alibaba/TinyNeuralNetwork issues and pull requests

#280 - [misc] update QR code

Pull Request - State: closed - Opened by peterjc123 5 months ago

#279 - how to do Post-training integer quantization with int16 activation

Issue - State: open - Opened by FengWu-PKU 5 months ago - 4 comments
Labels: question

#278 - It became larger after converting to tflite model

Issue - State: open - Opened by weinixuehao 6 months ago - 4 comments
Labels: enhancement, question, work-medium

#277 - [tracer] fix special float args

Pull Request - State: closed - Opened by Juelianqvq 6 months ago

#276 - jit.trace succeed but tinynn tracer failed

Issue - State: closed - Opened by Juelianqvq 6 months ago - 1 comment

#275 - Does tinynn support following int16 quantization?

Issue - State: closed - Opened by Juelianqvq 6 months ago - 1 comment
Labels: question

#274 - PyTorch 转 TFLite 使用 int8 量化

Issue - State: closed - Opened by Tongsen99 7 months ago - 4 comments
Labels: question

#273 - KeyError when executing quantization

Issue - State: closed - Opened by iksooman 7 months ago - 5 comments
Labels: bug

#273 - KeyError when executing quantization

Issue - State: closed - Opened by iksooman 7 months ago - 5 comments
Labels: bug

#272 - OneShotChannelPruner results in the miss of some operators

Issue - State: open - Opened by Xz-Alan 7 months ago - 4 comments
Labels: bug

#272 - OneShotChannelPruner results in the miss of some operators

Issue - State: open - Opened by Xz-Alan 7 months ago - 4 comments
Labels: bug

#271 - [converter] fix shared prelu

Pull Request - State: closed - Opened by peterjc123 7 months ago

#271 - [converter] fix shared prelu

Pull Request - State: closed - Opened by peterjc123 7 months ago

#270 - [converter] support elem logic for aten::{max,min}

Pull Request - State: closed - Opened by peterjc123 7 months ago

#270 - [converter] support elem logic for aten::{max,min}

Pull Request - State: closed - Opened by peterjc123 7 months ago

#269 - [converter] support aten::{maximum, minimum}

Pull Request - State: closed - Opened by peterjc123 7 months ago

#269 - [converter] support aten::{maximum, minimum}

Pull Request - State: closed - Opened by peterjc123 7 months ago

#268 - torch.max not working

Issue - State: closed - Opened by weinixuehao 7 months ago - 2 comments
Labels: bug, work/xx-small

#268 - torch.max not working

Issue - State: closed - Opened by weinixuehao 7 months ago - 2 comments
Labels: bug, work/xx-small

#267 - Do not support PReLU module?

Issue - State: closed - Opened by weinixuehao 7 months ago - 5 comments
Labels: bug, work/x-small

#267 - Do not support PReLU module?

Issue - State: closed - Opened by weinixuehao 7 months ago - 5 comments
Labels: bug, work/x-small

#266 - Request: clamp would be more efficient to go to Bounded Relu than Maximum + Minimum

Issue - State: closed - Opened by BmanClark 8 months ago - 3 comments
Labels: question, dependencies

#266 - Request: clamp would be more efficient to go to Bounded Relu than Maximum + Minimum

Issue - State: closed - Opened by BmanClark 8 months ago - 3 comments
Labels: question, dependencies

#265 - [converter] implement torch's `aten::scaled_dot_product_attention` operator

Issue - State: open - Opened by mjamroz 8 months ago - 2 comments
Labels: enhancement, work/medium

#265 - [converter] implement torch's `aten::scaled_dot_product_attention` operator

Issue - State: open - Opened by mjamroz 8 months ago - 2 comments
Labels: enhancement, work/medium

#264 - change input to INT8 after converting to tflite

Issue - State: closed - Opened by xyzxinyizhang 9 months ago - 2 comments
Labels: question

#264 - change input to INT8 after converting to tflite

Issue - State: closed - Opened by xyzxinyizhang 9 months ago - 2 comments
Labels: question

#263 - [converter] skip 0-bias handling for group deconv

Pull Request - State: closed - Opened by peterjc123 9 months ago

#263 - [converter] skip 0-bias handling for group deconv

Pull Request - State: closed - Opened by peterjc123 9 months ago

#262 - TransposeConv wrong shape?

