[Tracking] TorchDynamo ONNX exporter issues

[BowenBao]

Export Error Summary Dashboard

• This report is generated from branch of [RFC][WIP] Introduce PT2 dynamo onnx exporter #1712
• Produced by RUN_SLOW=1 pytest tests/exporters/onnx -k "test_exporters_cli_pytorch_cpu" -s -vvvvv -n 8 --disable-warnings
• The error aggregation report is created by calling

  from torch.onnx import testing
  
  testing.summarize_and_report_errors()

  by running the unittests and above script under PyTorch branch [ONNX] Add error dumping and aggregation feature in testing pytorch/pytorch#125929, where error dumping and aggregation utility is being added for onnx export.

=== 366 failed, 348 passed, 11 skipped, 16035 warnings in 4422.00s (1:13:41) ===

Summary Grouped by Error

See more

onnxruntime.capi.onnxruntime_pybind11_state.Fail (67 total):

• [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp7v398t9w/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())

        hf-internal-testing/tiny-random-CodeGenModel
        hf-internal-testing/tiny-random-BeitForImageClassification
        mohitsha/tiny-random-testing-bert2gpt2
        hf-internal-testing/tiny-random-gpt2
        hf-internal-testing/tiny-random-GPTNeoModel
        hf-internal-testing/tiny-random-VisionEncoderDecoderModel-vit-gpt2
        hf-internal-testing/tiny-random-DistilBertModel
        hf-internal-testing/tiny-random-GPTJModel
        hf-internal-testing/tiny-random-Data2VecVisionModel
        hf-internal-testing/tiny-random-GPTBigCodeModel

• [ONNXRuntimeError] : 1 : FAIL : This is an invalid model. Error: Duplicate definition of name (_inlfunc_transformers_models_roformer_modeling_roformer_RoFormerSelfAttention_orig_forward___self___encoder_layer_0_attention_self_1__val_47_split_0_unsqueeze).

        hf-internal-testing/tiny-random-RoFormerModel

FXE0010: fx-pass[ERROR] (49 total):

• Running Functionalize pass.

        hf-internal-testing/tiny-random-SpeechT5ForTextToSpeech
        hf-internal-testing/tiny-random-TableTransformerModel
        hf-internal-testing/tiny-random-wavlm
        timm/hrnet_w18.ms_aug_in1k
        hf-internal-testing/tiny-random-WavLMForCTC
        hf-internal-testing/tiny-random-BloomModel
        hf-internal-testing/tiny-random-YolosModel
        hf-internal-testing/tiny-random-DetrModel
        hf-internal-testing/tiny-random-ImageGPTModel
        hf-internal-testing/tiny-random-WavLMForXVector
        hf-internal-testing/tiny-random-GPTNeoXForCausalLM

• Running InsertTypePromotion pass.

        hf-internal-testing/tiny-random-SEWDModel

FXE0012: unsupported-fx-node-analysis[ERROR] (35 total):

• Unsupported FX nodes: {'call_function': ['aten.std.correction']}.

        hf-internal-testing/tiny-random-Wav2Vec2Model
        hf-internal-testing/tiny-random-UniSpeechSatForXVector
        hf-internal-testing/tiny-random-Data2VecAudioModel

• Unsupported FX nodes: {'call_function': ['<built-in function mod>']}.

        fxmarty/tiny-doc-qa-vision-encoder-decoder
        hf-internal-testing/tiny-random-SwinModel
        hf-internal-testing/tiny-random-DonutSwinModel

• Unsupported FX nodes: {'call_function': ['aten.sym_constrain_range.default', 'aten._assert_scalar.default']}.

        hf-internal-testing/tiny-random-XLMModel
        hf-internal-testing/tiny-random-flaubert

• Unsupported FX nodes: {'call_function': ['aten.glu.default']}.

        hf-internal-testing/tiny-random-wav2vec2-conformer

• Unsupported FX nodes: {'call_function': ['aten.glu.default', 'aten.std.correction']}.

        hf-internal-testing/tiny-random-wav2vec2-conformer

• Unsupported FX nodes: {'call_function': ['<built-in function mod>', 'aten.pixel_shuffle.default']}.

        hf-internal-testing/tiny-random-SwinModel

onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph (33 total):

