Revert "chore(deps): upgrade trl (#537)" (#539)

This reverts commit 6fc9f46.

Signed-off-by: Dushyant Behl <dushyantbehl@in.ibm.com>

Author: dushyantbehl

docs/offline-data-preprocessing.md

@@ -37,8 +37,6 @@ python scripts/offline_data_processing.py \
 
 Additionally, once the offline data processing is complete, users can leverage the shards stored in `output_dir` for tuning by passing it through the `--training_data_path` flag or passing it via `data_paths` argument in data config yaml, provided they find the sharded datasets beneficial for training.
 
-**NOTE**: The offline data preprocessing script is not compatible with processing image datasets for vision models. 
-
 ## Example Usage
 ### Applying Chat Template
 

pyproject.toml

@@ -34,7 +34,7 @@ dependencies = [
 "sentencepiece>=0.1.99,<0.3",
 "tokenizers>=0.13.3,<1.0",
 "tqdm>=4.66.2,<5.0",
-"trl>=0.13,<0.18",
+"trl>=0.13,<0.17",
 "peft>=0.8.0,<0.14",
 "protobuf>=5.28.0,<6.0.0",
 "datasets>=2.15.0,<4.0",