Issue - State: open - Opened by BmanClark 9 months ago - 15 comments
Labels: question, awaiting response, dependencies, work/medium

#261 - Whether to support pytorch to keras

Issue - State: open - Opened by cuiyong127 9 months ago - 1 comment
Labels: enhancement

#261 - Whether to support pytorch to keras

Issue - State: open - Opened by cuiyong127 9 months ago - 1 comment
Labels: enhancement

#260 - [CI] several tests for modifier failed

Issue - State: open - Opened by peterjc123 9 months ago
Labels: bug

#260 - [CI] several tests for modifier failed

Issue - State: open - Opened by peterjc123 9 months ago
Labels: bug

#259 - [CI] Fix env setup for full job matrix

Pull Request - State: closed - Opened by peterjc123 9 months ago

#259 - [CI] Fix env setup for full job matrix

Pull Request - State: closed - Opened by peterjc123 9 months ago

#258 - [contrib] bug fixes for tflite2tf

Pull Request - State: closed - Opened by Vincent-2017 9 months ago - 1 comment

#257 - [contrib] add tflite2tf

Pull Request - State: closed - Opened by peterjc123 9 months ago

#256 - tinynn.converter module not found!

Issue - State: closed - Opened by tanu-04 9 months ago - 2 comments
Labels: invalid, question

#256 - tinynn.converter module not found!

Issue - State: closed - Opened by tanu-04 9 months ago - 2 comments
Labels: invalid, question

#255 - [converter] refine reshape transpose pass

Pull Request - State: closed - Opened by peterjc123 10 months ago

#255 - [converter] refine reshape transpose pass

Pull Request - State: closed - Opened by peterjc123 10 months ago

#254 - [converter] fix elementwise pass for pack ops

Pull Request - State: closed - Opened by peterjc123 10 months ago

#254 - [converter] fix elementwise pass for pack ops

Pull Request - State: closed - Opened by peterjc123 10 months ago

#253 - [converter] map gather(+reshape) ops with seperate consecutive indices to split(unpack) ops

Issue - State: open - Opened by peterjc123 10 months ago
Labels: enhancement, work/medium

#253 - [converter] map gather(+reshape) ops with seperate consecutive indices to split(unpack) ops

Issue - State: open - Opened by peterjc123 10 months ago
Labels: enhancement, work/medium

#252 - Float model failed to convert to TFLite

Issue - State: closed - Opened by Juelianqvq 10 months ago
Labels: bug, work/medium

#252 - Float model failed to convert to TFLite

Issue - State: closed - Opened by Juelianqvq 10 months ago
Labels: bug, work/medium

#251 - [converter] remove custom Atan2

Pull Request - State: closed - Opened by Juelianqvq 10 months ago

#251 - [converter] remove custom Atan2

Pull Request - State: closed - Opened by Juelianqvq 10 months ago

#250 - [converter] stop tracking ops with used constants

Pull Request - State: closed - Opened by peterjc123 10 months ago

#249 - [quantizer] stop rewriting for quantized models

Pull Request - State: closed - Opened by peterjc123 10 months ago

#249 - [quantizer] stop rewriting for quantized models

Pull Request - State: closed - Opened by peterjc123 10 months ago

#248 - [quantizer] extra check for rewrite of {add,mul}

Pull Request - State: closed - Opened by peterjc123 10 months ago

#247 - [dequantizer] restore {add, mul}_scalar

Pull Request - State: closed - Opened by peterjc123 10 months ago

#247 - [dequantizer] restore {add, mul}_scalar

Pull Request - State: closed - Opened by peterjc123 10 months ago

#246 - [utils] fix fused bias calculation

Pull Request - State: closed - Opened by peterjc123 10 months ago

#246 - [utils] fix fused bias calculation

Pull Request - State: closed - Opened by peterjc123 10 months ago

#245 - [quantizer] fused {bn, conv}, {bn, conv, relu}

Pull Request - State: closed - Opened by peterjc123 10 months ago - 1 comment

#245 - [quantizer] fused {bn, conv}, {bn, conv, relu}

Pull Request - State: closed - Opened by peterjc123 10 months ago - 1 comment

#244 - [converter] add aten::sign support

Pull Request - State: closed - Opened by Juelianqvq 11 months ago

#244 - [converter] add aten::sign support

Pull Request - State: closed - Opened by Juelianqvq 11 months ago

#243 - [quantizer] fix convbn1d pass for nested modules

Pull Request - State: closed - Opened by zhaoxin111 11 months ago

#243 - [quantizer] fix convbn1d pass for nested modules

Pull Request - State: closed - Opened by zhaoxin111 11 months ago

#242 - [Converter] Need transpose optimization

Issue - State: closed - Opened by Juelianqvq 11 months ago - 2 comments
Labels: question

#242 - [Converter] Need transpose optimization

Issue - State: closed - Opened by Juelianqvq 11 months ago - 2 comments
Labels: question

