[ { "model_name_or_path": "elastic/distilbert-base-uncased-finetuned-conll03-english", "task": "token-classification", "dataset": { "path": "conll2003", "eval_split": "validation", "data_keys": { "primary": "tokens", "secondary": null }, "ref_keys": [ "ner_tags" ], "name": null, "calibration_split": "train" }, "quantization_approach": "dynamic", "operators_to_quantize": [ "Add", "MatMul" ], "node_exclusion": [], "aware_training": false, "per_channel": false, "calibration": { "method": "minmax", "num_calibration_samples": 100, "calibration_histogram_percentile": null, "calibration_moving_average": null, "calibration_moving_average_constant": null }, "framework": "onnxruntime", "framework_args": { "opset": 11, "optimization_level": 1 }, "hardware": "Architecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nAddress sizes: 46 bits physical, 48 bits virtual\nCPU(s): 8\nOn-line CPU(s) list: 0-7\nThread(s) per core: 2\nCore(s) per socket: 4\nSocket(s): 1\nNUMA node(s): 1\nVendor ID: GenuineIntel\nCPU family: 6\nModel: 85\nModel name: Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz\nStepping: 4\nCPU MHz: 3108.647\nBogoMIPS: 5000.01\nHypervisor vendor: KVM\nVirtualization type: full\nL1d cache: 128 KiB\nL1i cache: 128 KiB\nL2 cache: 4 MiB\nL3 cache: 33 MiB\nNUMA node0 CPU(s): 0-7\nVulnerability Itlb multihit: KVM: Vulnerable\nVulnerability L1tf: Mitigation; PTE Inversion\nVulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nVulnerability Meltdown: Mitigation; PTI\nVulnerability Spec store bypass: Vulnerable\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke\n", "versions": { "transformers": "4.20.1", "optimum": "1.2.3.dev0", "optimum_hash": "65ad733f16d4e34fee21f98cb1b7edd33b8c1662" }, "evaluation": { "time": [ { "batch_size": 1, "input_length": 64, "baseline": { "nb_forwards": 51, "throughput": 17.0, "latency_mean": 59.34628984313726, "latency_std": 0.778172196284656, "latency_50": 59.237776, "latency_90": 60.363116, "latency_95": 60.5019055, "latency_99": 61.45229, "latency_999": 61.5998972 }, "optimized": { "nb_forwards": 137, "throughput": 45.67, "latency_mean": 21.914979291970806, "latency_std": 0.10463730867869543, "latency_50": 21.891895, "latency_90": 22.0202298, "latency_95": 22.075110600000002, "latency_99": 22.195937479999998, "latency_999": 22.661796416 } } ], "others": { "baseline": { "precision": 0.9742268041237113, "recall": 0.9545454545454546, "f1": 0.9642857142857142, "accuracy": 0.9896296296296296 }, "optimized": { "precision": 0.9740932642487047, "recall": 0.9494949494949495, "f1": 0.9616368286445013, "accuracy": 0.9888888888888889 } } }, "max_eval_samples": 100, "time_benchmark_args": { "duration": 3, "warmup_runs": 1 }, "model_type": "distilbert" }, { "model_name_or_path": "elastic/distilbert-base-uncased-finetuned-conll03-english", "task": "token-classification", "dataset": { "path": "conll2003", "eval_split": "validation", "data_keys": { "primary": "tokens", "secondary": null }, "ref_keys": [ "ner_tags" ], "name": null, "calibration_split": "train" }, "quantization_approach": "dynamic", "operators_to_quantize": [ "Add" ], "node_exclusion": [], "aware_training": false, "per_channel": false, "calibration": { "method": "minmax", "num_calibration_samples": 100, "calibration_histogram_percentile": null, "calibration_moving_average": null, "calibration_moving_average_constant": null }, "framework": "onnxruntime", "framework_args": { "opset": 11, "optimization_level": 1 }, "hardware": "Architecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nAddress sizes: 46 bits physical, 48 bits virtual\nCPU(s): 8\nOn-line CPU(s) list: 