Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update model analysis docs #985

Merged
merged 1 commit into from
Dec 30, 2024
Merged

Update model analysis docs #985

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
94 changes: 67 additions & 27 deletions model_analysis_docs/Models/albert/pt_albert_base_v1_masked_lm.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -90,43 +90,83 @@
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 1, 1, 128), dtype=int64)</td>
<td>dtype : torch.float32</td>
<td>Operand(type=Parameter, shape=(30000, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(30000, 128), dtype=float32)</td>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 128, 128), dtype=bfloat16)</td>
<td>dtype : torch.float32</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Parameter, shape=(2, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(2, 128), dtype=float32)</td>
<td>Cast</td>
<td>Operand(type=Parameter, shape=(512, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 1, 1, 128), dtype=int64)</td>
<td>dtype : torch.float32</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(512, 128), dtype=float32)</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(30000, 128), dtype=bfloat16)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(2, 128), dtype=bfloat16)</td>
<td></td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(512, 128), dtype=bfloat16)</td>
<td></td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Gelu</td>
Expand Down Expand Up @@ -186,7 +226,7 @@
<td>&#x274C;</td>
<td>&#x274C;</td>
<td></td>
<td>[FORGE][Runtime Data mismatch] RuntimeError Tensor data type mismatch: expected got</td>
<td>[FORGE][Runtime Datatype mismatch] RuntimeError Tensor data type mismatch: expected got</td>
</tr>
<tr>
<td>Index</td>
Expand All @@ -196,7 +236,7 @@
<td>&#x274C;</td>
<td>&#x274C;</td>
<td></td>
<td>[FORGE][Runtime Data mismatch] RuntimeError Tensor data type mismatch: expected got</td>
<td>[FORGE][Runtime Datatype mismatch] RuntimeError Tensor data type mismatch: expected got</td>
</tr>
<tr>
<td>Layernorm</td>
Expand Down Expand Up @@ -242,11 +282,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(128, 768), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(768, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Matmul</td>
Expand All @@ -272,11 +312,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(1, 128, 768), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(768, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Matmul</td>
Expand All @@ -292,11 +332,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(1, 128, 3072), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(3072, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
<td>[TT_METAL][TT-Metal vs Forge Output Data mismatch] ValueError Data mismatch -> AutomaticValueChecker (compare_with_golden): framework_model , compiled_model</td>
</tr>
<tr>
<td>Matmul</td>
Expand Down
94 changes: 67 additions & 27 deletions model_analysis_docs/Models/albert/pt_albert_base_v1_token_cls.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,43 +80,83 @@
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 1, 1, 128), dtype=int64)</td>
<td>dtype : torch.float32</td>
<td>Operand(type=Parameter, shape=(30000, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(30000, 128), dtype=float32)</td>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 128, 128), dtype=bfloat16)</td>
<td>dtype : torch.float32</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Parameter, shape=(2, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(2, 128), dtype=float32)</td>
<td>Cast</td>
<td>Operand(type=Parameter, shape=(512, 128), dtype=float32)</td>
<td>dtype : torch.bfloat16</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
</tr>
<tr>
<td>Cast</td>
<td>Operand(type=Activation, shape=(1, 1, 1, 128), dtype=int64)</td>
<td>dtype : torch.float32</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Parameter, shape=(512, 128), dtype=float32)</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(30000, 128), dtype=bfloat16)</td>
<td></td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td>[TT_METAL][ttnn.embedding validation] RuntimeError tt-metal/ttnn/cpp/ttnn/operations/embedding/device/embedding_device_operation.cpp a.get_dtype() == DataType::UINT32 or a.get_dtype() == DataType::BFLOAT16 Input must be UINT32 or BFLOAT16</td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(2, 128), dtype=bfloat16)</td>
<td></td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Embedding</td>
<td>Operand(type=Activation, shape=(1, 128), dtype=int64)<br><div align='center'>X</div>Operand(type=Activation, shape=(512, 128), dtype=bfloat16)</td>
<td></td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Gelu</td>
Expand Down Expand Up @@ -166,7 +206,7 @@
<td>&#x274C;</td>
<td>&#x274C;</td>
<td></td>
<td>[FORGE][Runtime Data mismatch] RuntimeError Tensor data type mismatch: expected got</td>
<td>[FORGE][Runtime Datatype mismatch] RuntimeError Tensor data type mismatch: expected got</td>
</tr>
<tr>
<td>Index</td>
Expand All @@ -176,7 +216,7 @@
<td>&#x274C;</td>
<td>&#x274C;</td>
<td></td>
<td>[FORGE][Runtime Data mismatch] RuntimeError Tensor data type mismatch: expected got</td>
<td>[FORGE][Runtime Datatype mismatch] RuntimeError Tensor data type mismatch: expected got</td>
</tr>
<tr>
<td>Layernorm</td>
Expand Down Expand Up @@ -212,11 +252,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(128, 768), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(768, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Matmul</td>
Expand All @@ -242,11 +282,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(1, 128, 768), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(768, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
</tr>
<tr>
<td>Matmul</td>
Expand All @@ -262,11 +302,11 @@
<td>Matmul</td>
<td>Operand(type=Activation, shape=(1, 128, 3072), dtype=float32)<br><div align='center'>X</div>Operand(type=Activation, shape=(3072, 768), dtype=float32)</td>
<td></td>
<td>&#x2705;</td>
<td>&#x2705;</td>
<td>&#x274C;</td>
<td></td>
<td></td>
<td></td>
<td>&#xFFFD;</td>
<td></td>
<td>[TT_METAL][TT-Metal vs Forge Output Data mismatch] ValueError Data mismatch -> AutomaticValueChecker (compare_with_golden): framework_model , compiled_model</td>
</tr>
<tr>
<td>Matmul</td>
Expand Down
Loading
Loading