[GitHub] [tvm] cxx122 opened a new issue, #12373: [Bug] Misuse of reduce_axis

[GitBox <gi...@apache.org>]

cxx122 opened a new issue, #12373:
URL: https://github.com/apache/tvm/issues/12373

   ```
   REDUCE_0 = te.reduce_axis((5, 0), name="REDUCE_0")
   REDUCE_1 = te.reduce_axis((7, 0), name="REDUCE_1")
   TENSOR_0 = te.placeholder([10,10,10,10], dtype="int8", name="TENSOR_0")
   TENSOR_1 = te.compute([10,10], lambda mcq,uch:te.sum(expr=TENSOR_0[REDUCE_0,mcq,uch,REDUCE_1], axis=[REDUCE_0,REDUCE_1]), name ="TENSOR_1")
   ```
   The args in the reduce_axis should be like (0,5), the args like (5,0) will cause a wrong result.
   ### Expected behavior
   
   An error message to point out the misuse.
   
   ### Actual behavior
   
   ```
   Traceback (most recent call last):
     File "/Scuzer/src/bugs/IncorrectResult__f3f22bd7-feec-4f44-890c-2cd7d1c86814/Incorrect_bug.py", line 48, in <module>
       tvm.testing.assert_allclose(pre_list[1].numpy(), after_list[1].numpy(),rtol=1e-5)
     File "/Scuzer/tvm_cov_patch/tvm/python/tvm/testing/utils.py", line 114, in assert_allclose
       np.testing.assert_allclose(actual, desired, rtol=rtol, atol=atol, verbose=True)
     File "/root/miniconda3/envs/py38/lib/python3.8/site-packages/numpy/testing/_private/utils.py", line 1527, in assert_allclose
       assert_array_compare(compare, actual, desired, err_msg=str(err_msg),
     File "/root/miniconda3/envs/py38/lib/python3.8/site-packages/numpy/testing/_private/utils.py", line 840, in assert_array_compare
       raise AssertionError(msg)
   AssertionError: 
   Not equal to tolerance rtol=1e-05, atol=1e-07
   
   Mismatched elements: 100 / 100 (100%)
   Max absolute difference: 92
   Max relative difference: 1.
    x: array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
          [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
          [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],...
    y: array([[75, 68, 41, 69, 74, 68, 66, 63, 65, 72],
          [65, 67, 58, 88, 82, 74, 63, 87, 66, 66],
          [71, 69, 80, 69, 52, 71, 75, 51, 69, 72],...
   ```
   
   ### Environment
   
   Operating System: Ubuntu 18.04 TVM version: tag0.9.0 [d361585]
   
   ### Steps to reproduce
   
   ```
   import os
   import numpy as np
   import tvm
   from tvm import te, auto_scheduler, topi
   import tvm.testing
   
   REDUCE_0 = te.reduce_axis((0, 5), name="REDUCE_0")
   REDUCE_1 = te.reduce_axis((7, 0), name="REDUCE_1")
   TENSOR_0 = te.placeholder([10,10,10,10], dtype="int8", name="TENSOR_0")
   TENSOR_1 = te.compute([10,10], lambda mcq,uch:te.sum(expr=TENSOR_0[REDUCE_0,mcq,uch,REDUCE_1], axis=[REDUCE_0,REDUCE_1]), name ="TENSOR_1")
   s = te.create_schedule(TENSOR_1.op)
   tensor_list = [TENSOR_0,TENSOR_1]
   
   dev = tvm.cpu(0)
   pre_list = []
   after_list = []
   for tensor in tensor_list:
       shape = [x.value if 'value' in dir(x) and isinstance(x.value, int) else 1 for x in tensor.shape]
       params = (5*np.random.uniform(size=shape)).astype(tensor.dtype)
       pre_list.append(tvm.nd.array(params.copy(), dev))
       after_list.append(tvm.nd.array(params.copy(), dev))
   
   pre_mod = tvm.lower(s, tensor_list, simple_mode=True)
   with tvm.transform.PassContext(opt_level=4):
       f = tvm.build(pre_mod)
   f(*pre_list)
   
   # Schedule
   TENSOR_1_mcq, TENSOR_1_uch, TENSOR_1_REDUCE_0, TENSOR_1_REDUCE_1 = tuple(TENSOR_1.op.axis) + tuple(TENSOR_1.op.reduce_axis)
   TENSOR_1_REDUCE_0_REDUCE_1_fused = s[TENSOR_1].fuse(TENSOR_1_REDUCE_0, TENSOR_1_REDUCE_1)
   
   now_mod = tvm.lower(s, tensor_list, simple_mode=True)
   with tvm.transform.PassContext(opt_level=4):
       f = tvm.build(now_mod)
   f(*after_list)
   
   tvm.testing.assert_allclose(pre_list[1].numpy(), after_list[1].numpy(),rtol=1e-5)
   ```
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [tvm] cxx122 closed issue #12373: [Bug] Misuse of reduce_axis

["cxx122 (via GitHub)" <gi...@apache.org>]

cxx122 closed issue #12373: [Bug] Misuse of reduce_axis
URL: https://github.com/apache/tvm/issues/12373


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org