Hugging Face's logo

fpaissan
/
tinyCLAP

contrastive learning
CLAP
audio classification
zero-shot classification
Model card Files Community
fpaissan commited on
Commit
c85e695
•
1 Parent(s): 619e7c2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -81,9 +81,10 @@ This command will automatically download the checkpoint, if present in the zoo o
81
 
82
  A list of available models with their computational cost is described in the follwing table:
83
 
84
- | alpha | beta | t0 | N | Params [M] | ESC-50 | UrbanSound8K | TUT17 |
85
- |:-----:|:----:|:--:|:-:|:----------:|:------:|:------------:|:-----:|
86
- | 1.5 | 0.75 | 6 | 7 | 4.4 | 77.3% | 69.7% | 21.9% |
 
87
 
88
  **Note**: I am currently computing the confidence intervals for these results. Due to seed variability and CUDA version mismatch, the numbers might be slightly different from those reported in the paper.
89
  The original paper's checkpoints are available through [this link](https://www.dropbox.com/scl/fi/e3aj76vxwlb4w6hs3mclv/tinyCLAP_results.zip?rlkey=7fl426tz1vf686oyosvja8i9s&dl=0).
 
81
 
82
  A list of available models with their computational cost is described in the follwing table:
83
 
84
+ | audioenc_name_student | Params [M] | ESC-50 | UrbanSound8K | TUT17 |
85
+ |:-----:|:----------:|:------:|:------------:|:-----:|
86
+ | Cnn14 | 82.8 | 81.3% | 72.3% | 23.7% |
87
+ | phinet_alpha_1.50_beta_0.75_t0_6_N_7 | 4.4 | 77.3% | 69.7% | 21.9% |
88
 
89
  **Note**: I am currently computing the confidence intervals for these results. Due to seed variability and CUDA version mismatch, the numbers might be slightly different from those reported in the paper.
90
  The original paper's checkpoints are available through [this link](https://www.dropbox.com/scl/fi/e3aj76vxwlb4w6hs3mclv/tinyCLAP_results.zip?rlkey=7fl426tz1vf686oyosvja8i9s&dl=0).