{
  "id": 519175,
  "title": "keep getting xla problem and cant train",
  "url": "/competitions/leap-atmospheric-physics-ai-climsim/discussion/519175",
  "author_name": "Johnsmz",
  "post_date": "2024-07-10T03:15:34.643000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>`UnimplementedError: Graph execution error:</p>\n<p>Detected at node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad defined at (most recent call last):<br>\n<br>\nGPU MaxPool gradient ops do not yet have a deterministic XLA implementation.<br>\n     [[{{node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad}}]]<br>\n    tf2xla conversion failed while converting __inference_one_step_on_data_45556[]. Run with TF_DUMP_GRAPH_PREFIX=/path/to/dump/dir and --vmodule=xla_compiler=2 to obtain a dump of the compiled functions.<br>\n     [[StatefulPartitionedCall]] [Op:__inference_one_step_on_iterator_46783]`</p>\n<p>i am keep getting xla problem and cant train my model,<br>\nenvironment   tf 2.16<br>\ncuda 12.5</p>",
  "messages": [
    {
      "id": 2914560,
      "postDate": "2024-07-10T03:15:34.643Z",
      "content": "<p>`UnimplementedError: Graph execution error:</p>\n<p>Detected at node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad defined at (most recent call last):<br>\n<br>\nGPU MaxPool gradient ops do not yet have a deterministic XLA implementation.<br>\n     [[{{node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad}}]]<br>\n    tf2xla conversion failed while converting __inference_one_step_on_data_45556[]. Run with TF_DUMP_GRAPH_PREFIX=/path/to/dump/dir and --vmodule=xla_compiler=2 to obtain a dump of the compiled functions.<br>\n     [[StatefulPartitionedCall]] [Op:__inference_one_step_on_iterator_46783]`</p>\n<p>i am keep getting xla problem and cant train my model,<br>\nenvironment   tf 2.16<br>\ncuda 12.5</p>",
      "rawMarkdown": "`UnimplementedError: Graph execution error:\n\nDetected at node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad defined at (most recent call last):\n<stack traces unavailable>\nGPU MaxPool gradient ops do not yet have a deterministic XLA implementation.\n\t [[{{node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad}}]]\n\ttf2xla conversion failed while converting __inference_one_step_on_data_45556[]. Run with TF_DUMP_GRAPH_PREFIX=/path/to/dump/dir and --vmodule=xla_compiler=2 to obtain a dump of the compiled functions.\n\t [[StatefulPartitionedCall]] [Op:__inference_one_step_on_iterator_46783]`\n\ni am keep getting xla problem and cant train my model,\nenvironment   tf 2.16\ncuda 12.5"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2914560": "`UnimplementedError: Graph execution error:\n\nDetected at node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad defined at (most recent call last):\n<stack traces unavailable>\nGPU MaxPool gradient ops do not yet have a deterministic XLA implementation.\n\t [[{{node gradient_tape/functional_1_1/functional_1/max_pooling1d_2_1/MaxPool1d/MaxPoolGrad}}]]\n\ttf2xla conversion failed while converting __inference_one_step_on_data_45556[]. Run with TF_DUMP_GRAPH_PREFIX=/path/to/dump/dir and --vmodule=xla_compiler=2 to obtain a dump of the compiled functions.\n\t [[StatefulPartitionedCall]] [Op:__inference_one_step_on_iterator_46783]`\n\ni am keep getting xla problem and cant train my model,\nenvironment   tf 2.16\ncuda 12.5"
  }
}