mirror of
https://github.com/nod-ai/AMD-SHARK-Studio.git
synced 2026-04-03 03:00:17 -04:00
96b1560da5
* Fix sharktank generation and add batch_size pytest option for torch. * Disable torch dynamo until py3.11 supported * Compile torchmodel without dynamo if torch.compile fails * Use release versions of TF/Keras for importer. * Pin torchvision and remove debug prints. * Remove duplicates from torch model list. * Update generate_sharktank.py * xfail a few models that fail sharktank generation/ numerics
52 lines
723 B
Plaintext
52 lines
723 B
Plaintext
-f https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html
|
|
--pre
|
|
|
|
numpy>1.22.4
|
|
pytorch-triton
|
|
torchvision==0.16.0.dev20230322
|
|
tabulate
|
|
|
|
tqdm
|
|
|
|
#iree-compiler | iree-runtime should already be installed
|
|
iree-tools-tflite
|
|
iree-tools-xla
|
|
iree-tools-tf
|
|
|
|
# TensorFlow and JAX.
|
|
gin-config
|
|
tensorflow>2.11
|
|
keras
|
|
#tf-models-nightly
|
|
#tensorflow-text-nightly
|
|
transformers
|
|
diffusers
|
|
#tensorflow-probability
|
|
#jax[cpu]
|
|
|
|
|
|
# tflitehub dependencies.
|
|
Pillow
|
|
|
|
# Testing and support.
|
|
lit
|
|
pyyaml
|
|
python-dateutil
|
|
sacremoses
|
|
sentencepiece
|
|
|
|
# web dependecies.
|
|
gradio
|
|
altair
|
|
scipy
|
|
|
|
#ONNX and ORT for benchmarking
|
|
#--extra-index-url https://test.pypi.org/simple/
|
|
#protobuf
|
|
#coloredlogs
|
|
#flatbuffers
|
|
#sympy
|
|
#psutil
|
|
#onnx-weekly
|
|
#ort-nightly
|