Update on "[quant][graphmode][fx] Add support for dynamic quant for R…

…NN and RNNCell"

Summary:

Test Plan:
python test/test_quantization.py TestQuantizeFxOps.test_rnn
python test/test_quantization.py TestQuantizeFxOps.test_rnn_cell

Reviewers:

Subscribers:

Tasks:

Tags:

[ghstack-poisoned]

torch/quantization/fx/quantization_patterns.py

@@ -492,7 +492,7 @@ def convert(self, quantizer: QuantizerCls, node: Node, load_arg: Callable,
             load_arg(quantized=False)(emb_node.args),
             load_arg(quantized=False)(emb_node.kwargs))
 
-# TODO: merge with embedding quantize handler
+# TODO (maybe): merge with embedding quantize handler
 @register_quant_pattern(torch.nn.GRUCell)
 @register_quant_pattern(torch.nn.LSTMCell)
 @register_quant_pattern(torch.nn.RNNCell)