerNorm.bias', 'bert.encoder.layer.11.attention.output.LayerNorm.weight', 'bert.encoder.layer.11.attention.output.LayerNorm.bias', 'bert.encoder.layer.11.output.LayerNorm.weight', 'bert.encoder.layer.11.output.LayerNorm.bias']

    Please notice the line saying : 2020-05-17 17:26:56,475 Weights from pretrained model not used in BertModel: however the training has continued regular , I'm afraid the weights are not inilaized by the pre trained bert model and thus won't get similar results, maybe the link is broken ?
    Thanks

    opened by AmeenAli 2
  • model_best_ndcg.pth.tar cannot be opened

    model_best_ndcg.pth.tar cannot be opened

    Hi,

    I downloaded model_best_ndcg.pth.tar from Drive, to test the model. However, the tar file seems to be corrupted, since it is not a valid tar file that can be opened. I get the following error : Error opening archive: Failed to open 'model_best_ndcg.pth.tar' Would it be possible to share a valid tar file?

    Thanks, Maurits

    opened by MauritsBleeker 1
  • Eval code clarification

    Eval code clarification

    Hello
    I have noticed in the following function :
    https://github.com/mesnico/TERN/blob/e147f8369ac80b71f77649f282fa92c7193263c1/evaluation.py#L197
    you are doing everything shifted by 5, would love to know why. Thanks!

    opened by AmeenAli 1
  • Deprecated PyPi Package imdirect

    Deprecated PyPi Package imdirect

    Hello there. Sadly the package imdirect is no longer available via pip and therefore the requirements can no longer be installed. It would be much appreciated if you would address this issue.

    opened by florianq-two 1
  • reproducing the VSRN paper

    reproducing the VSRN paper

    Hi, recently we are reproducing the VSRN paper.During the replication process, the running environment of the code was changed from Python 2.7 to Python 3.6, resulting in a much lower model accuracy.Do you have any good idea?Thank you very much!

    opened by DZZ95 1
  • Segmentation fault (core dumped)

    Segmentation fault (core dumped)

    After I configured the environment according to readme, I started this command "python3 test.py model_best_ndcg.pth --config configs/tern.yaml --size 1k ", I found this error "Segmentation fault (core dumped)". Then I tried debugging test.py and found that there was something wrong after the code"loaded_config = checkpoint['config']" , I couldn't print checkpoint['config'] and found this error "Segmentation fault (core dumped)". I would appreciate it if you told me how to solve it.

    opened by littlejerry411 6
Owner
Nicola Messina
PhD student at ISTI-CNR, Pisa, Italy. I'm interested in the secrets of intelligence and nature, and I'm passionate about Computer Vision and Deep Learning.
Nicola Messina
GitHub
Official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers, a novel method to visualize any Transformer-based network. Including examples for DETR, VQA.

PyTorch Implementation of Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers 1 Using Colab Please notic

Hila Chefer 489 Jan 7, 2023
Pytorch implementation of "A simple neural network module for relational reasoning" (Relational Networks)

Pytorch implementation of Relational Networks - A simple neural network module for relational reasoning Implemented & tested on Sort-of-CLEVR task. So

Kim Heecheol 800 Dec 5, 2022
An implementation of a sequence to sequence neural network using an encoder-decoder

Keras implementation of a sequence to sequence model for time series prediction using an encoder-decoder architecture. I created this post to share a

Luke Tonin 195 Dec 17, 2022
VSR-Transformer - This paper proposes a new Transformer for video super-resolution (called VSR-Transformer).

