fix reduce_mean_op_xpu op bug for cvrq #7
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
fix reduce_mean_op_xpu op bug for cvrq:
gtest_broadcast_mul(api::kXPU2, "GM_0", "GM", "GM_0", {287, 990, 8}, {287, 990});
ASSERT-FAIL[src/wrapper/math_broadcast_calc_common.cpp:587]
ExternalError: XPU broadcast_mul kernel return wrong value[1 xpu api invalid param]
[Hint: Expected r == XPU_SUCCESS, but received r:1 != XPU_SUCCESS:0.] (at /workspace/workspace/Paddle/paddle/fluid/operators/reduce_ops/reduce_mean_op_xpu.cc:138)
[operator < reduce_mean_grad > error]. (at /workspace/workspace/Paddle/paddle/fluid/framework/threadpool.h:46)
Compile Traceback (most recent call last):
File "./framework/train.py", line 352, in
update_model = model.Model(slot, False, config.update_model_lr, param_sgd_lr_map)
File "./model.py", line 167, in init
self._base_net(self.slot_data, self.label_cast)
File "./model.py", line 640, in _base_net
self._update_net()
File "./model.py", line 422, in _update_net
self.ctr_squeeze_input = self._squeeze_inputs(self.ctr_base_normalized, self.ctr_slot_stop[0]-self.ctr_slot_start[0], 11, ["mean", "max"], 8)
File "./model.py", line 364, in _squeeze_inputs
squeeze_tensor = paddle.mean(inputs_group, axis=2, keepdim=False)
PR types
PR changes
Describe