Comments (1)
Hi,
Sorry for my late reply.
I only keep this training log of captioning based on X-VLM(16M):
{"train_lr": "0.00001", "train_loss": "3.09465", "test_Bleu_1": 0.7888877725308794, "test_Bleu_2": 0.63571035330044, "test_Bleu_3": 0.5010730538646152, "test_Bleu_4": 0.3913670270575186, "test_METEOR": 0.30496697421985314, "test_ROUGE_L": 0.5970224740919123, "test_CIDEr": 1.317922681426818, "test_SPICE": 0.23711267501362399, "epoch": 0}
{"train_lr": "0.00001", "train_loss": "2.98785", "test_Bleu_1": 0.7908520616358936, "test_Bleu_2": 0.6381588587372502, "test_Bleu_3": 0.5034545904748693, "test_Bleu_4": 0.3940724470046297, "test_METEOR": 0.3044191834423089, "test_ROUGE_L": 0.5968370063641272, "test_CIDEr": 1.324427769391428, "test_SPICE": 0.23648371743116095, "epoch": 1}
{"train_lr": "0.00001", "train_loss": "2.95646", "test_Bleu_1": 0.7900966620445572, "test_Bleu_2": 0.6394082741826733, "test_Bleu_3": 0.5067956381747537, "test_Bleu_4": 0.3990826836166066, "test_METEOR": 0.30812466366805014, "test_ROUGE_L": 0.6003008324478327, "test_CIDEr": 1.335514654758239, "test_SPICE": 0.23785805830856616, "epoch": 2}
{"train_lr": "0.00000", "train_loss": "2.93689", "test_Bleu_1": 0.7885322668167135, "test_Bleu_2": 0.6375335548575466, "test_Bleu_3": 0.5049943737820617, "test_Bleu_4": 0.3974852071709018, "test_METEOR": 0.3076980686063203, "test_ROUGE_L": 0.5984061051446633, "test_CIDEr": 1.332406512277844, "test_SPICE": 0.23666827912166746, "epoch": 3}
{"train_lr": "0.00000", "train_loss": "2.92356", "test_Bleu_1": 0.7897854626961097, "test_Bleu_2": 0.6385248029045398, "test_Bleu_3": 0.5062238463521957, "test_Bleu_4": 0.3988569646467052, "test_METEOR": 0.30886155569362295, "test_ROUGE_L": 0.6002198552354895, "test_CIDEr": 1.3395548597453977, "test_SPICE": 0.2384228247143562, "epoch": 4}
from x-vlm.
Related Issues (20)
- Fine-tuning HOT 1
- Drawing Attention Heatmap HOT 1
- Training log for the pretrain stage HOT 2
- Script to generate RegionTextJsonDataset? HOT 4
- add web demo/models/datasets to ICML organization on Hugging Face
- inferece api for referring expression comprehension
- About swin_B_480 HOT 1
- Fine-tune on VQA HOT 2
- Performance of different vision encoders HOT 1
- About batch sampling `iter_perc` HOT 1
- Finetuning On NLVR2 HOT 1
- NLVR Pretrain HOT 1
- The code saves the best testing results on Image-Text Retrieval
- Code for Grad-CAM visualization HOT 2
- VQA: Understanding how the model provides us an answer? Need of answer list? HOT 1
- VQA: Limitations in questions and answers HOT 1
- apply an entire BERT as text encoder HOT 1
- Dear Author,Is there any inference code for image retrieval?How can I use this project to inference on my own image-text pairs. HOT 1
- About license HOT 2
- pretrain-base-4m for the X_VLM HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from x-vlm.