VSR-Transformer By Jiezhang Cao, Yawei Li, Kai Zhang, Luc Van Gool This paper proposes a new Transformer for video super-resolution (called VSR-Transf

Jiezhang Cao 225 Nov 13, 2022
Data and Code for ACL 2021 Paper "Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning"

Introduction Code and data for ACL 2021 Paper "Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning". We cons

Pan Lu 81 Dec 27, 2022
The code of “Similarity Reasoning and Filtration for Image-Text Matching” [AAAI2021]

SGRAF PyTorch implementation for AAAI2021 paper of “Similarity Reasoning and Filtration for Image-Text Matching”. It is built on top of the SCAN and C

Ronnie_IIAU 149 Dec 22, 2022
Code Repo for the ACL21 paper "Common Sense Beyond English: Evaluating and Improving Multilingual LMs for Commonsense Reasoning"

Common Sense Beyond English: Evaluating and Improving Multilingual LMs for Commonsense Reasoning This is the Github repository of our paper, "Common S

INK Lab @ USC 19 Nov 30, 2022
Code and data for "Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning" (EMNLP 2021).

GD-VCR Code for Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning (EMNLP 2021). Research Questions and Aims: How well can a model perform o

Da Yin 24 Oct 13, 2022
Code for "FGR: Frustum-Aware Geometric Reasoning for Weakly Supervised 3D Vehicle Detection", ICRA 2021

FGR This repository contains the python implementation for paper "FGR: Frustum-Aware Geometric Reasoning for Weakly Supervised 3D Vehicle Detection"(I

Yi Wei 31 Dec 8, 2022
The code for MM2021 paper "Multi-Level Counterfactual Contrast for Visual Commonsense Reasoning"

The Code for MM2021 paper "Multi-Level Counterfactual Contrast for Visual Commonsense Reasoning" Setting up and using the repo Get the dataset. Follow

null 4 Apr 20, 2022
This repository contains the data and code for the paper "Diverse Text Generation via Variational Encoder-Decoder Models with Gaussian Process Priors" (SPNLP@ACL2022)

GP-VAE This repository provides datasets and code for preprocessing, training and testing models for the paper: Diverse Text Generation via Variationa

Wanyu Du 18 Dec 29, 2022
A curated list of neural network pruning resources.

A curated list of neural network pruning and related resources. Inspired by awesome-deep-vision, awesome-adversarial-machine-learning, awesome-deep-learning-papers and Awesome-NAS.

Yang He 1.7k Jan 9, 2023
Code for the paper "Adversarial Generator-Encoder Networks"

This repository contains code for the paper "Adversarial Generator-Encoder Networks" (AAAI'18) by Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky. Pr

Dmitry Ulyanov 279 Jun 26, 2022
Deep Learning and Logical Reasoning from Data and Knowledge

Logic Tensor Networks (LTN) Logic Tensor Network (LTN) is a neurosymbolic framework that supports querying, learning and reasoning with both rich data

null 171 Dec 29, 2022
All the essential resources and template code needed to understand and practice data structures and algorithms in python with few small projects to demonstrate their practical application.

Data Structures and Algorithms Python INDEX 1. Resources - Books Data Structures - Reema Thareja competitiveCoding Big-O Cheat Sheet DAA Syllabus Inte

Shushrut Kumar 129 Dec 15, 2022
PyTorch implementation of "Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning"

Transparency-by-Design networks (TbD-nets) This repository contains code for replicating the experiments and visualizations from the paper Transparenc

David Mascharka 351 Nov 18, 2022
🌈 PyTorch Implementation for EMNLP'21 Findings "Reasoning Visual Dialog with Sparse Graph Learning and Knowledge Transfer"

SGLKT-VisDial Pytorch Implementation for the paper: Reasoning Visual Dialog with Sparse Graph Learning and Knowledge Transfer Gi-Cheon Kang, Junseok P

Gi-Cheon Kang 9 Jul 5, 2022
A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning

CLEVR Dataset Generation This is the code used to generate the CLEVR dataset as described in the paper: CLEVR: A Diagnostic Dataset for Compositional

Facebook Research 503 Jan 4, 2023
PyTorch implementations of the paper: "DR.VIC: Decomposition and Reasoning for Video Individual Counting, CVPR, 2022"

DRNet for Video Indvidual Counting (CVPR 2022) Introduction This is the official PyTorch implementation of paper: DR.VIC: Decomposition and Reasoning

tao han 35 Nov 22, 2022