• [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_3_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_3_shape",) , Error Node (_inlfunc_aten_sym_size|folded_3_n0)'s input 0 is marked single but has an empty string in the graph

        Ericwang/tiny-random-ast
        hf-internal-testing/tiny-random-Data2VecAudioModel
        hf-internal-testing/tiny-random-UniSpeechSatForPreTraining
        hf-internal-testing/tiny-random-CvTModel
        hf-internal-testing/tiny-random-unispeech
        hf-internal-testing/tiny-random-HubertModel
        hf-internal-testing/tiny-random-Wav2Vec2Model
        hf-internal-testing/tiny-random-unispeech-sat

Exception (21 total):

• The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 1, 16, 2]

        hf-internal-testing/tiny-random-mbart
        hf-internal-testing/tiny-random-t5
        sshleifer/tiny-marian-en-de
        lewtun/tiny-random-mt5
        hf-internal-testing/tiny-random-m2m_100
        microsoft/trocr-small-handwritten
        hf-internal-testing/tiny-random-PegasusModel
        hf-internal-testing/tiny-random-bart
        hf-internal-testing/tiny-random-BlenderbotModel

onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument (21 total):

• [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))

        hf-internal-testing/tiny-random-MptForCausalLM
        hf-internal-testing/tiny-stable-diffusion-torch
        fxmarty/really-tiny-falcon-testing
        echarlaix/tiny-random-stable-diffusion-xl
        fxmarty/pix2struct-tiny-random
        fxmarty/tiny-testing-falcon-alibi
        hf-internal-testing/tiny-random-GPTBigCodeModel

• [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Invalid rank for input: bbox Got: 3 Expected: 2 Please fix either the inputs/outputs or the model.

        hf-internal-testing/tiny-random-LayoutLMv3Model

FXE0015: fx-node-insert-type-promotion[ERROR] (5 total):

• Applied type promotion for div_1.

        hf-internal-testing/tiny-random-SEWDModel

FXE0011: no-symbolic-function-for-call-function[ERROR] (3 total):

• Unknown call_function target: <function _sym_sqrt at 0x7f70b92496c0>

        hf-internal-testing/tiny-random-DPTModel

Summary Grouped by Model

See more

hf-internal-testing/tiny-random-BloomModel (10 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-SEWDModel (10 total):

• FXE0015: fx-node-insert-type-promotion[ERROR]

        Applied type promotion for div_1. 

• FXE0010: fx-pass[ERROR]

        Running InsertTypePromotion pass. 

hf-internal-testing/tiny-random-gpt2 (10 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpiwzky7zf/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmprp8vxxik/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpul44gh3y/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpl9pvboia/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpcvw8w9lf/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp6de0g3hn/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpb38g_cjk/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpfh402mbq/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpy26k2poj/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp83qhih35/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-GPTJModel (10 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmplizuo_dw/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp3s4iiaoo/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmphiy8f9h3/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpedt5h21b/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpa74e88hd/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpczhzc1j9/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpy9u2jqev/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpp38vr6hk/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp9_11kmyt/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp8v5vi7t2/model.onnx failed:ONNX Schema transformers_models_gptj_modeling_gptj_GPTJAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-GPTNeoModel (9 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp_42s7wu8/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp91hew9ew/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___transformer_h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpv81kg1e7/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___transformer_h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp8t3vt14t/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmprsa34v0w/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___transformer_h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp8ahh1_86/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___transformer_h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpwzixyjr5/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___transformer_h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp2tjrrdyb/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp02zgz4gf/model.onnx failed:ONNX Schema transformers_models_gpt_neo_modeling_gpt_neo_GPTNeoSelfAttention_orig_forward___self___h_0_attn_attention_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-GPTNeoXForCausalLM (9 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

fxmarty/tiny-doc-qa-vision-encoder-decoder (9 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['<built-in function mod>']}. 

hf-internal-testing/tiny-random-CodeGenModel (8 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpyjflchsr/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpq1wd4yuh/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpxipgd6s9/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpzhojqtbd/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpa1f5b8ex/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmptjerqsn7/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpcb72_kpm/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp7v398t9w/model.onnx failed:ONNX Schema transformers_models_codegen_modeling_codegen_CodeGenAttention_orig_forward___self___h_0_attn_1: failed validating the check: !(it.GetName().empty())

fxmarty/pix2struct-tiny-random (7 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))

hf-internal-testing/tiny-random-Data2VecAudioModel (7 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_2_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_2_shape",) , Error Node (_inlfunc_aten_sym_size|folded_2_n0)'s input 0 is marked single but has an empty string in the graph

