[TVM Integration] Support TVM conversion with FP16 data type

[Xingjian Shi]

Description

We fixed TVM integration in GluonNLP recently for fp32 dtype. However, we still do not support fp16 dtype. We should

• Revise the test to add FP16:

  gluon-nlp/tests/test_models.py

  Lines 65 to 77 in e8d4c8a

   @pytest.mark.serial 
   @pytest.mark.seed(123) 
   @pytest.mark.parametrize('model_name', 
                            ['google_albert_base_v2', 
                             'google_en_cased_bert_base', 
                             'google_electra_small', 
                             'fairseq_bart_base']) 
   @pytest.mark.parametrize('batch_size,seq_length', [(2, 4), (1, 4)]) 
   @pytest.mark.parametrize('layout', ['NT', 'TN']) 
   @pytest.mark.skipif(not tvm_enabled(), 
                       reason='TVM is not supported. So this test is skipped.') 
   # @pytest.mark.skip('TVM issue https://github.com/dmlc/gluon-nlp/issues/1425.') 
   def test_tvm_integration(model_name, batch_size, seq_length, layout, ctx): 

• Revise the benchmark to add TVM + FP16:

  gluon-nlp/scripts/benchmarks/benchmark_utils.py

  Lines 610 to 623 in e8d4c8a

   def compile_tvm_graph_runtime(model, model_name, layout, compute_layout, 
                                 batch_size, seq_length, dtype, instance_type): 
       key = (model_name, layout, compute_layout, batch_size, seq_length, dtype, instance_type) 
       if key in _TVM_RT_CACHE: 
           return _TVM_RT_CACHE[key] 
       tvm = try_import_tvm() 
       from tvm import relay 
       from tvm.contrib import graph_runtime 
       from gluonnlp.utils.tvm_utils import get_ec2_tvm_flags, update_tvm_convert_map 
       flags = get_ec2_tvm_flags()[instance_type] 
       update_tvm_convert_map() 
       token_ids_shape = (batch_size, seq_length) if layout == 'NT' else (seq_length, batch_size) 
       valid_length_shape = (batch_size,) 
       if 'bart' in model_name: