hanzo-ml/trainer
1
0
Fork 0
mirror of https://github.com/hanzo-ml/trainer.git synced 2026-09-24 00:31:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(examples): use namespaced SQuAD dataset (#3560)

Browse Source 
* fix(examples): use namespaced SQuAD dataset

Signed-off-by: XploY04 <[email protected]>

* fix(runtimes): cap kagglehub below 1.0

Signed-off-by: XploY04 <[email protected]>

---------

Signed-off-by: XploY04 <[email protected]>
This commit is contained in:
Yash Agarwal
2026-05-28 23:21:00 +00:00
committed by GitHub
parent 4f8920c902
commit 0f4c4be96d
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
cmd/trainers/torchtune/requirements.txt
+1 -1
View File
@@ -1,4 +1,4 @@
torchao==0.15.0
torchtune==0.6.1
bitsandbytes>=0.49.2
kagglehub>=1.0.1
kagglehub<1.0.0
examples/pytorch/question-answering/fine-tune-distilbert.ipynb
+1 -1
View File
@@ -166,7 +166,7 @@
" )\n",
"\n",
" # Download the dataset and tokenizer\n",
" squad = load_dataset(\"squad\", split=\"train[:100]\") \n",
" squad = load_dataset(\"rajpurkar/squad\", split=\"train[:100]\") \n",
"\n",
" squad = squad.train_test_split(test_size=0.2, shuffle=False)\n",
" \n",