0-7\nThread(s) per core: 2\nCore(s) per socket: 4\nSocket(s): 1\nNUMA node(s): 1\nVendor ID: GenuineIntel\nCPU family: 6\nModel: 85\nModel name: Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz\nStepping: 4\nCPU MHz: 2898.018\nBogoMIPS: 5000.01\nHypervisor vendor: KVM\nVirtualization type: full\nL1d cache: 128 KiB\nL1i cache: 128 KiB\nL2 cache: 4 MiB\nL3 cache: 33 MiB\nNUMA node0 CPU(s): 0-7\nVulnerability Itlb multihit: KVM: Vulnerable\nVulnerability L1tf: Mitigation; PTE Inversion\nVulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nVulnerability Meltdown: Mitigation; PTI\nVulnerability Spec store bypass: Vulnerable\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke\n", "versions": { "transformers": "4.20.1", "optimum": "1.2.3.dev0", "optimum_hash": "65ad733f16d4e34fee21f98cb1b7edd33b8c1662" }, "evaluation": { "time": [ { "batch_size": 1, "input_length": 64, "baseline": { "nb_forwards": 51, "throughput": 17.0, "latency_mean": 59.18282125490196, "latency_std": 0.6426837410053724, "latency_50": 59.074956, "latency_90": 60.217791, "latency_95": 60.3066755, "latency_99": 60.4462285, "latency_999": 60.50652175 }, "optimized": { "nb_forwards": 103, "throughput": 34.33, "latency_mean": 29.235091689320388, "latency_std": 0.0994740876199717, "latency_50": 29.216962, "latency_90": 29.3426696, "latency_95": 29.4535227, "latency_99": 29.55608076, "latency_999": 29.56924647 } } ], "others": { "baseline": { "precision": 0.9742268041237113, "recall": 0.9545454545454546, "f1": 0.9642857142857142, "accuracy": 0.9896296296296296 }, "optimized": { "precision": 0.9742268041237113, "recall": 0.9545454545454546, "f1": 0.9642857142857142, "accuracy": 0.9896296296296296 } } }, "max_eval_samples": 100, "time_benchmark_args": { "duration": 3, "warmup_runs": 1 }, "model_type": "distilbert" }, { "model_name_or_path": "elastic/distilbert-base-uncased-finetuned-conll03-english", "task": "token-classification", "dataset": { "path": "conll2003", "eval_split": "validation", "data_keys": { "primary": "tokens", "secondary": null }, "ref_keys": [ "ner_tags" ], "name": null, "calibration_split": "train" }, "quantization_approach": "static", "operators_to_quantize": [ "Add", "MatMul" ], "node_exclusion": [], "aware_training": false, "per_channel": false, "calibration": { "method": "minmax", "num_calibration_samples": 100, "calibration_histogram_percentile": null, "calibration_moving_average": null, "calibration_moving_average_constant": null }, "framework": "onnxruntime", "framework_args": { "opset": 11, "optimization_level": 1 }, "hardware": "Architecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nAddress sizes: 46 bits physical, 48 bits virtual\nCPU(s): 8\nOn-line CPU(s) list: 0-7\nThread(s) per core: 2\nCore(s) per socket: 4\nSocket(s): 1\nNUMA node(s): 1\nVendor ID: GenuineIntel\nCPU family: 6\nModel: 85\nModel name: Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz\nStepping: 4\nCPU MHz: 3100.046\nBogoMIPS: 5000.01\nHypervisor vendor: KVM\nVirtualization type: full\nL1d cache: 128 KiB\nL1i cache: 128 KiB\nL2 cache: 4 MiB\nL3 cache: 33 MiB\nNUMA node0 CPU(s): 0-7\nVulnerability Itlb multihit: KVM: Vulnerable\nVulnerability L1tf: Mitigation; PTE Inversion\nVulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nVulnerability Meltdown: Mitigation; PTI\nVulnerability Spec store bypass: Vulnerable\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke\n", "versions": { "transformers": "4.20.1", "optimum": "1.2.3.dev0", "optimum_hash": "65ad733f16d4e34fee21f98cb1b7edd33b8c1662" }, "evaluation": { "time": [ { "batch_size": 1, "input_length": 64, "baseline": { "nb_forwards": 