#241 - [Converter] Support GRU separated_rnn_gate_calc=False

Pull Request - State: closed - Opened by Juelianqvq 11 months ago - 1 comment

#241 - [Converter] Support GRU separated_rnn_gate_calc=False

Pull Request - State: closed - Opened by Juelianqvq 11 months ago - 1 comment

#240 - Meet Detailed error: Tensor-likes are not close! using TFLiteConverter

Issue - State: closed - Opened by Juelianqvq 11 months ago - 2 comments
Labels: bug

#240 - Meet Detailed error: Tensor-likes are not close! using TFLiteConverter

Issue - State: closed - Opened by Juelianqvq 11 months ago - 2 comments
Labels: bug

#239 - [quantizer] fix GRU for PTQ

Pull Request - State: closed - Opened by peterjc123 11 months ago

#239 - [quantizer] fix GRU for PTQ

Pull Request - State: closed - Opened by peterjc123 11 months ago

#238 - Output 0 of UnbindBackward0 is a view and is being modified inplace in GRU

Issue - State: closed - Opened by Juelianqvq 11 months ago - 1 comment

#238 - Output 0 of UnbindBackward0 is a view and is being modified inplace in GRU

Issue - State: closed - Opened by Juelianqvq 11 months ago - 1 comment

#237 - [optimizer] reset hints & more expands

Pull Request - State: closed - Opened by peterjc123 12 months ago

#237 - [optimizer] reset hints & more expands

Pull Request - State: closed - Opened by peterjc123 12 months ago

#236 - [quantizer] fix broadcast bug

Pull Request - State: closed - Opened by zk1998 12 months ago

#236 - [quantizer] fix broadcast bug

Pull Request - State: closed - Opened by zk1998 12 months ago

#235 - [converter] fix fusion with conv_transpose_with_bias=False

Pull Request - State: closed - Opened by peterjc123 12 months ago - 1 comment

#235 - [converter] fix fusion with conv_transpose_with_bias=False

Pull Request - State: closed - Opened by peterjc123 12 months ago - 1 comment

#234 - [converter] add aten::frobenius_norm support

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#234 - [converter] add aten::frobenius_norm support

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#233 - [converter] add aten::frobenius_norm support

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#233 - [converter] add aten::frobenius_norm support

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#232 - 模型是监督 + 多任务

Issue - State: closed - Opened by weinixuehao about 1 year ago - 2 comments
Labels: enhancement

#232 - 模型是监督 + 多任务

Issue - State: closed - Opened by weinixuehao about 1 year ago - 2 comments
Labels: enhancement

#231 - [converter] improve opcode compatibility

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#231 - [converter] improve opcode compatibility

Pull Request - State: closed - Opened by Juelianqvq about 1 year ago

#230 - 模型post量化后转tflite出现如下错误

Issue - State: closed - Opened by pyl62112991 about 1 year ago - 2 comments
Labels: question

#230 - 模型post量化后转tflite出现如下错误

Issue - State: closed - Opened by pyl62112991 about 1 year ago - 2 comments
Labels: question

#229 - [converter] add aten::broadcast_tensors

Pull Request - State: closed - Opened by peterjc123 about 1 year ago

#228 - [quantizer] fix tensor broadcasting for q_add ops

Pull Request - State: closed - Opened by peterjc123 about 1 year ago

#228 - [quantizer] fix tensor broadcasting for q_add ops

Pull Request - State: closed - Opened by peterjc123 about 1 year ago

#227 - [llm_quant]fix fc's weight in_features not 4-aligned bug

Pull Request - State: closed - Opened by zk1998 about 1 year ago

#227 - [llm_quant]fix fc's weight in_features not 4-aligned bug

Pull Request - State: closed - Opened by zk1998 about 1 year ago

#226 - 多输入，单输出，tinynn 从Pytorch直接QAT量化生成int8 tflite模型有些问题

Issue - State: closed - Opened by puyiwen about 1 year ago - 50 comments
Labels: enhancement, awaiting response

#226 - 多输入，单输出，tinynn 从Pytorch直接QAT量化生成int8 tflite模型有些问题

Issue - State: closed - Opened by puyiwen about 1 year ago - 50 comments
Labels: enhancement, awaiting response

#225 - [llm_quant] bug fix for torch2.0

Pull Request - State: closed - Opened by zk1998 about 1 year ago

#225 - [llm_quant] bug fix for torch2.0

Pull Request - State: closed - Opened by zk1998 about 1 year ago

#224 - Could not export Python function call 'HardSwishJitAutoFn'.

Issue - State: closed - Opened by weinixuehao about 1 year ago - 2 comments
Labels: question

GitHub / alibaba/TinyNeuralNetwork issues and pull requests