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.std.correction']}. 

hf-internal-testing/tiny-random-RoFormerModel (7 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : This is an invalid model. Error: Duplicate definition of name (_inlfunc_transformers_models_roformer_modeling_roformer_RoFormerSelfAttention_orig_forward___self___roformer_encoder_layer_0_attention_self_1__val_47_split_0_unsqueeze).
        [ONNXRuntimeError] : 1 : FAIL : This is an invalid model. Error: Duplicate definition of name (_inlfunc_transformers_models_roformer_modeling_roformer_RoFormerSelfAttention_orig_forward___self___encoder_layer_0_attention_self_1__val_47_split_0_unsqueeze).

hf-internal-testing/tiny-random-wav2vec2-conformer (7 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.glu.default']}. 
        Unsupported FX nodes: {'call_function': ['aten.glu.default', 'aten.std.correction']}. 

hf-internal-testing/tiny-random-Wav2Vec2Model (7 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_3_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_3_shape",) , Error Node (_inlfunc_aten_sym_size|folded_3_n0)'s input 0 is marked single but has an empty string in the graph

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.std.correction']}. 

hf-internal-testing/tiny-random-DistilBertModel (7 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpdmgbdle6/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpmxs0po84/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___distilbert_transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpcw8vqcvy/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___distilbert_transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp3mvo02ej/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___distilbert_transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpy26bpfcw/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpda2rsw8c/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___distilbert_transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpmxntee_c/model.onnx failed:ONNX Schema transformers_models_distilbert_modeling_distilbert_FFN_orig_forward___self___distilbert_transformer_layer_0_ffn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-GPTBigCodeModel (6 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp0n0o2ijk/model.onnx failed:ONNX Schema transformers_models_gpt_bigcode_modeling_gpt_bigcode_GPTBigCodeSdpaAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp4p90x_74/model.onnx failed:ONNX Schema transformers_models_gpt_bigcode_modeling_gpt_bigcode_GPTBigCodeSdpaAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int64)) , expected: (tensor(float))
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpqa4ms0c4/model.onnx failed:ONNX Schema transformers_models_gpt_bigcode_modeling_gpt_bigcode_GPTBigCodeSdpaAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpf_1e1d9_/model.onnx failed:ONNX Schema transformers_models_gpt_bigcode_modeling_gpt_bigcode_GPTBigCodeSdpaAttention_orig_forward___self___transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-unispeech (5 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_2_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_2_shape",) , Error Node (_inlfunc_aten_sym_size|folded_2_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-wavlm (5 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-XLMModel (5 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.sym_constrain_range.default', 'aten._assert_scalar.default']}. 

hf-internal-testing/tiny-random-flaubert (5 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.sym_constrain_range.default', 'aten._assert_scalar.default']}. 

hf-internal-testing/tiny-random-HubertModel (5 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_3_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_3_shape",) , Error Node (_inlfunc_aten_sym_size|folded_3_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-unispeech-sat (4 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_2_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_2_shape",) , Error Node (_inlfunc_aten_sym_size|folded_2_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-BlenderbotModel (4 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

hf-internal-testing/tiny-random-DetrModel (4 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-SwinModel (4 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['<built-in function mod>', 'aten.pixel_shuffle.default']}. 
        Unsupported FX nodes: {'call_function': ['<built-in function mod>']}. 

hf-internal-testing/tiny-random-VisionEncoderDecoderModel-vit-gpt2 (4 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp4h4xrx0r/decoder_model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp3sjnjrvc/decoder_model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpoyc0lyic/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpq_xdrjqw/decoder_model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-YolosModel (3 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-Data2VecVisionModel (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpx1u3hy53/model.onnx failed:ONNX Schema transformers_models_data2vec_modeling_data2vec_vision_Data2VecVisionEmbeddings_orig_forward___self___data2vec_vision_embeddings_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpo8wp3kp2/model.onnx failed:ONNX Schema transformers_models_data2vec_modeling_data2vec_vision_Data2VecVisionEmbeddings_orig_forward___self___embeddings_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp4im7xeay/model.onnx failed:ONNX Schema transformers_models_data2vec_modeling_data2vec_vision_Data2VecVisionEmbeddings_orig_forward___self___embeddings_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-m2m_100 (3 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

