googlenet

LiteRT TFLite conversion of the TorchVision googlenet model.

Files

• googlenet.tflite
• googlenet_dynamic_wi8_afp32.tflite
• googlenet_int8_channelwise.tflite — Static INT8 with channelwise INT8 weights.

Source and preprocessing

TorchVision GoogLeNet_Weights.IMAGENET1K_V1 (checkpoint). Convert to RGB, resize the short side to 256 with bilinear interpolation and antialiasing, then center crop to 224 × 224. Divide pixels by 255, subtract [0.485, 0.456, 0.406], and divide by [0.229, 0.224, 0.225]. The input layout is NCHW, shape [1, 3, 224, 224]; output is ImageNet class scores with shape [1, 1000].

Quantization

googlenet_int8_channelwise.tflite: Static INT8 with channelwise INT8 weights and INT8 input/output. The base recipe is ai_edge_quantizer.recipe.static_wi8_ai8(), with local scale/bias safeguards to avoid overflowing INT32 biases. Intermediate calibration ranges use an exponential moving average; the output logits use the global observed range.

For INT8 input, apply the preprocessing above, then compute clip(round(x / input_scale) + input_zero_point, -128, 127) and cast to INT8, using nearest rounding with ties away from zero, as in this validation. Dequantize output scores with (q - output_zero_point) * output_scale. Read scales and zero points from the selected model file; do not assume that they match an earlier revision. The existing FP32 usage example, where present, requires this additional I/O handling before it can be used with the INT8 variant.

Compatibility

File	CPU	GPU	NPU
googlenet.tflite	YES	YES	N/A
googlenet_int8_channelwise.tflite	YES	NO	QC/MTK/...