51, "throughput": 17.0, "latency_mean": 59.248113901960785, "latency_std": 0.614252094367223, "latency_50": 59.360647, "latency_90": 59.876494, "latency_95": 60.110628, "latency_99": 60.3347135, "latency_999": 60.33478775 }, "optimized": { "nb_forwards": 106, "throughput": 35.33, "latency_mean": 28.305577603773585, "latency_std": 0.1333294316560668, "latency_50": 28.283224, "latency_90": 28.4987405, "latency_95": 28.56634, "latency_99": 28.63837805, "latency_999": 28.83994 } } ], "others": { "baseline": { "precision": 0.9742268041237113, "recall": 0.9545454545454546, "f1": 0.9642857142857142, "accuracy": 0.9896296296296296 }, "optimized": { "precision": 0.08073394495412844, "recall": 0.2222222222222222, "f1": 0.11843876177658141, "accuracy": 0.4666666666666667 } } }, "max_eval_samples": 100, "time_benchmark_args": { "duration": 3, "warmup_runs": 1 }, "model_type": "distilbert" }, { "model_name_or_path": "elastic/distilbert-base-uncased-finetuned-conll03-english", "task": "token-classification", "dataset": { "path": "conll2003", "eval_split": "validation", "data_keys": { "primary": "tokens", "secondary": null }, "ref_keys": [ "ner_tags" ], "name": null, "calibration_split": "train" }, "quantization_approach": "static", "operators_to_quantize": [ "Add" ], "node_exclusion": [], "aware_training": false, "per_channel": false, "calibration": { "method": "minmax", "num_calibration_samples": 100, "calibration_histogram_percentile": null, "calibration_moving_average": null, "calibration_moving_average_constant": null }, "framework": "onnxruntime", "framework_args": { "opset": 11, "optimization_level": 1 }, "hardware": "Architecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nAddress sizes: 46 bits physical, 48 bits virtual\nCPU(s): 8\nOn-line CPU(s) list: 0-7\nThread(s) per core: 2\nCore(s) per socket: 4\nSocket(s): 1\nNUMA node(s): 1\nVendor ID: GenuineIntel\nCPU family: 6\nModel: 85\nModel name: Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz\nStepping: 4\nCPU MHz: 3100.104\nBogoMIPS: 5000.01\nHypervisor vendor: KVM\nVirtualization type: full\nL1d cache: 128 KiB\nL1i cache: 128 KiB\nL2 cache: 4 MiB\nL3 cache: 33 MiB\nNUMA node0 CPU(s): 0-7\nVulnerability Itlb multihit: KVM: Vulnerable\nVulnerability L1tf: Mitigation; PTE Inversion\nVulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nVulnerability Meltdown: Mitigation; PTI\nVulnerability Spec store bypass: Vulnerable\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke\n", "versions": { "transformers": "4.20.1", "optimum": "1.2.3.dev0", "optimum_hash": "65ad733f16d4e34fee21f98cb1b7edd33b8c1662" }, "evaluation": { "time": [ { "batch_size": 1, "input_length": 64, "baseline": { "nb_forwards": 52, "throughput": 17.33, "latency_mean": 58.76881473076923, "latency_std": 0.964794406899851, "latency_50": 58.816893, "latency_90": 60.1946519, "latency_95": 60.455782799999994, "latency_99": 60.82715915, "latency_999": 60.950738015 }, "optimized": { "nb_forwards": 95, "throughput": 31.67, "latency_mean": 31.801677631578947, "latency_std": 1.0425164191094953, "latency_50": 31.513748, "latency_90": 33.2824478, "latency_95": 34.0926499, "latency_99": 34.60494526, "latency_999": 34.640664226000005 } } ], "others": { "baseline": { "precision": 0.9742268041237113, "recall": 0.9545454545454546, "f1": 0.9642857142857142, "accuracy": 0.9896296296296296 }, "optimized": { "precision": 0.07346938775510205, "recall": 0.18181818181818182, "f1": 0.10465116279069768, "accuracy": 0.2903703703703704 } } }, "max_eval_samples": 100, "time_benchmark_args": { "duration": 3, "warmup_runs": 1 }, "model_type": "distilbert" } ]