SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Point to the right file for hybrid CLIP (#12599)

Browse Source
This commit is contained in:
Eduardo Gonzalez Ponferrada
2021-07-11 23:46:22 -07:00
committed by GitHub
parent de23ecea36
commit 2dd9440d08
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/research_projects/jax-projects/hybrid_clip/README.md
View File

@@ -65,7 +65,7 @@ Next, let's add a symbolic link to the `run_hybrid_clip.py`.
```bash ```bash
export MODEL_DIR="./clip-roberta-base export MODEL_DIR="./clip-roberta-base
ln -s ~/transformers/examples/flax/summarization/run_hybrid_clip.py run_hybrid_clip.py ln -s ~/transformers/examples/research_projects/jax-projects/hybrid_clip/run_hybrid_clip.py run_hybrid_clip.py
``` ```
## How to use the `FlaxHybridCLIP` model: ## How to use the `FlaxHybridCLIP` model:
@@ -169,4 +169,4 @@ python run_hybrid_clip.py \
--push_to_hub --push_to_hub
``` ```
This should finish in ~1h50 mins with min validation loss 2.43. Training statistics can be accessed on [tfhub.de](https://tensorboard.dev/experiment/RUNPYd1yRgSD5kZSb9hDig/#scalars) This should finish in ~1h50 mins with min validation loss 2.43. Training statistics can be accessed on [tfhub.de](https://tensorboard.dev/experiment/RUNPYd1yRgSD5kZSb9hDig/#scalars)