hf-internal-testing/tiny-random-BeitForImageClassification (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmptqokkbrh/model.onnx failed:ONNX Schema transformers_models_beit_modeling_beit_BeitEmbeddings_orig_forward___self___beit_embeddings_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpk6jidulq/model.onnx failed:ONNX Schema transformers_models_beit_modeling_beit_BeitEmbeddings_orig_forward___self___embeddings_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmpdn8_ixl6/model.onnx failed:ONNX Schema transformers_models_beit_modeling_beit_BeitEmbeddings_orig_forward___self___beit_embeddings_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-DPTModel (3 total):

• FXE0011: no-symbolic-function-for-call-function[ERROR]

        Unknown call_function target: <function _sym_sqrt at 0x7f70b92496c0>

hf-internal-testing/tiny-random-MptForCausalLM (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))

lewtun/tiny-random-mt5 (3 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

fxmarty/really-tiny-falcon-testing (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int64)) , expected: (tensor(float))

fxmarty/tiny-testing-falcon-alibi (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(float)) , expected: (tensor(int64))

hf-internal-testing/tiny-random-t5 (3 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 8]

hf-internal-testing/tiny-random-TableTransformerModel (3 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-ImageGPTModel (3 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

Ericwang/tiny-random-ast (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_7_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_7_shape",) , Error Node (_inlfunc_aten_sym_size|folded_7_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-CvTModel (3 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_17_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_17_shape",) , Error Node (_inlfunc_aten_sym_size|folded_17_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-mbart (2 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

hf-internal-testing/tiny-random-WavLMForXVector (2 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

timm/hrnet_w18.ms_aug_in1k (2 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-PegasusModel (2 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

mohitsha/tiny-random-testing-bert2gpt2 (2 total):

• onnxruntime.capi.onnxruntime_pybind11_state.Fail

        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp6ct6ewo8/model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())
        [ONNXRuntimeError] : 1 : FAIL : Load model from /tmp/tmp48ktht0x/decoder_model.onnx failed:ONNX Schema transformers_models_gpt2_modeling_gpt2_GPT2Attention_orig_forward___self___decoder_transformer_h_0_attn_1: failed validating the check: !(it.GetName().empty())

hf-internal-testing/tiny-random-WavLMForCTC (2 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

hf-internal-testing/tiny-random-DonutSwinModel (2 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['<built-in function mod>']}. 

hf-internal-testing/tiny-random-bart (2 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 4, 16, 4]

sshleifer/tiny-marian-en-de (1 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 1, 16, 2]

hf-internal-testing/tiny-random-UniSpeechSatForPreTraining (1 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidGraph

        [ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. In Node, ("_inlfunc_aten_sym_size|folded_3_n0", Shape, "", -1) : ("",) -> ("_inlfunc_aten_sym_size|folded_3_shape",) , Error Node (_inlfunc_aten_sym_size|folded_3_n0)'s input 0 is marked single but has an empty string in the graph

hf-internal-testing/tiny-random-UniSpeechSatForXVector (1 total):

• FXE0012: unsupported-fx-node-analysis[ERROR]

        Unsupported FX nodes: {'call_function': ['aten.std.correction']}. 

hf-internal-testing/tiny-stable-diffusion-torch (1 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int64)) , expected: (tensor(int32))

echarlaix/tiny-random-stable-diffusion-xl (1 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int64)) , expected: (tensor(int32))

hf-internal-testing/tiny-random-LayoutLMv3Model (1 total):

• onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument

        [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Invalid rank for input: bbox Got: 3 Expected: 2 Please fix either the inputs/outputs or the model.

hf-internal-testing/tiny-random-SpeechT5ForTextToSpeech (1 total):

• FXE0010: fx-pass[ERROR]

        Running Functionalize pass. 

microsoft/trocr-small-handwritten (1 total):

• Exception

        The post-processing of the ONNX export failed. The export can still be performed by passing the option --no-post-process. Detailed error: Unable to merge decoders. Detailed error: Expected at least one dynamic input shape for the output present.0.encoder.key, found a static shape: [2, 8, 16, 32]

cc @thiagocrepaldi, @fxmarty